KR100225326B1 - Digital video signal record and playback device and method for recording and playing back the same - Google Patents

Abstract

The digital video signal encoded using this function is divided into functional directions in the digital video signal, arranged at the head of 1 GOP in area units, and output from the recording medium at the time of reproduction for a predetermined time. You can efficiently run all of the edits inside.

Description

Digital video signal recording and reproducing apparatus and its recording and reproducing method

1 is a block diagram of a conventional optical disc recording and reproducing apparatus. FIG. 2 is a block diagram of a conventional MPEG video signal coding unit.

3 is a block diagram of a conventional motion compensation prediction circuit.

4 is a block diagram of a conventional MPEG signal decoding unit.

5 is a diagram showing a data reordering structure of a conventional MPEG video encoding algorithm.

6 is a diagram showing an example of a structure of a GOP of a conventional MPEG video encoding algorithm.

7A and 7B show an example of a conventional video bit stream of MPEG.

8 is a diagram showing one example of a system stream of a conventional MPEG PS.

9 is a diagram showing one example of a stream of a conventional MPEG PES packet.

10 is a block diagram of a conventional digital signal recording and reproducing apparatus.

11 is a block diagram of a video signal decoding unit in a conventional digital signal recording and reproducing apparatus.

12 is a block diagram of a video signal decoding unit in a conventional digital signal recording and reproducing apparatus.

13A and 13B show a moving picture of encoding in a conventional digital signal recording and reproducing apparatus.

An explanatory diagram showing the concept of image processing.

14 is a block diagram of a video signal encoder in a conventional digital signal recording and reproducing apparatus.

Fig. 15 is a block diagram of a recording system of the digital video signal recording and reproducing apparatus of the first embodiment.

16 is a block diagram of a reproduction system of the digital video signal recording and reproducing apparatus of the first embodiment.

17 is a conceptual diagram for explaining a macroblock.

18 is a conceptual diagram for explaining screen division.

19 is a conceptual diagram for explaining a data array.

20A to 20D are conceptual views for explaining a reproduction method of special reproduction.

21A and 21B are conceptual views for explaining a method of reproducing special reproduction when data extension is performed.

FIG. 22 is a conceptual diagram for explaining screen division of Example 3. FIG.

23 is a conceptual diagram for explaining the data arrangement of the third embodiment.

24A to 24E are conceptual views for explaining a reproduction method of special reproduction in the third embodiment.

25A and 25B are conceptual views for explaining the arrangement of the error correction blocks of the third embodiment.

26A to 26D are conceptual views for explaining the playback method of special playback in the fourth embodiment.

27A to 27F are conceptual views for explaining a method of reproducing special reproduction when data interpolation in Embodiment 4 is performed.

FIG. 28 is a conceptual diagram for explaining a data array of Embodiment 5. FIG.

29A to 29E are conceptual views for explaining the reproduction method of the special playback of the fifth embodiment.

30 is a conceptual diagram for explaining a data array of Example 6. FIG.

31A to 31F are conceptual views for explaining the playback method of special playback in the sixth embodiment.

32A and 32B are conceptual views for explaining the arrangement of the error correction blocks of the sixth embodiment.

33A to 33G are conceptual views for explaining the playback method of special playback in the seventh embodiment.

34A to 34F are conceptual views for explaining a reproduction method of special reproduction in the case of performing data interpolation in Example 7. FIG.

FIG. 35 is a conceptual diagram for explaining a data array of Example 8. FIG.

36A to 36F are conceptual views for explaining the playback method of special playback in the eighth embodiment.

37 is a block diagram of a digital video signal encoding unit of the ninth embodiment;

38 is a diagram showing the concept of frequency division according to the embodiments 9 and 11;

39 is a flowchart of the digital video signal encoding process according to the ninth embodiment.

40 is an explanatory diagram of a header of a bit stream of a ninth embodiment;

41A to 41D show rearrangements of the bit streams of the ninth embodiment.

FIG. 42 is a diagram showing one example of system address information of the ninth embodiment; FIG.

43 is a diagram showing the concept of a decoding process of Example 9. FIG.

45 is a flowchart of a decoding process of Example 9;

FIG. 46 is a block diagram of a digital video signal encoder according to the tenth embodiment;

Fig. 47 is a block diagram of a digital video signal decoding unit of the tenth embodiment.

48 is a diagram showing an example of an area of the screen of the tenth embodiment;

FIG. 49 is a diagram showing an example of a bit stream when rearranged by area units of the screen of Example 10; FIG.

50 is a flowchart of a digital video signal encoding process according to the tenth embodiment.

FIG. 51 is a diagram showing an example of address information of the system stream of Embodiment 10; FIG.

FIG. 52 shows an example of the system stream of Embodiment 10; FIG.

53A to 53E show an example in which a screen is output until playback can be performed during playback as an example of the playback screen of the tenth embodiment.

54A to 54E are examples of the playback screen of the tenth embodiment, in which only the center portion of the screen is output during playback;

55A and 55B are examples of the playback screen of the tenth embodiment, showing an enlarged display of an area in the center of the screen during playback.

56 is a flowchart of the digital video signal decoding processing of the tenth embodiment.

57 is a block diagram of a digital video signal encoding unit of the eleventh embodiment.

58 is a flowchart of a digital video signal encoding process according to the eleventh embodiment.

59A to 59D show an example of the system stream of Example 11;

60 is a diagram showing one example of address information of the system stream of Embodiment 11;

61 is a block diagram of a digital video signal decoding unit of the eleventh embodiment.

62 is a flowchart of a digital video signal decoding process according to the eleventh embodiment.

63 is a block diagram of a digital video signal encoder according to a twelfth embodiment;

64 is a diagram showing on the image a concept of the resolution conversion of the twelfth embodiment;

65 is an explanatory diagram showing an example of the data configuration results of Examples 12, 13, and 14;

66 is a block diagram of a digital video signal encoding unit of the thirteenth embodiment;

67 is an explanatory diagram showing an example of data arrangement of DCT coefficients within a DCT block.

68 is a block diagram of a digital video signal encoding unit of the fourteenth embodiment;

FIG. 69 is an explanatory diagram showing an example of statistics of encoded data in Examples 12, 13, and 14;

70 is an explanatory diagram showing an example of the processing sequence of Examples 12, 13, and 14;

71A to 71D are explanatory diagrams showing an example of the arrangement of the DCT blocks of the fifteenth embodiment and the arrangement of frequency components in one block of bit streams;

72A is a block diagram of a digital video signal decoding unit of the fifteenth embodiment.

72B is an explanatory diagram showing the operation concept of the digital video signal decoding processing of the fifteenth embodiment;

73A is a block diagram of a digital video signal encoder according to the sixteenth embodiment.

73B is an explanatory diagram showing the concept of operation of the digital video signal encoding process of the sixteenth embodiment;

74 is a block diagram of a digital video signal decoding unit of the sixteenth embodiment;

75 is a block diagram of a digital video signal decoding unit of the seventeenth embodiment;

76 is a block diagram of a GOP address generating section and disk control section of the eighteenth embodiment;

77 is a block diagram of a GOP address generating section and disk control section including the playback process according to the eighteenth embodiment;

78 is a block diagram of a digital video signal decoding unit in the case of performing division by frequency and division by quantization according to the nineteenth embodiment.

79 is a block diagram of a digital video signal decoding unit in the case of performing division by bit length of the nineteenth embodiment;

80 is a block diagram of a digital video signal decoding unit in the case of performing division by the resolution of the nineteenth embodiment;

81 is a block diagram of a digital video signal coding unit of the twentieth embodiment;

FIG. 82 is a block diagram of a digital video signal decoding unit of the twentieth embodiment; FIG.

83A and 83B are explanatory views showing the concept of processing of the digital signal recording and reproducing apparatus of the twentieth embodiment;

84 is a block diagram of a digital video signal decoding unit in the case of performing division by frequency or division by quantization of the twenty-first embodiment;

85 is a block diagram of a digital video signal decoding unit in the case of performing division by bit length or division by quantization according to the twenty-first embodiment;

Fig. 86 is a block diagram of a digital video signal decoding unit in the case of performing division by resolution or division by quantization of the twenty-first embodiment.

87 is a block diagram of a digital video signal decoding unit in the case of performing division by frequency or division by quantization of the twenty-second embodiment;

88 is a block diagram of a digital video signal decoding unit in the case of performing division by bit length or division by quantization in the twenty-second embodiment;

89A and 89B are explanatory diagrams showing the concept of processing during skip search in the twenty-second embodiment;

FIG. 90 is an explanatory diagram showing the concept of processing in reverse playback of Example 22; FIG.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital video signal recording and reproducing apparatus for recording and reproducing digital video signals. In particular, a digital video signal for recording and reproducing a digital video signal encoded according to a motion compensation prediction and an orthogonal transformation onto a medium such as an optical disc. A recording and reproducing apparatus.

1 is a block circuit diagram of a conventional optical disc recording and reproducing apparatus disclosed in Japanese Patent Laid-Open Publication No. Hei 4-114369. In Fig. 1, reference numeral 201 denotes an A / D converter for converting a video signal, an audio signal, etc. into digital information, 202 denotes an information compression circuit, and 203 denotes compressed information as sector information equal to an integer multiple of the frame period. A frame sector converting circuit for converting, 204 is an error correcting code circuit for adding an error correction signal to sector information, 205 is a modulator for converting into a predetermined modulation code to reduce inter-code interference in a recording medium, Denoted at 206 is a laser drive circuit for modulating laser light according to a modulation code, and 207 is a laser output switch. In addition, 208 is an optical head for emitting laser light, 209 is an actuator for tracking the light beam emitted from the optical head 208, 210 is a trackbus motor for sending the optical head 208, Reference numeral 211 denotes a disc motor for rotating the optical disc 212, 219 denotes a motor driving circuit, 220 denotes a first motor control circuit, and 221 denotes a second motor control circuit. Reference numeral 213 denotes a reproduction amplifier for amplifying a reproduction signal from the optical head 208, 214 denotes a demodulator for obtaining data from a recorded modulation signal, 215 denotes an error correction decoding circuit, and 216 denotes The frame center inverse conversion circuit 217 is an information extension circuit for decompressing compressed information, and 218 is a D / A converter for converting the decompressed information into, for example, an analog video signal and an audio signal.

FIG. 2 is a block circuit diagram showing the internal configuration of the information compression circuit 202 shown in FIG. In FIG. 2, the digital video signal input from the A / D converter 201 is input to the memory circuit 301. In FIG.

The video signal 321 output from the memory circuit 301 is applied to the first input of the subtractor 302 and the second input of the motion compensation prediction circuit 310. The output of the subtractor 3020 is input to the quantizer 304 via a discrete cosine transform (DCT) circuit 303.

The output of quantizer 304 is fed to the input of transmit buffer 306 via variable length encoder 305. The output of the transmission buffer 306 is output to the frame center conversion circuit 203. On the other hand, the output of the quantizer 304 is also input to the inverse DCT circuit 308 via the inverse quantizer 307. The output of the inverse DCT circuit 308 is applied to the first input of the adder 309. An output 322 of the adder 309 is provided to the first input of the motion compensation prediction circuit 310.

It is applied to the first input of the motion compensation prediction circuit 310.

FIG. 3 is a block circuit diagram showing the internal structure of the motion compensation prediction circuit 310 in FIG. In FIG. 3, the output 322 of the adder 309 is provided to the input terminal 401a, and the output 321 of the memory circuit 301 is provided to the input terminal 401b, respectively. The signal 322 input from the input terminal 401a is input to the frame memory 404a or the frame memory 404b via the switch 403. The reference picture output from the frame memory 404a is given to the first input of the motion vector detection circuit 405a. The video signal 321 input from the input terminal 401b is provided to the second input of the motion vector detection circuit 405a. The output of the motion vector detection circuit 405a is input to the prediction mode selector 406. On the other hand, the reference picture output from the frame memory 404b is given to the first input of the motion vector detection circuit 405b. The video signal 321 input from the input terminal 401b is provided to the second input of the motion vector detection circuit 405b. An output of the motion vector detection circuit 405b is provided to the second input of the prediction mode selector 406. The third input of the prediction mode selector 406 is provided with a video signal 321 input from the input terminal 401b. A second signal is applied to the second input of the switch 407. A second output of the prediction mode selector 406 is provided to the third input of the switcher 407. The output 323 of the switcher 407 is output from the output terminal 402.

4 is a block circuit diagram showing the internal configuration of the information extension circuit 217 in FIG. In FIG. 4, the video signal input from the frame sector inverse conversion circuit 216 is input to the reception buffer 501.

The output from the receiving buffer 501 is input to the variable length decoder 502, dequantized by the inverse quantizer 503, and inverse DCT circuit 5040 is subjected to inverse ideal cosine transformation to add the adder 506. On the other hand, the output of the reception buffer 501 is also input to the predictive data decoding circuit 505, and the output of the predictive data decoding circuit 505 is applied to the second input of the adder 506. The output of the adder 506 is also output to the D / A converter 218 via the memory circuit 507.

Next, the operation will be described. One of the high-efficiency encoding methods for encoding a video signal is an encoding algorithm by the MPEG (Moving Picture Expert Group) method. This is a hybrid coding method that combines inter-picture prediction coding and intra-picture transform coding using motion compensation prediction. In the past, the MPEG system is adopted using the information compression circuit 202 having the configuration shown in FIG.

5 is a simplified diagram of an MPEG data bank structure (layer structure). In FIG. 5, reference numeral 621 denotes a sequence layer composed of a Group of Pictures (hereinafter, simply referred to as a "GOP") composed of a plurality of frame information, 622 denotes a GOP layer composed of several pictures (pictures); 623 is a slice divided into several blocks, 624 is a slice layer composed of several macroblocks, 625 is a macroblock layer, and 626 is a block composed of 8 pixels x 8 pixels. Layer.

The macroblock layer 625 is, for example, a blur of 8x8 pixels in the minimum unit of coding in the MPEG system, and this block is a unit for performing DCT. At this time, six blocks of four adjacent Y signal blocks, one Cb block and one Cr block corresponding to these positions are called macroblocks. This macroblock is the minimum unit of the motion compensation prediction, and the motion vector for the motion compensation prediction is formed in the macroblock unit.

Next, the process of inter-picture predictive encoding will be described. 6 shows an outline of inter picture prediction coding. Each picture is divided into three methods: an intra picture coded picture (hereinafter referred to as an I picture), a unidirectional predictive coded picture (hereinafter referred to as a P picture), and a bidirectional predictive coded picture (hereinafter referred to as a B picture). .

For example, in the case where N pictures are I pictures and I pictures are P pictures or I pictures, n and m are integers and 1≤n≤N / M (N x The n + M) th picture is an I picture, the (N × n + M × m) th picture (m ≠ 1) is the P picture, and the (N × n + M × m + 1) th (N × n +) The M x m + M-1) th image is a B picture. At this time, the (Nxn + 1) th image to the (Nxn + N) th image is collectively referred to as a GOP (Group of Pictures).

6 shows the case of N = 15 and M = 3. In Fig. 6, the I picture executes only intra-picture conversion encoding without performing inter-image prediction. The P picture performs prediction on the immediately preceding I picture or P picture. For example, the sixth picture in the figure is a P picture, but this performs prediction in three I pictures. The 9th P picture is predicted by the 6th P picture. The B picture is predicted by the I picture or P picture immediately before and after. For example, the 4th and 5th B pictures in the drawing are predicted by both the 3rd I picture and the 6th P picture. Therefore, the fourth and fifth images are encoded after the sixth image is encoded.

Next, the operation of the information compression circuit 202 will be described with reference to FIG. The memory circuit 301 rearranges the input digital video signals in coding order and outputs them. That is, as described above, in FIG. 6, for example, the first B picture is encoded after the third I picture, so that rearrangement of the images is performed here. 7 illustrates the operation of this rearrangement. The input image sequence of Fig. 7a is output in the order of Fig. 7b.

Also, the video signal 321 output from the memory circuit 301 is between the image with the predictive image 323 output from the motion compensation prediction circuit 310 by the subtractor 302 in order to reduce the redundancy in the time axis direction. After the difference is taken, DCT is performed in the space axis direction. The transformed coefficients are quantized, variable length coded, and then output via the transmission buffer 306. On the other hand, the quantized conversion coefficient is inversely quantized, and after inverse DCT is performed, the predictor image 323 is also virtualized in the adder 309 to obtain a decoded image 322. The decoded image 322 is input to the motion compensation prediction circuit 310 for encoding the next picture.

Next, the motion compensation prediction circuit 310 performs motion compensation prediction on the image signal 321 output from the memory circuit 301 by using two reference pictures stored in the frame memory 404a and the frame memory 404b. The predicted picture 323 is output.

First, when the image 322 encoded and decoded as described above is an I picture or a P picture, this picture 322 is stored in the frame memory 404a or the frame memory 404b for encoding the next picture. . The switching unit 403 is switched so as to select one of the frame memory 404a and the frame memory 404b, which is updated earlier in the market. However, if the decoded image 322 is a B picture, writing to the frame memory 404a and the frame memory 404b is not executed.

By this switching, for example, when the first and second B pictures of FIG. 7 are encoded, 0 P pictures and 3 I pictures are stored in the frame memory 404a and the frame memory 404b, respectively. After that, when the sixth P-picture is encoded and decoded, the frame memory 404a is rewritten on the decoded image of the sixth P-picture.

Therefore, when the next 4th and 5th B pictures are encoded, 6 P pictures and 3 I pictures are stored in the frame memories 404a and 404b, respectively. When the 9th P picture is encoded and decoded, the frame memory 404b is rewritten on the decoding of the 9th P picture. As a result, when the seventh and eighth B-pictures are encoded, six P-pictures and nine P-pictures are stored in the frame memories 404a and 404b, respectively.

When the video signal 321 output from the memory circuit 301 is input to the motion compensation prediction circuit 310, the two motion vector detection circuits 405a and 405b are respectively input to the frame memories 404a and 404b. A motion vector is detected based on the stored reference image, and a motion compensated predictive image is output. That is, the video signal 321 is divided into a plurality of blocks, a block is selected so that the predictive distortion is smallest among the reference pictures for each block, the relative position of the block is output as a motion vector, and the block is moved. It outputs as a compensation prediction image.

On the other hand, the prediction mode selector 406 selects the smallest prediction distortion among the two motion compensated prediction images output from the motion vector detection circuits 405a and 405b and their average images and outputs them as the prediction images. At this time, if the video signal 321 is not a B picture, the motion compensation predictive image corresponding to the reference image inputted earlier in time is always selected and thrust. Further, the prediction mode selector 406 selects the one having the best coding efficiency among intra-picture encoding that does not perform prediction and inter-picture prediction encoding by the selected predictive picture.

At this time, if the video signal 321 is an I picture, image internal encoding is always selected. When intra picture encoding is selected, a signal indicating the intra picture coding mode is output as the prediction mode, and when inter picture prediction coding is selected, a signal indicating the selected predictive picture is output as the prediction mode. The converter 407 outputs a zero when the prediction mode output from the prediction mode selector 406 is an intra picture coding mode, and outputs a prediction image output from the prediction mode selector 406.

When the video signal 321 output from the memory circuit 301 is an I picture, the motion compensation prediction circuit 310 always outputs a 0 signal as the side image 323, so that the I picture does not perform inter-picture prediction. In-image conversion encoding is performed. When the video signal 321 output from the memory circuit 301 is, for example, the sixth P-picture of FIG. 6, the motion compensation prediction circuit 310 predicts the motion-compensation of the third I-picture of FIG. The image 323 is output. Also, when the video signal 321 output from the memory circuit 301 is, for example, the fourth B picture of FIG. 6, the motion compensation prediction circuit 310 uses the third I picture and the sixth P picture of FIG. The predictive image 323 is output by predicting motion compensation.

Next, the operation of the transmission buffer 306 will be described. In the transmission buffer 306, the variable length coder 305 converts the variable length coded video data into a bit stream of an MPEG video signal. Here, the strip of MPEG has a six-layered layer structure as shown in FIG. 5, and includes a sequence layer 621, a GOP layer 622, a picture layer 623, a slice layer 624, and a macroblock layer. Each of the headers 625 and 626 is provided with header information which is an identification code.

In addition, the transmission buffer 306 decomposes the bit stream of the video signal and the bit stream of the audio signal into several packets, respectively, and multiplexes these packets including the synchronization signal to form a system stream of MPEG2-PS (program stream). Doing. Here, MPEG2-PS is composed of a pack layer and a packet layer, as shown in FIG. 8. The header information is added to the packet layer and the pack layer, respectively. In this conventional example, the system stream is constructed such that the data shown in FIG. 8 includes data for 1 GOP of video data.

Here, the pack layer is an upper layer of the packet layer, and has a structure in which the packet layer is integrated, and each packet layer constituting the pack layer is called a PES packet. In addition, the header information of the pack layer shown in FIG. 8 includes a pack identification signal, a synchronization signal that is the basis of video and audio signals, and the like.

On the other hand, as shown in FIG. 9, three types of PES packets exist in the packet which comprises a packet layer. Here, the second-stage packet of FIG. 9 is a video / audio / private 1 packet, and the time stamp information (PTS and DTS) necessary for decoding each packet is code and header information for identifying the head of the packet before the packet data. Added. However, the time stamp information PTS is time management information of the reproduction output, which is information for managing the decoding order of the data stream of each packet during reproduction. The DTS is time management information at the start of decoding, which is information for managing the order of sending the decoded data.

The third packet in FIG. 9 is a private 2 packet, which is a packet for writing user data. The lowest packet in FIG. 9 is a padding packet, which is a packet that masks all packet data by one. The header information of the private 2 packet and the padding packet is composed of the packet start code and the packet length.

As described above, video and audio data are converted into a system stream of MPEG2-PS by the transmission buffer 306, and modulation is performed to reduce the influence of inter-code interference for each frame sector, and is recorded on the optical disc 212. At this time, for example, the data amount in each GOP unit is made to be almost the same amount, and it is clear that the editing in the GOP unit can be performed by dividing the data into sectors equal to integer multiples of the frame period.

Next, the operation during reproduction will be described. At the time of reproduction, the video information recorded on the optical disc 212 is amplified by the reproduction amplifier 213 and restored to digital data by the demodulator 214 and the error correction decoding circuit 215. As a result, the data is restored as pure video source data from which data such as address and parity are removed, and input to the information extension circuit 217 having the configuration shown in FIG. The system stream composed of MPEG2-PS is input to the reception buffer 501.

The receiving buffer 501 decomposes the input system stream into pack units. Subsequently, each PES packet is decomposed according to the header information to reconstruct a bit stream of video and audio data divided into PES packets. Regarding the video data, the stream is decomposed to the block layer shown in Fig. 5, and the block data and the motion getter data are separated and output.

The block data output from the reception buffer 501 is input to the variable length decoder 502, and the variable length data is inversely quantized into fixed length data, inverse DCT is performed, and output to the adder 506. On the other hand, the predictive data decoding circuit 505 decodes the predictive image according to the motion vector output from the reception buffer 501 and outputs it to the adder 506.

In this case, the predictive data decoding circuit 505 is provided with a frame memory for storing the I picture and the P picture data decoded by the adder 506 in the same way as the motion compensation prediction circuit 310. Next, the predictive image is reproduced from the I picture and the P picture as the target according to the input motion vector and output to the adder 506. Note that the method of updating the reference picture data at the time of the I picture and the P picture is the same as in the case of encoding, and therefore description thereof is omitted.

The adder 506 adds the output of the predictive data decoding circuit 505 and the output of the inverse DCT circuit 504, and goes out to the memory circuit 507. Here, at the time of encoding, rearrangement of frames is performed in the order of encoding video signals that are temporally continuous as shown in FIG. For this reason, the memory circuit 507 rearranges the input data in the order shown in FIG. 7B so that the image data continues in time in the order of FIG. 7A and outputs it to the D / A converter 218.

Next, image retrieval and high-speed reproduction are started when data having such a coding structure is recorded on the optical disc. In the case of having an encoding structure as shown in FIG. 6, playback in I picture units enables high-speed playback. In this case, after the I picture is played back, the jump is executed immediately and the next or previous GOP is accessed, and the I picture is played there. By repeating such an operation, in the case of FIG. 6, high-speed send reproduction and rewind reproduction can be realized.

However, since the rate of this GOP is a variable bit rate, it cannot know at all where the head of the next GOP actually exists.

Therefore, it was not possible to determine which track should be accessed since the optical head was jumped to find the head of the GOP.

In addition, since the I picture has a very large amount of data and only the I picture is continuously played like a special playback, the read speed from the disc is limited. Therefore, it is impossible to reproduce the I picture at a frequency of 30 Hz as in a normal moving picture. Even if the reproduction of the I picture is completed and the optical head is jumped, the interval for updating to the next I picture is increased after the I picture is finished playing, thus lacking smooth movement.

The conventional digital video signal recording and reproducing apparatus is constructed as described above, and when the high speed reproduction is performed in the defensive speed using the I picture and the P picture, the head portion of the GOP is detected on a bit stream that is recorded on a recording medium such as an optical disc. After that, data of the I picture and the P picture are read. For this reason, when the data amount of the I picture and the P picture becomes very large, or when a large amount of time is spent searching for the head of the GOP, the time for reading the data from a recording medium such as an optical disc is insufficient.

For this reason, when the data amount of the I picture and the P picture becomes very large, or when a large amount of time is spent searching for the head of the GOP, the time for reading data from a recording medium such as an optical disc is insufficient.

For this reason, there is a problem that data of all I pictures and P pictures cannot be read and high-speed playback cannot be realized.

In addition, in the conventional digital video signal recording and reproducing apparatus, since the amount of data of the I picture is large, one picture function? Even when high-speed playback is performed, it takes time for data input of an I picture, and therefore, special playback over several tens of times cannot be realized. In this case, by reproducing an I picture once in a few GOPs, it is possible to realize a special high speed reproduction, but there is a problem that the contents of the image become unknown due to a long interval between update of the reproduced images.

Since the conventional digital video signal recording and reproducing apparatus is encoded as described above, only the I picture having a large amount of data at the time of skip search (as seen by fast forwarding), like a video tape recorder, does not reproduce enough data to be decoded. When the optical head is jumped or the data is played back, the data playing time is long. Therefore, if you want to obtain the same double speed, the jump destination of the GOP must be set farther away. There is the problem that there is little number.

In addition, since the sector address of the next GOP cannot be recognized due to the variable rate, there is no guarantee that the jumped track has a head of the GOP. For this reason, there is a problem that the number of commers that are output to the screen during the special playback is further reduced due to the multiple disk rotation standby in the jumped place. In addition, even if the sector address can be recognized, there is no means of knowing which data should be reproduced by the system layer to jump to the optical head. Therefore, if it does not pass through the video decoder, it is difficult to judge the efficiency of jumping the optical head. There was also the problem of deterioration.

Other conventional digital video signal recording and reproducing apparatuses are disclosed in, for example, Japanese Patent Application Laid-Open No. Hei 6-98314, Japanese Patent Application Laid-Open No. Hei 6-78289, and one example thereof is shown in FIG. .

In FIG. 10, reference numeral 775 denotes a video signal generator such as a camera and a VTR, reference numeral 776 denotes an audio signal generator such as a microphone and a VTR, reference numeral 762 denotes an image signal encoder, and 763 denotes an audio signal encoder. ) Denotes a system layer bit stream generator, 778 denotes an error corrector, 779 denotes a digital modulator, 780 denotes an optical disc, 756 denotes a play amplifier, 786 denotes when detected, and 781 denotes a digital demodulator. Reference numeral 758 denotes an error corrector, 759 denotes a system stream processor, 782 denotes a video signal decoder, 773 denotes an audio signal decoder, 784 denotes a monitor, and 785 denotes a speaker.

At present, the most common optical disc is 120 mm in diameter, and this optical disc is capable of recording data of 600 Mbytes of steel. Most recently, a product capable of recording a 74 minute video signal and an audio signal with a data rate of about 1.2 Mbps with respect to the optical disc is available.At the time of recording, the video signal generator 775 sends the video signal to the video signal encoder 7620. The audio signal generator 776 inputs the audio signal to the audio signal encoder 763 to perform encoding of the audio signal, and multiplexes a header or the like to these two encoded signals. Is executed by the system layer bit stream generator 777, an error correcting code is added by the error correcting coder 778, and then digitally modulated by the digital modulator 779 to generate a bit stream for recording. A mother disk is created by recording means (not shown) for recording this bit stream, and the contents of the mother disk are copied to the optical disk 780 for commercialization. The soft disk is created.

In the user's player, the ?? signal obtained by the optical head in the video soft disk is amplified by the reproduction amplifier 756, and the reproduction signal is input to the detector 786. The reproduction signal is detected by the detector 786, digitally demodulated by the digital demodulator 781, and error corrected by the error corrector 758. Thereafter, the video signal region is extracted from the error-corrected signal, the extracted data is decoded by the video signal decoder 782, and the audio signal decoder 783 is decoded to match the result of the audio signal. 784 and the sputter 785.

Representative methods of this video signal include MPEG1 and MPEG2, which are called MPEG methods, which are international standard encoding methods.

Hereinafter, a specific example of the encoding method will be described taking MPEG2 as an example.

FIG. 11 is a block diagram of a video signal encoding unit in accordance with a conventional digital signal recording and reproducing apparatus for explaining the MPEG2 coding scheme, and FIG. 12 is a conventional digital signal recording and reproducing apparatus for explaining the decoding scheme in FIG. A block diagram of a video signal decoder in FIG. FIG. 13 is a diagram showing the concept of motion image processing of video signal encoding in a conventional de-zipper signal recording / reproducing apparatus for explaining the motion image and grouping by the MPEG2 coding method. FIG. In IBBPBBP… … I denotes an I picture, B denotes a B picture, and P denotes a P picture. For example, in FIG. 13A, the moving images from I to before the appearance of the next I are grouped by a certain number of sheets.

The number of pictures constituting this group is usually 15, but is not particularly limited.

The GOP, which is a group of pictures constituting this group, has at least one picture that can be completely decoded with only one picture. In addition, a P picture that encodes by performing motion compensation prediction by one direction prediction of the temporal sequence based on an I picture; It consists of B pictures which are encoded based on I pictures and P pictures by bidirectional prediction of their time series. Incidentally, arrows in FIG. 13A and FIG. 13B indicate a prediction relationship.

In other words, the B picture can be coded and decoded only after the I and P pictures are provided, and the first P picture in the GOP is provided with one picture before the picture and can be coded and decoded. The P pictures after the second sheet in the GOP are equipped with the P picture before that one, and the encoding and decoding are possible. Therefore, if there is no I picture, coding decoding of both P and B pictures is impossible.

In Fig. 11, reference numeral 787 denotes an image rearranger, reference numeral 788 denotes a scan converter, reference numeral 789 denotes an encoder buffer, reference numeral 790 denotes a mode determiner, reference numeral 702 denotes a motion vector detector, reference numeral 706 denotes a subtractor, 708 has a field memory, a frame memory, and a DCT calculating section as a DCT circuit. 710 is a quantizer, 714 is an inverse quantizer, 716 is an inverse DCT circuit, 718 is an adder, 720 is an image memory, 722 is a rate controller, and 726 is a variable length encoder to be.

In FIG. 12, 733 is a variable length decoder, 736 is an inverse DCT circuit, 737 is an image memory, 738 is an adder, and 739 is an inverse scan converter. In addition, the motion vector detector 702 and the mode determiner 790 will be combined to represent a motion vector detector.

Next, operations of a digital video signal recording and reproducing apparatus using a conventional optical disc and the like will be described with reference to FIGS. 11 to 13. FIG. In FIG. 11, the image rearranger 787 rearranges the images for encoding in the order shown in FIG. 13, and converts them from the raster scan to the block scan by the scan converter 788. Subsequently, the rearrangement process of this image and the conversion process from the raster scan to the block scan are collectively referred to as preprocessing, and the image rearranger 787 and scan converter 788 are collectively referred to as a preprocessor. The input image data is block-scanned in the order of encoding, and subtracter 706 is passed if it is an I picture, and subtracted by the subtractor 706 if it is a P picture and a B picture. At this time, the input to the motion vector detector 702 (the input to the motion vector detector 702 may be an image after image rearrangement or an image after block scanning as an original image, but the circuit size becomes smaller in the latter case. The image must also be input from the image memory 720, but the reference arrow in the drawing is omitted.) The signal is read out from the image memory 720 in consideration of the direction and amount of the movement. The mode determiner 790 determines whether to use both direction prediction or one direction prediction.

The subtraction with the reference picture taking into account the motion vector is performed by the subtractor 706, resulting in an image having a small power and having a high coding efficiency. The output of the subtractor 706 is integrated in a field unit or a frame unit by the DCT circuit 708 to perform DCT and convert the data into frequency component data. This data is input to a quantizer 710 having different loads for each frequency, and zigzag-scanned two-dimensionally over a low frequency component and a high frequency component by a variable length encoder 726 to perform run length coding and Huffman coding. The variable length coded data is normally output via the encoder buffer 789. The quantized data is returned by the inverse quantizer 714 to the original image space data by the inverse DCT circuit 716 and added by the adder 718 with the data referenced by the subtractor 706. By executing, the same data as in ?? is obtained.

Fig. 12 shows a schematic block configuration of the decoding apparatus. The variable length decoder 733 decodes image data including header information such as motion-getter, encoding mode, picture mode, and the like and decodes the decoded data. After the L quantization, an inverse DCT operation is performed (in addition, the inverse quantizer at the front end of the inverse DCT circuit 736 is omitted in FIG. 12), and the image memory 737 is inversed in consideration of the motion vector. Decode the motion compensation prediction by adding the data after the DCT. The data is converted into a raster scan by the inverse scanning converter 739 to obtain and output an interlaced image.

In addition, according to the variable transmission rate disc system introduced in "Variable transfer rate disc system and its code amount control method" in a presentation by Sugiyama et al. In the 1994 Conference of Television Society, the proposal of a higher quality digital video signal coding method is proposed. consist of. This is a method in which the coding rate of the GOP is fixed to one program (for example, the first set), and the coding rate is set for each of the GOPs at a rate due to difficulty in drawing. Fig. 14 is a block diagram showing a video signal encoder in the conventional digital signal recording and reproducing apparatus by this method. In Fig. 14, reference numeral 791 denotes a motion compensation predictor, numeral 792 denotes a code amount memory, numeral 793 denotes a GOP rate setter, numeral 794 denotes a code amount allocator, numeral 795 denotes a subtractor, and numeral 796 denotes a motion compensation predictor. Code amount counter 797 is a switch. The GOP setter 793 of FIG. 14 changes the setting of the quantization value to the complexity of the drawing. That is, the output of the variable length encoder 726 is input to the code amount counter 796 while the switch 797 is connected to the provisional encoding side. In this code amount counter 796, the code amount is counted and the code amount memory Accumulate at 792.

The GOP rate setting device 793 calculates and sets the optimal code rate for each GOP by obtaining the provisional code amount for the entire program from the value of the code amount stored in the code amount memory 792. At this time, the code allocation is performed by the code amount allocator 794, which is also prepared for this coding. When the switch 797 is connected to this encoding side, the code quantity allocator 794 compares the value of the code quantity allocator with the value of the code quantity counter 796 and operates to control the quantizer 710 according to the actual code quantity. . In this way, in order to absorb the difficulty of the coding which changes slowly in the program, a small amount of code is allocated in an easy degree, and a large amount of code is allocated in a complicated degree. As a result, it has been reported that the image quality recorded at 3Mbps by this system was almost the same as that encoded at the rate of 6Mbps using the method up to that time.

Considering the possibility of skip search in the digital video signal recording and reproducing apparatus using the optical disc as described above, even if the head of the GOP can be accessed at high speed, for example, I picture and P picture are reproduced for fast delivery. In consideration of this, since the P picture is drawn at a proper position in the GOP, it is necessary to operate the optical head while searching the data on the bit stream. However, this cannot make such control in time in servo time constants such as actuators and the like. In order to find the head of a specific GOP, 15 frames of pictures are normally included per 1 GOP, and there is a possibility of searching for 0.5 seconds by NTSC scanning. However, for example, when trying to lead an I picture or a P picture during skip search, Even if the bit stream for reading 1/3 of the image at the frame rate requires 1/2 or more reads, a read speed of 2.5 times or more is required when the head movement time is 200 milliseconds. This exceeds the response limit of the actuator, and practically skip search is impossible by an appropriate regeneration method.

Since the conventional digital video signal recording and reproducing apparatus is encoded in this manner, when a skip search or the like is performed with a video tape recorder, data such as a B picture that cannot be completely restored from the original image can be reproduced. In this case, a complete playback image cannot be obtained. In particular, skip search or the like causes low-keyness (unnaturalness of movement) related to output processing in units of frames, or when variable rate coding with good reproduction quality is executed, Since the position changes, it becomes more difficult to access the head of the GOP itself, which causes problems such as uneven GOP units resulting in a space in the disk area.

SUMMARY OF THE INVENTION An object of the present invention is to provide a digital video signal recording and reproducing apparatus and a recording and reproducing method capable of performing special reproduction using an I picture having a large data amount and obtaining a high quality reproduction image.

It is another object of the present invention to provide a digital video signal recording / reproducing apparatus and a recording / reproducing method thereof capable of performing high-speed reproduction using an I picture and a P picture having a large data amount and obtaining a high quality reproduction image.

It is another object of the present invention to provide a digital video signal recording / reproducing apparatus and a recording / reproducing method capable of realizing an improvement in accessibility of a GOP on the premise of obtaining a good capture search and employing a variable bit rate encoding.

It is still another object of the present invention to enable skip search and to record video data in GOP units on the premise of employing variable rate encoding, and I frame is divided into n areas for each I picture and encoded for each area. In this case, data is recorded at the head of the 1GOP in order from the area located at the center of the screen, and the address information of each area of the I picture is simultaneously recorded as the header information. During special playback, only the I picture data in the center area of the screen is read and the data is read. For areas that have not been read, a special reproduced image is output by masking the data to a certain value. Therefore, only the data of the area located in the center of the screen of the I picture is decoded and reproduced at the time of special reproduction, so that special reproduction can be achieved at a higher speed than when all the I pictures with a large amount of data are reproduced.

In the digital video signal recording and reproducing apparatus, a specially reproduced image is outputted by extending the leaded area to the entire screen. Therefore, since the data of the center portion of the screen is expanded to synthesize the reproduced images, the area where data cannot be read is not found well, and the reproduced images are not decent.

In the digital video signal reproducing apparatus of the present invention, video data is recorded in units of GOP. For an I picture, one frame is divided into n areas, encoded for each area, and the head of one GOP is sequentially selected from the area located at the center of the screen. When reading and playing back video signals from a recording medium such as an optical disc in which the address information of each area of an I picture is recorded simultaneously as header information, only the I picture data in the center of the screen is read during special playback. In areas where data has not been read, a specially reproduced image is output by masking the data to a certain value. Therefore, it is possible to realize special playback at a higher speed than when all the I pictures are played back.

In the digital video signal recording and reproducing apparatus, a special reproduced image is output by extending the center area of the lead to the whole screen. Therefore, the area where data could not be read is not found well, and the reproduced image is not deemed unsightly.

According to another digital video signal recording / reproducing apparatus of the present invention, when recording video data in GOP units, one frame is divided into n areas for an I picture, encoded for each area, and sequentially encoded from the area located at the center of the screen. The address information of each area of an I picture is simultaneously recorded as header information at the head of one GOP, and only the I picture data is read in area units during special playback. 2, ..., n areas are read, one screen image is synthesized and output as a reproduced image, and if all areas of an I picture cannot be read within a predetermined time period, interpolation is performed by data of the previous screen. Outputs a special playback image. Therefore, the area located in the center of the screen is preferentially reproduced, and since one screen is synthesized from n I pictures, the interpolated reproduced image is hardly found.

According to another digital video signal recording / reproducing apparatus of the present invention, when recording video data in units of GOPs, one frame is divided into n areas for I pictures, encoded for each area, and recorded at each head of each GOP. In this case, the position of the area first recorded in the GOP unit is scrolled and recorded, and at the same time, the address information of each area of the I picture is simultaneously recorded as the header information, and during the special playback, only the I picture data is read in area units and output as a reproduced picture. If the entire area of the I picture cannot be read within a predetermined time period, the special reproduced image is output by interpolating the data of the previous screen.

Therefore, since the position of the area is scrolled in units of GOP during recording, one screen can be reproduced on average.

According to another digital video signal recording and reproducing apparatus of the present invention, when recording image data in GOP units, one frame is divided into n areas for I pictures, encoded for each area, and divided into error correction block units to display the picture. At the beginning of the 1GOP in order from the area to be centered, the address information of each area of the I picture is simultaneously recorded as header order information.In the case of special playback, the I picture and P picture data are read in area units and output as a reproduced picture. If all areas of the I picture or the P picture cannot be read within a predetermined time period, the special playback image is output by interpolating the data of the previous screen. Therefore, the area located in the center of the screen is preferentially reproduced, so that the interpolated reproduced image is unsightly.

According to another digital video signal recording and reproducing apparatus of the present invention, when recording image data in GOP units, an I picture and a P picture are divided into n areas and coded for each area, and the area becomes the center of the screen. First, at the head of 1 GOP, and simultaneously, address information of each area of an I picture is recorded as header information, and during special playback, data of an I picture and a P picture is read in area units, and n consecutive I pictures and When an area of areas 1, 2, ..., n is read from a P picture, and one picture is synthesized and output as a reproduced picture, and all areas of an I picture or a P picture cannot be read within a predetermined time period. Outputs a special playback image by interpolating the data of the previous screen. Therefore, the area located in the center of the screen is preferentially reproduced and synthesized from n consecutive I-pictures and P-pictures on one screen, so that the interpolated reproduction picture is unsightly.

According to another digital video signal recording and reproducing apparatus of the present invention, when recording image data in GOP units, I-frames and P-pictures are divided into n areas and encoded for each area in frame units, and the head of 1GOP is obtained. When recording in each area, the area of the first recording area is scrolled in frame units, and the address information of each area of the I picture and the P picture is simultaneously recorded as header information, and only the I picture data is used for special playback. The error correction block is read in units, and output as a reproduced image. If all I pictures cannot be read within a predetermined time period, a special reproduced image is output by interpolating with the data of the previous screen. Therefore, the area located in the center of the screen is preferentially reproduced, so that the interpolated reproduced image is unsightly.

The digital video signal reproducing method (apparatus) of the present invention divides at least an I picture for performing intra-frame encoding by a frequency domain, a quantization level, or a spatial resolution, so that at least data that is more important as an image is divided into at least I pictures. A bit stream of video data arranged at the head is formed, and the address information of the divided data is arranged at the head of the bit stream of the video data as header information to form a packet. The data is rearranged and output in the order of data before division according to the header information present, and during the special playback, the special playback is performed by decoding and outputting the data arranged at the head. Therefore, by dividing the data into the frequency domain, the quantization level or the spatial resolution, the data to be accessed during special playback is reduced, thereby obtaining a smooth special playback image. In addition, since the address of the divided data is recorded as the header information of the system stream, the number of bytes to be reproduced instantaneously can be known at the time of reproduction, so that the jump of the optical head during special reproduction can be efficiently performed. In normal playback, the data is rearranged according to the address, so that playback can be performed without causing irrationality due to division.

The digital video signal recording and reproducing method (apparatus) of the present invention divides at least an I picture for performing intra-frame encoding by a frequency domain, a quantization level, or a spatial resolution, and is more important as an image among data divided at least for an I picture. Constitutes a bit stream of video data arranged at the head, and sets the packet by arranging the address information of the divided data at the head of the bit stream of the video data as header information, so that the data recorded on the recording medium is The data is rearranged and output in the order of the data before division according to the header information present, and during the special playback, the special playback is performed by decoding and outputting the data arranged at the head. Therefore, by dividing the data into the frequency domain, the quantization level or the spatial resolution, the data to be accessed during special playback is reduced, thereby obtaining a smooth special playback image. In addition, since the address of the divided data is recorded as the header information of the system stream, the number of bytes to be reproduced instantaneously can be known at the time of reproduction, so that the jump of the optical head during special reproduction can be efficiently performed. In normal playback, the data is rearranged according to the address, so that recording and playback can be performed without causing an unreasonable result of division.

Another digital video signal recording / reproducing method (device) of the present invention divides (n1) at least the I picture for performing intra-frame encoding at the time of recording, rearranges the n-divided I picture data in area units, and displays them on the screen. The area that comes to the center constitutes a bit stream of video data arranged at the head, the address information of the n-divided area is arranged at the head of the bit stream of video data as header information, and a packet is formed and recorded on a recording medium. At the time of reproduction, special reproduction is executed by rearranging and outputting data in order of data before division according to the header information in the packet, and at the time of special reproduction, decoding and outputting the data arranged at the head. Therefore, since data to be accessed during special playback is reduced in dividing the data in the area of the screen at the time of recording, and the address of the divided data is recorded as header information of the system stream, the number of bytes to be instantaneously reproduced at the time of reproduction is reduced. As can be seen, it is possible to efficiently jump the optical head during special playback, and to perform address jumping on a regular basis.

In normal playback, the data is rearranged according to the address, so that recording and playback can be performed without causing irrationality due to division.

Another digital video signal recording and reproducing method (apparatus) according to the present invention divides an I picture that performs at least intra-frame encoding at the time of recording by a low frequency region, a high frequency region, a quantization level, or a spatial resolution, and divides the I picture. Rearrange the basic data among the data in each area unit on the screen to form a bit stream of video data in which the center area on the screen is located at the head of the I picture, and divide the divided area, data division, and address information of the picture. The header is formed at the head of the bit stream of the video data as the header information, and the packet is formed and recorded on the recording medium. In normal playback, the data is rearranged and output in area units according to the header information placed at the head of the packet. The rearranged data is rearranged in the original data order, and during special playback, the data can be read from the beginning of the packet within a certain time. Was executes a special reproduction by only the I picture data output. Therefore, data is divided into four frequencies, quantization or spatial resolution at the time of recording, and divided into area units on the screen. As a result, data to be accessed during special reproduction is reduced, and smooth special reproduction image is obtained by gradually decreasing the amount of data to be accessed during special reproduction. In addition, since the address of the divided data is recorded as the header information of the system stream, the number of bytes to be reproduced instantaneously can be known at the time of reproduction, so that the jump of the optical head during special reproduction can be efficiently performed. In addition, the data divided by the plurality of dividing means can adjust the read data amount in accordance with the special reproduction speed, and can cope with a wide range of special reproduction speeds. In normal playback, the data is rearranged according to the address, so that recording and playback can be performed without causing an unreasonable result of division.

Another digital video signal recording and reproducing method (apparatus) of the present invention divides at least an I picture for performing intra-frame encoding into a low frequency region, a high frequency region, a quantization level, or a spatial resolution, and among the divided I picture data, The basic data is rearranged in units of areas on the screen to form a bit stream of video data in which an area in the center of the I picture is located at the head, and the divided area, data division, and address information of the picture are header information. In this case, the packet is formed by arranging the packet at the head of the bit stream of the video data, and during normal playback on the recording medium on which the data is recorded, the data is rearranged and output in area units according to the header information arranged at the head of the packet. The original data is rearranged, and the special data is reproduced within a certain time from the beginning of the packet. Executes a special reproduction by outputting only I-picture data could I load. Therefore, by dividing the data into frequency, quantization, or spatial resolution and dividing it into area units on the screen, data to be accessed during special playback is reduced, and address information of the divided data is recorded as a header of the system stream. Since the number of bytes to be reproduced instantaneously is known at the time of reproduction, it is possible to efficiently jump the optical head during special reproduction. In addition, the data divided by the plurality of dividing means can adjust the read data amount in accordance with the special reproduction speed, and can cope with a wide special reproduction speed. In normal playback, the data is rearranged according to the address, so that playback can be performed without causing irrationality due to division.

Another digital video signal recording and reproducing method (device) (or digital video signal reproducing method (apparatus) of the present invention) reads only the area which comes to the center of the screen of the I picture at the time of special playback, and the data of the unread area is read. The reproduced images are synthesized by masking the data to a certain value. Therefore, it is possible to realize special playback at a higher speed than when all the I pictures with a large amount of data are reproduced.

Another digital video signal recording / reproducing method (or device) (or digital video signal reproducing method (apparatus) of the present invention) reads only the area which comes to the center of the screen of the I picture during the special playback, and displays the read area. The expanded image is synthesized by stretching with. Therefore, compared to the case of reproducing all I pictures with a large amount of data, it is possible to realize special playback at a higher speed, so that the area where data cannot be read is hardly found.

The digital video signal recording apparatus of the present invention comprises a first encoding for encoding a video signal consisting of a coded image including at least an intra-frame coded image of a digital video signal encoded using motion compensation prediction and an orthogonal transformation according to a predetermined condition. Means, the second encoding means for performing encoding on the residual difference component by the encoding using the first encoding means among the video signals, and the respective output data output from the first and second encoding means for each image group data. And a data arranging means for arranging at a predetermined position in each image group data. Compared to encoding all video signals, the first encoding means reduces the area to be accessed at the minimum by encoding a basic portion of the motion picture. The second encoding means encodes the video information that is not encoded by the first encoding means, and encodes all the video information by the data of the two encoding means. Further, the data arranging means rearranges the data obtained by the two encoding means so as to be suitable for the head access. Therefore, encoding can be performed with fewer codes to be accessed at the minimum during special reproduction, and the arrangement on the recording medium of data to be accessed at the minimum during special reproduction can be efficiently performed.

In the digital video signal recording apparatus, the first encoding means encodes the video information subtracted at predetermined intervals from the video signal composed of the encoded image including at least the intra-frame encoded image. Therefore, if the first encoding means encodes the video data spatially swept, reduces the minimum access area, and accesses only the data of the first encoding means, the encoding can be performed to sufficiently understand the scene. have.

In the digital video signal recording apparatus, the first encoding means encodes only the low frequency region that is red-transformed. If the first encoding means encodes the partial video data in frequency, and reduces the minimum access area and accesses only the data of the first encoding means, the encoding can be encoded so that the scene can be sufficiently understood.

In the digital video signal recording apparatus, the first encoding means is roughly quantized to the quantization level and encoded. By substantially quantizing the first encoding means, the data of higher bits having a high influence on the video can be encoded, the area to be accessed to the lowest can be reduced, and the encoding can be performed without degrading the resolution. If only data is accessed, it can be encoded so that the scene can be sufficiently understood even if it is decoded.

Another digital video signal recording apparatus of the present invention is a low frequency in a data string partitioning a video signal composed of a coded image including at least an intra-frame coded image among digital video signals encoded using motion compensation prediction and orthogonal transformation. Extract as much data as the area. The low frequency band of the video signal is partitioned by each predetermined bit for each block. Therefore, it is easy to restrict the code amount to a fixed length or less, and if the data of the low frequency region is decoded, the approximate contents of the image can be understood. Can be encoded.

The digital video signal reproducing apparatus of the present invention rearranges the data of the low frequency region and the data of the high frequency region in a predetermined order, and either the mode of decoding the rearranged data or the mode of selectively decoding the data of the low frequency region. Choose. In normal playback, two decoded encoded data are connected to each other to obtain a complete decoded image. In special reproduction, only data of a low frequency region is decoded among two decoded encoded data.

Therefore, an image can be obtained which can be decoded according to the operation status of the apparatus and can grasp an approximate content of the image.

In the digital video signal reproducing apparatus, when only the data in the low frequency region is decoded in the mode, only the decodable data is decoded, the undecryptable data near the predetermined bit boundary is discarded, and the fixed value is obtained for the obtained high frequency region data. Substitute or transform to inverse orthogonal transformation. In the case of special playback, in the case of decoding the low frequency region of the coded data divided into two, only decodable data can be decoded, and bits that could not be decoded can be discarded to decode non-normal data, and the rest of the block can be avoided. In the high frequency region of, it is replaced with a fixed value and decoded, so that a decoded image can be obtained without data formation.

According to another aspect of the present invention, a digital video signal recording apparatus includes a low-frequency region for encoded data of each block of a video signal consisting of a coded image including at least an intra-frame coded image among digital video signals encoded using motion compensation prediction and orthogonal transformation. When the predetermined number of bits as the data is reached, a block terminating code is attached, and the encoded data over the predetermined number of bits is encoded as high frequency region data. Since both the low frequency region and the high frequency region of the block are coded as if the block is apparently terminated by the block termination code (EOB), when decoding only the data of the low frequency region, it is possible to decode without unnecessary circuitry such as discarding data. Encoded data can be obtained.

Another digital video signal recording apparatus of the present invention reconstructs data according to each of the data of the low frequency region, the data of the high frequency region, and the block termination code, and decodes only the data of the low frequency region with a mode or a selective low-frequency decoding of the reconstructed data. One of the modes is selected, the reconstructed encoded data is decoded according to the selection result, and the inverse orthogonal transformation is performed by substituting a fixed value for the high frequency region. In normal playback, a complete decoded image is obtained from the encoded data divided into EOBs, and during special playback, only the data in the low frequency region of the encoded data is decoded, so that both playback modes can be operated in accordance with the operation conditions of the apparatus, so that the scene can be understood. You can get an approximate image. When the low frequency region of the encoded data is decoded, the high frequency region remaining in the block is replaced with a fixed value and decoded, so that it can be decoded without data. Both the low frequency region and the high frequency region of the block are coded as if the block was apparently terminated by EOB.

Another digital video signal recording apparatus of the present invention is a low-resolution component that performs pixel thinning on a video signal consisting of a coded image including at least an intra-frame coded image among digital video signals encoded using motion compensation prediction and orthogonal transformation. Low resolution encoding means for encoding data of the data, and low resolution encoding means for dividing the output into predetermined regions to form data, and adding an audio signal, header information, etc., and adding information correction means for attaching an error correction code. If the video data spatially decimated is encoded and only this encoded data is accessed, it can be encoded so that the scene can be sufficiently understood even if it is decoded. By interpolating the decoded data from the low resolution encoding means and comparing it with the image before performing the low resolution conversion, a difference component is obtained, and image information other than the low resolution is encoded by encoding the image data of the high resolution which is not obtained by the low resolution encoding means. Can be encoded.

Another digital video signal reproducing apparatus of the present invention synthesizes and decodes data of a low resolution component and data of a difference component. In normal playback, since encoded data of a low resolution and encoded data of a high resolution component that is a difference component of data of a low resolution of a data before being subtracted into a low resolution are synthesized, an image having a perfect resolution can be decoded.

In the digital video signal reproducing apparatus, a mode of synthesizing and decoding data of low resolution components and data of differential components and a mode of decoding only low resolution components are performed. In normal playback, two-part low resolution coded data, low resolution coded data, and high resolution component encoded data that is the difference between low resolution data can be synthesized to decode a full resolution image. The decoding mode is switched in accordance with the operation state of the apparatus so that only the data can be decoded by decoding only the data.

In the digital video signal reproducing apparatus, when decoding a low resolution image, an interpolated image is generated after decoding. In the case of special playback, when decoding only the low resolution encoded data, the video data of the low resolution component is interpolated to return to the original image size.

Another digital video signal recording apparatus of the present invention adaptively determines the data rate in accordance with the determination output for judging the degree of image degradation when encoding and protecting according to the motion compensation prediction and the orthogonal transformation. Adaptive encoding means for converting and encoding, information addition means for adding additional information such as an audio signal, a header, and an error correction code, and data rate setting means for setting discrete values to adaptively varying data rates. In a variable rate encoding means, only a very limited value is allowed. Therefore, the data rate information (corresponding to the code amount of the GOP) of the GOP can be expressed with a small number of bits.

According to another aspect of the present invention, a digital video signal recording apparatus adaptively changes a data rate in accordance with determination means for determining the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation. It has an adaptive encoding means for encoding and encoding, and additional information means for adding additional information such as an audio signal, a header, and an error correction code, and writes data rate information into a header or the like or writes to a predetermined area on a recording medium. . The data rate setting information when the variable rate is encoded is recorded separately from the image data on the recording medium. Therefore, it is possible to read data rate information collectively and to record information for immediately calculating the position occupied on the disc of the predetermined GOP.

According to still another aspect of the present invention, a digital video signal recording apparatus includes determining means for determining a degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation, and adding additional information such as an audio signal, a header, and an error correcting code. First encoding means for encoding a video signal subtracted at a predetermined interval with respect to a video signal composed of an encoded image including at least an intra-frame coded image, and encoding using a first encoding means among the video signals. And a second encoding means for encoding the residual difference component by the data, and adaptively adapts the data rate in at least one of the first and second encoding means, in accordance with the determination output from the determination means. It is set as a structure which changes and encodes with the following. High-definition encoding can be realized by the variable rate, and even in a GOP having a very high rate due to the variable rate, the image data spatially thinned out can be coded so as to reduce the minimum access area.

According to another aspect of the present invention, there is provided a digital video signal recording apparatus comprising: determining means for determining a degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation, and adding additional information such as an audio signal, a header, and an error correcting code. First encoding means for encoding only a low frequency region orthogonally transformed with respect to a video signal composed of an encoded picture including at least an intra-frame coded image, and a residual difference component by encoding using a first encoding means among the video signals And a second encoding means for encoding the data, and adaptively changing the data rate in at least one of the first and second encoding means in accordance with the determination output from the drawing determination means. It is assumed that the configuration. High-definition encoding can be realized by the variable rate, and even in a GOP having a very high rate due to the variable rate, the image data of the partial frequency domain can be coded for each block to be encoded so as to reduce the minimum access area during reproduction. .

According to still another aspect of the present invention, a digital video signal recording apparatus includes determining means for determining a degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation, and adding additional information such as an audio signal, a header, and an error correcting code. First encoding means for roughly quantizing and encoding a video signal composed of an encoded image including at least an intra-frame coded image at an quantization level, and a residual difference component by encoding using a first encoding means among the video signals And a second encoding means for encoding the data, and adaptively changing the data rate in at least one of the first and second encoding means in accordance with the determination output from the drawing determination means. It is assumed that the configuration. High-quality coding can be realized by variable rate, and even in a GOP whose rate is greatly increased by variable rate, by roughly quantizing, data of higher bits having much influence on the image can be coded so as to reduce the minimum access area. Can be encoded.

Another digital video signal reproducing apparatus of the present invention executes the switching between the normal playback mode and the special playback mode, and extracts data rate information, and in the special playback mode, data for special playback exists in accordance with the data rate information. The position on the recording medium is calculated. In the case of extracting data rate information of each GOP and reproducing a GOP having a different data rate, during normal playback, both of the encoded data divided into two are synthesized and decoded, and during special playback, the position on the recording medium of the GOP targeted for access is determined. Calculate and access the lowest accessed data to access the next target GOP. The position information on any recording medium of the GOP to be accessed is calculated from the data rate information obtained at this time, so that special reproduction and retrieval of a high-quality variable rate can be easily performed. can do.

In the digital video signal reproducing apparatus, the head position is controlled to a position on the recording medium in accordance with the position calculation result and the special reproduction speed. According to the special playback speed, the position information on any disc of the GOP to be accessed can be calculated, and the optical head position can be controlled at the position of the target GOP according to the special playback speed. It becomes possible.

In another digital video signal recording apparatus of the present invention, encoding means is performed by controlling the amount of code corresponding to an area allocated to one group of images formed by a digital video signal encoded according to motion compensation prediction and orthogonal transformation. And a data supplement means for embedding the extra data in the blank area of the image group having the blank area in accordance with the output from the encoding means. Even in the case of coding with variable rate and recording, the access time is shortened by GOP to a position where the head is easy to access. Therefore, it can be coded and recorded so as to increase the amount of read data in special reproduction. In addition, unnecessary portions such as the margin part on the disc can be embedded as much as possible, so that it can be used for improving the image quality or contributing to longer recording time.

Another digital video signal recording apparatus of the present invention has data reconstruction means for reconstructing the embedded video signal encoded data into the original image group, and data decoding means for decoding the reconstructed data from the data reconstruction means. Since encoded data of other GOPs embedded in other blank areas can be reconstructed, the encoded data of the original GOP can be reconstructed without making it into data, and the amount of lead data increases in special playback, and also high quality playback images can be obtained. You can get it.

According to another aspect of the present invention, there is provided a digital video signal recording apparatus comprising: first decoding means for decoding first and second coded data to obtain a reproduced image, and low frequency region of a coded image in a frame by decoding the first coded data. A second decoding means for obtaining a reconstructed image number corresponding to the reduced number of pixels or the approximate quantization, the low-frequency region of at least the intra-frame coded image, the inter-frame predictive-coded image, and the first number of encoded data; Among the third decoding means for obtaining a reproduced image corresponding to the approximate quantization, which decoding means is used for special reproduction is switched in accordance with the special reproduction speed. Depending on the speed of special playback, the mode for decoding and displaying only I pictures and the mode for displaying I and P pictures is switched. Therefore, special playback of I and P pictures can be realized in relatively slow special playback (for example, double speed). It is possible to realize very fine special playback with less frame jump compared to the special playback only for the I picture, and to cope with various playback speeds such as performing a special playback for the I picture during the fast special playback.

Another digital video signal reproducing apparatus of the present invention includes video data extracting means for extracting data corresponding to a video signal from a reproduction code, video data decoding / playback means for decoding and reproducing video data output from the video data extracting means, and video data. And mode switching means for switching the display mode for reproducing and reversing the field configuration of the odd or even field in the common reproduction mode and the odd field or even field. Optimize the field configuration according to the mode during special playback. In reverse playback, up to the field display is displayed to be reversed, and during frame jump such as fast forwarding, the even field outputs the same video signal as the radix field so that the number of jumping fields is kept constant. It is possible to obtain a special reproduced image easily.

The above and other objects and novel features of the present invention will become apparent from the description and the accompanying drawings.

Example 1

Embodiment 1 of the present invention will be described. FIG. 15 is a block circuit diagram showing a recording system of the digital video signal recording and reproducing apparatus of the first embodiment. In FIG. 15, the digital video signal output from the input terminal 1 is input to the formatting circuit 3. The video signal output from the formatting circuit 3 is applied to the first input to the subtractor 4 and the second input of the motion compensation prediction circuit 11. The output of the subtractor 4 is input to the quantizer 6 via the DCT circuit 5. The output of the quantizer 6 is fed to the first input of the buffer memory 12 via the variable length encoder 7. On the other hand, the output of the quantizer 6 is also input to the inverse DCT circuit 9 via the inverse quantizer 8. The output of the inverse DCT circuit 9 is applied to the first input of the adder 10.

The output of the adder 10 is given to the first input of the motion compensation prediction circuit 11. The first output of the motion compensation prediction circuit 11 is applied to the second input of the adder 10 and the second input of the subtractor 4. The second output of the motion compensation prediction circuit 11 is also provided to the second input of the buffer memory 12. The output of the buffer memory 12 is input to the modulation circuit 14 via the format encoder 13. The output of the modulation circuit 14 is recorded on a recording medium such as an optical disc via the output terminal 2.

16 is a block circuit diagram showing a reproduction system of the digital video signal recording and reproducing apparatus of the first embodiment. In FIG. 16, video information read from a recording medium such as an optical disc is inputted from the input terminal 20, and inputted into the demodulation circuit 21. FIG. The output from the demodulation circuit 21 is input to the format decoder 23 via the buffer memory 22. The first output of the format decoder 23 is input to the variable length decoder 24, de-inversed in the inverse quantizer 25, DCT is performed by the inverse DCT circuit 26, and the adder 28 Is given to the first input. On the other hand, the second output of the format decoder 23 is input to the predictive data decoding circuit 27, and the second output of the format decoder 23 is input to the predictive data decoding circuit 27, and the predictive data decoding circuit 27 is input. The output from is added to the second input of adder 28. The output of the adder 28 is output from the output terminal 30 via the unformatting circuit 29.

Next, the operation will be described. The digital video signal is input line by line from the input terminal 1 and supplied to the formatting circuit 3. In the motion compensation prediction, as shown in Fig. 6, as in Fig. 6, one GOP is set to 15 frames, one I picture, four P pictures (P1 to P4), and ten B pictures (B1 to B). B10) performs predictive encoding. In this case, the formatting circuit 3 rearranges the video data continuously input as in the conventional example in the unit of frame in the order as shown in FIG.

Further, the data input in line units is rearranged in blocks of 8x8 pixels, and macroblocks as shown in FIG. 17 (four adjacent luminance signal Y blocks and positions corresponding thereto) are shown. 6 blocks together with two color difference signals Cr and Cb blocks) to output data in units of macroblocks. Here, the macroblock is the minimum unit of the motion compensation prediction, and the motion vector for the motion compensation prediction is obtained in the macroblock unit.

In addition, the formatting circuit 3 divides the video data of one frame into three areas as shown in FIG. 18, and blocks 8x8 pixel units in this area. ?? Configure and print. Here, the three divided areas are defined as areas 1, 2, and 3 from the top of the screen as shown in FIG.

In Fig. 18, area 2 located at the center of the screen is 720 pixels x 288 lines, and areas 1 and 3 at both ends of the screen are 720 pixels x 96 lines. On the other hand, P pictures and B pictures execute blocking without dividing by area and are output in units of macroblocks.

The output of the formatting circuit 3 is input to the subtractor 4 and the motion compensation prediction circuit 11, but the subtractor 4, the DCT circuit 5, the quantizer 6, the variable length encoder 7, the inverse quantization The operations of the group 8, the inverse DCT circuit 9, the adder 10, and the motion compensation prediction circuit 11 are the same as in the conventional embodiment, and thus description thereof is omitted.

The image data output from the variable length encoder 7 and the motion vector output from the motion compensation prediction circuit 11 are input to the buffer memory 12. The buffer memory 12 records video data and motion vectors for 1 GOP, and outputs data to the sequential format encoder 13.

The output of the format encoder 13 is input to the modulation circuit 14, and an error correction code or the like is added and recorded on a recording medium such as an optical disc.

The format encoder 13 rearranges the video data for one GOP into a data array as shown in FIG. 19 and outputs the same to the modulation circuit 14. Here, I and pictures are divided into three areas as shown in FIG. 18, and data of I pictures for the areas 1 to 3 are divided into I (1), I (2), and I (3). In FIG. 19, the I picture data is recorded in the order of I (1), I (2), and I (3) at the head of the data string for 1 GOP.

In FIG. 19, an address in which data of each I picture area is stored at the head of a GOP is recorded as header information. As the header information, the number of bytes occupied by the data of each divided area on the data format of FIG. 19 is recorded. For this reason, the end position of each area can be determined as the relative address at the head of the GOP by the bypass after occupying each area recorded in the header information at the time of reproduction. This makes it possible to jump to the head address of the GOP by a unit of time at the time of special playback and to read data in units of areas according to the header information at the head of the GOP.

In the general consular mark recording arch, the I-picture is recorded in frame units at the time of recording. On the other hand, in Fig. 19, the area located at the center of the screen among the three-divided I picture data is preferentially placed at the head of 1 GOP, so even if only one area of the I picture can be decoded within a certain time during high-speed playback. It is possible to output at least the reproduced image of the center portion of the screen.

Next, operation during reproduction will be described with reference to FIG.

The video signal recorded in the format of FIG. 19 in the buffer memory 22 by the error correction process in the demodulation circuit 21 is divided into motion vectors and video data by the format decoder 23, respectively, and the predictive data decoding circuit 27 is used. And the variable length decoder 24 is output. Since the operation during normal playback is the same as in the conventional embodiment, the description is omitted.

During high-speed playback, the data recorded in 1GOP units on an optical disc or the like is jumped to the head of the GOP by a predetermined time unit, and the data portion of the I picture is read out in area units according to the header information recorded at the head. The demodulation is carried out by the inquiry path 21 and input to the buffer memory 22. Here, in the case of reading data from a recording medium such as an optical disc during high-speed playback, even if the head address of the GOP recorded on the disc is known, the rotation waiting time of the disc occurs when jumping to the head of the GOP. For this reason, when the high speed playback speed increases, the time for reading data on the disc is shortened, and the rotation waiting time of the disc is not constant, so that all I pictures cannot be stably read. Since this mode can not stably read data for one screen in high-speed playback, outputting them to the screen only when the data of areas 1 and 3 can be read out does not output these areas at regular intervals. This is because it becomes unnatural.

As shown in FIG. 19, the I picture used for the special playback is arranged so that the area located at the center of one screen at the head of the 1GOP is preferentially recorded on the recording medium. Since the high-speed playback is performed by reading the data function of the area 2 located in the center of the frame, it is easy to understand the contents of the playback picture. In addition, since only the data in the area 2 area is read from the recording medium, it is possible to realize special playback at a higher speed than when the entire I picture is read.

In addition to the first embodiment, as shown in FIG. 18, the I picture is divided into three areas in the vertical direction and recorded. However, it is not necessary to divide the picture into three parts, and n is a slice unit defined by the international standard MPEG. You may record by dividing (n1). Here, the slice has a one-dimensional structure of an arbitrary length (more than one) of macroblocks, and when the left end of the screen is reached, the slice is continuous to the left end of one row below.

Example 2

Next, Embodiment 2 of the present invention will be described with reference to the drawings. 21 is a conceptual diagram for explaining a special playback method in the case of executing the data mountain book in the second embodiment. In Example 1, the image is divided into three areas of Wawaktai shown in FIG. 18, and only data of the area 2 in the center of the screen is read and reproduced during special playback, and mask data is output for the areas 1 and 3. FIG. As shown in Fig. 21, the data of the area 2 may be expanded to one screen size and output.

In this case, the unformatting circuit 29 interpolates the data of the area 2 when converting the image data input in units of blocks into data in units of lines, and expands and outputs the size of one screen. In the case of Fig. 21, the area 2 is 720 pixels x 288 lines in size, and is composed of 144 lines of vertical symmetry at the center of one screen.

Here, in the case of special regeneration, if the upper half portion of area 2 is AR2a and the lower half portion is AR2b in FIG. 21a, the regeneration volcanoes AR2a and AR2b extended 1.5 times in the vertical direction in AR2a and 2b, respectively, as shown in FIG. Synthesize. As for the expansion method, the data in each line unit of AR2a is AT (1) (1: line number 1≤1≤144), and the line data of the upper half of the stretched screen is DT (m) (1≤m ≤ 240)

Run the kidney represented by.

On the other hand, if the data in each line unit of AR2b is AB (1) (1: line number 1≤1≤144) and the line data in the lower half of the stretched screen is DC (m) (1≤m≤240) ,

Run the kidney represented by.

As described above, since only the data of the area 2 located at the center of the screen is read at the time of high-speed playback, and expanded to a size for one screen and output as a reproduced image, both ends of the reproduced image at the time of high-speed playback are not masked. The image does not become unsightly.

In the second embodiment, the screen is stretched in the vertical direction by simply inserting data in line units. However, the line data may be linearly interpolated in the vertical direction.

Example 3

The third embodiment of the present invention will be described. The structures of the recording system and the reproduction system of the digital video signal recording and reproducing apparatus in the third embodiment are the same as those in the first embodiment (see Figs. 15 and 16).

Next, the operation will be described. The digital video signal is input line by line from the input terminal 1 and supplied to the formatting circuit 3. In the motion compensation prediction, as shown in FIG. 6, as in the conventional example, 1 GOP is set to 15 frames, one I picture, four P pictures (P1 to P4), and ten B pictures (B1 to B). Predictive encoding is performed as B10). In this case, the formatting circuit 3 rearranges and outputs the video data continuously input in the same manner as in the conventional example in units of frames in the order as shown in FIG. In addition, the data inputted in units of lines is rearranged in units of blocks of 8 x 8 pixels, and macroblocks as shown in FIG. Two blocks of chrominance signals Cr and Cb in total) and output data in units of macroblocks. Here, the macroblock is the minimum unit of the motion compensation prediction, and the motion vector for the motion compensation prediction is obtained in the macroblock unit.

In the formatting circuit 3, as shown in Fig. 22 for the I picture, the video data of one frame is divided into five areas for every 720 pixels x 96 lines in the vertical direction, and 8 x in this area. Blocks 8 pixels and constructs and outputs a macro block. In this case, the divided five areas are the areas 1, 2, 3, 4, and 5 from above, as shown in FIG. 21 and FIG. On the other hand, P pictures and B pictures execute blocking without dividing by area and are output in units of macroblocks.

The output of the formatting circuit 3 is input to the subtractor 4 and the motion compensation prediction circuit 11, but the subtractor 4, the DCT circuit 5, the quantizer 6, the variable length coder 7, the nutritionist Operations of the firearm 8, the inverse DCT circuit 9, the adder 10, and the motion compensation prediction circuit 11 are the same as in the conventional embodiment, and thus description thereof is omitted.

The image data output from the variable length encoder 7 and the motion vector output from the motion compensation prediction circuit 11 are input to the buffer memory 12. The buffer memory 12 records video data and a motion vector for 1 GOP, and outputs the data to the format encoder 13 sequentially.

The output of the format encoder 13 is input to the modulation circuit 14, and an error correction code or the like is added and recorded on a recording medium such as an optical disc.

The format encoder 13 rearranges the video signals in the data array as shown in FIG. 23 and outputs the data for 1 GOP to the modulation circuit 14. As shown in FIG. 22, the I picture is divided into five areas, and the data of the I picture for the areas 1 to 5 are divided into I (1), I (2), I (3), and I (4). In the case of I (5), in Fig. 23, I (3), I (2), I (4), It is configured to record in the order of I (1) and I (5). In FIG. 23, the address stored in the data of each I picture area at the head of the GOP is written as header information. As the header information, the number of bytes occupied by the data of each divided area on the data format shown in FIG. 23 is recorded. Therefore, the end position of each area can be determined as the relative address at the head of the GOP based on the number of bytes occupied by each area recorded in the header information at the time of reproduction. This makes it possible to jump to the head address of the GOP in a unit of predetermined time during the special playback, and to read the I picture data in each area unit in accordance with the header information from the head of the GOP.

In a general video signal recording / reproducing apparatus, an I picture is recorded in units of frames in the data format at the time of recording. On the other hand, in Fig. 23, since the position of the center of the screen among the data of the I picture divided into 5 is placed at the head of the 1GOP, the area is preferentially decoded, so that only a part of the area of the I picture can be decoded within a certain time. It is also possible to output at least the reproduction of the center portion of the screen.

Next, the operation during reproduction will be described with reference to FIG.

The video signal recorded in the format shown in FIG. 23 in the buffer memory 22 by the error correction process in the demodulation circuit 21 is divided into the motion? Getter and the video data in the format decoder 23, respectively. 27) and the variable length decoder 24. Here, since the operation during normal reproduction is the same as in the conventional embodiment, description thereof is omitted.

In high-speed playback, data recorded in a recording medium such as an optical disc in units of 1 GOP is jumped to the head of the GOP by a predetermined time unit, and the data portion of the I picture is read in area units according to the header information, and the demodulation circuit 21 Demodulates the data and inputs it to the buffer memory 22. However, since the amount of information in the I picture is large, if all the I pictures cannot be read within a certain time, the data being read is read to the end of the area, and then jumped to the beginning of the next GOP. Only the data of the area is input to the buffer memory 22. In this case, the format decoder 23 decodes only the area of the I picture that can be read and outputs it as a high speed playback image. Therefore, when 1 GOP is set to 15 frames, a special playback image of 15 times speed is obtained.

FIG. 24 shows a playback image in the case where high speed playback is performed by reproducing only one IOP of 1 GOP. In this case, if the information amount of the I picture is large and the areas of all the I pictures cannot be read from the recording medium within a predetermined time, the high-speed playback is performed by retaining and outputting the data of the area of one screen before the area which could not be read. Combine the images. In FIG. 24, when the area 5 of the n + 1th GOP and the area 1 and 5 of the G + of the n + 3th cannot be read, the reproduction image before one screen is kept as it is.

As shown in FIG. 23, the I picture used for the special playback is arranged so that the area located at the center of one screen at the head of the 1GOP is preferentially recorded on the recording medium, so that all I pictures are read within a predetermined time. Even if it cannot be done, the center part of the screen is played first, so that the contents of the playback picture are easy to understand.

In the third embodiment, when all the I pictures cannot be read, the reproduced images are interpolated in area units, but they are not necessarily executed in area units but may be executed in error correction block units.

In this case, for the data arrangement as shown in FIG. 23, the demodulation circuit 21 divides the data into packets every few bytes, and adds an error correction code to each packet. FIG. 25 shows an example in which data of five areas input continuously in FIG. 25 is divided into packets in units of error correction blocks. FIG. 25A is a data string before packet division. FIG. 25B is data after packet division. The five areas of the I picture are divided into packets of a predetermined capacity, and the area of I (3) is divided into packets of 1 to i. I (4) is divided into i to j packets and inputted.

In high-speed playback, data recorded in a recording medium such as an optical disc in units of 1 GOP is jumped to the head of the GOP by a predetermined time unit, and the data portion of the I picture is read in area units according to the header information, and the demodulation circuit 21 Demodulates the data and inputs it to the buffer memory 22. However, since the information amount of the I picture is large, if all the I pictures cannot be read within a predetermined time, the jump to the head of the next GOP is performed even during reading of data for one area. In addition, an error correction process is performed on the data that can be read, and only the data that has been able to perform error correction is input into the buffer memory 22. In this case, the format decoder 23 decodes the data decoded up to halfway and outputs it as a high speed playback image. In this case, the macroblock that could not be decoded retains the data before one screen and outputs it.

In the third embodiment, data in each area of the I picture is divided into packets in succession, but data may be divided so that one packet does not contain data of two or more areas. In this case, since one area of data is closed by an integer multiple of the error correction block, the data can be rearranged in area units immediately after the error correction process is executed. However, when dividing the data of each area into packet units, only the data is input to the last packet of each area, and the rest of the data must be masked (for example, all masked to 1).

In the third embodiment, as shown in FIG. 23, priority is given in order of areas 3, 2, 4, 1, and 5, but the order is not necessarily the order, for example, 3, 4, 2, The order of 5 and 1 may be sufficient.

In the third embodiment, as shown in Fig. 22, the I picture is divided into five areas in the vertical direction and recorded. However, it is not necessary to divide the image into five pieces, and n is a slice unit defined by the international standard MPEG. You may record by dividing (n1). Here, the slice has a one-dimensional structure of any length (more than one) of macroblocks, and when the left end of the screen is reached, the slice is a continuous band at the left end under one row.

Example 4

Next, Embodiment 4 of the present invention will be described with reference to the drawings. 26 is a diagram showing a special reproducing method of the fourth embodiment. In the third embodiment, the special reproduction is executed by the reproduction method as shown in FIG. 24, but the special reproduction may also be executed so that the reproduction image as shown in FIG. In this case, as shown in FIG. 26, the format decoder 23 synthesizes one screen by reproducing one area for each successive I picture of five GOPs. For example, in FIG. 26A, reproduction images for one screen are synthesized from the I pictures of the nth to n + 4th GOPs, n + 4 in area 1, n + 3 in area 2, and area 3 The I picture of the nth GOP in n + 2, area 4 in n + 1, and area 5 is reproduced. In addition, in FIG. 26, if time 5 is focused on the area 5, the image data to be reproduced is n, n + 1, n + 2,... The I picture of the first GOP is played.

When the information of the I picture is large and all the I pictures cannot be read within a predetermined time, the high-speed playback image is synthesized by retaining and outputting the data before one screen. FIG. 27 shows a reproduced image when the area 5 of the n + 1th GOP and the areas 1 and 5 of the n + 3th GOP cannot be read. In this case, as shown in FIG. 23, the area located at the center of one screen is preferentially recorded on the recording medium, so that even if all I pictures cannot be read in time, the center of the screen is given priority. As you play, the playback image does not become unsightly. Also, even when data reading of two or more areas is impossible, one screen is divided into five areas, and frames reproduced in each area are different, so it is difficult to know that data is missing in the reproduced image.

Example 5

Next, Embodiment 5 of the present invention will be described with reference to the drawings. 28 is a diagram showing an arrangement structure of digital image data of the fifth embodiment. In the third embodiment, the data arrangement is written in the order of areas 3, 2, 4, 1, and 5 for the I picture as shown in FIG. 23, but may be arranged as shown in FIG. . In FIG. 28, when recording I picture data at the head of the data string for one GOP, the head area number is scrolled for each GOP.

That is, as shown in FIG. 28, in the nth GOP, if I picture data is recorded in the order of I (5), I (1), I (2), I (3), and I (4), n In the + 1th GOP, I (1), I (2), I (3), I (4), and I (5) are in order. In the n + 2th GOP, I (2) is at the head and the GOP numbers are n + 3, n + 4... If it is, the head area is scrolled and recorded as I (300 I (4), I (5), I (1) ...) in order.

In addition, at the head of the GOP, an address in which data of each I picture area is stored and information for discriminating the type of the head area are written as header information. As the ladder information, the number of the area recorded at the head and the number of bytes of the data amount occupied on the data format of FIG. 27 of each area divided by five are recorded. Therefore, the end position of each area on the data order recording medium of the I picture area can be determined as the relative address from the head of the GOP based on the number of the leading area and the number of bytes occupied by each area at the time of reproduction. have.

As a result, when the special address is jumped to the head address of the GOP by a predetermined time unit, I picture data can be read in each area unit from the head of the GOP according to the header information.

In this case, the position at which the five-divided I picture area is recorded is scrolled in units of GOP, so that even if only a part of the area of the I picture can be decoded during special playback, the area that cannot be decoded is moved to a fixed position on the screen. There is no concentration.

In high-speed playback, data recorded in a recording medium such as an optical disc in units of 1 GOP is jumped to the head of the GOP by a predetermined time unit, and the data portion of the I picture is read in area units according to the header information, and the demodulation circuit 21 Demodulates the data and inputs it to the buffer memory 22. However, since the amount of information in the I picture is large, if all the I pictures cannot be read within a certain time, the data in the area of the reading is read out to the end, and then jumps to the beginning of the next GOP. Only the data of the area that can be read is written to the buffer memory 22. In this case, the format decoder 23 decodes only the area of the I picture that can be read and outputs it as a high speed playback image. Therefore, when 1 GOP is set to 15 frames, a special playback image of 15 times speed is obtained.

FIG. 29 shows a reproduction image when high-speed reproduction is performed by reproducing only one picture of 1 GOP. In this case, the I pictures are recorded on the recording medium in the order as shown in FIG. Here, when the information amount of the I picture is large and all the I pictures cannot be read in time, the high-speed playback image is synthesized by keeping the output of the data before one screen as it is. 29 shows the case where the area 5 of the n + 1th GOP and the areas 1 and 2 of the n + 3th GOP cannot be read, and the data before one screen is kept as it is.

As shown in FIG. 28, since the order of recording the I picture used for the special playback is scrolled in units of GOP, even when only a part of the area of the I picture can be decoded during the special playback, it cannot be decoded. The area does not concentrate on a fixed position on the screen.

Example 6

Next, Example 6 of the present invention will be described with reference to the drawings. 30 illustrates Example 6 according to the drawings. 30 is a diagram showing an arrangement structure of digital image data of the sixth embodiment. In this case, as shown in Fig. 22, the I picture and the P picture are divided into five areas of 720 pixels x 96 lines, and are blocked and encoded in macroblock units for each area. However, since the P picture is divided into five areas, the motion compensation prediction is performed and encoded so that the search range of the reference pattern of the motion compensation prediction is closed in the area. Here, five divided areas are defined as areas 1, 2, 3, 4, and 5 from above, as shown in FIG. In addition, for the B picture, motion compensation prediction is performed without division for each area and encoded.

The format encoder 13 rearranges the video signals as data arrays as shown in FIG. 30 and outputs the data for one GOP to the conversion circuit 14. Here, for the I picture and the P picture, as shown in FIG. 22, one screen is divided into five areas, and the I pictures for the areas 1 to 5 and the data of P1, P2, P3, and P4 pictures are I. (1) to I (5) and Pi (5) (i = 1 to 4), in Fig. 30, the I pictures, p1, p2, The data of p3 and p4 pictures are recorded in the order of 3, 2, 4, 1 and 5 areas.

In FIG. 30, the data amount of each area is recorded at the head of the 1GOP as the header information so that the address of the data of each I rule and P picture area can be identified.

In case of the high speed reproduction, the data recorded in the recording medium of the optical disc is jumped to the head of the GOP for a certain period of time, and the data part of I rule and P picture is read in area unit according to the header information. The demodulation is performed by the furnace 21 and input to the buffer memory 22. However, if all I pictures and P pictures cannot be read within a certain time because the amount of information of I and P pictures is large, after reading data to the end of the area during reading, jump to the beginning of the next GOP. Only the data of the area that can be read is input to the buffer memory 22. In this case, the format decoder 23 decodes only the areas of the I picture and the P picture that can be read and outputs them as a high speed playback image. Therefore, when 1 GOP is set to 15 frames, a special playback image of 3 times speed is obtained.

In addition, since the area located at the center of the screen among the five divided I pictures is first placed at the head of the 1GOOP, even if only a part of the I picture or the P picture can be decoded during special playback, at least the reproduced image of the middle part of the screen is reproduced. It is possible to output. In addition, since the recording medium is recorded in the order of I, P1, P2, P3, and P4 pictures, the reference data cannot be reproduced by the predictive data recovery circuit 27 even when all data cannot be read.

FIG. 31 shows a playback image in the case where high-speed playback is performed by reproducing only an I picture and a P picture in one GOP. In this case, since the information amount of the I picture and the P picture is large, if the area of all the I pictures and the P picture cannot be read from the recording medium within a predetermined time, the data of the area before one screen is retained as it is. The high speed reproduction image is synthesized by outputting the same. 31 shows that the areas 3, 4, and 5 of P4 of the nth GOP cannot be read, and data before one screen is retained.

As described above, as shown in FIG. 30, the I pictures and P pictures used for special playback are gathered and arranged in area units at the head of the 1GOP, and the I pictures or P pictures are reproduced to regenerate the special playback image during special playback. do. In addition, even when the area of all I pictures or P pictures cannot be read in time, a reproduced image can be output by interpolating data before one screen in area units.

In the sixth embodiment, when all the I pictures and P pictures cannot be read, the reproduced images are interpolated in area units, but they may not necessarily be executed in area units but may be executed in error correction block units. In this case, the demodulation circuit 21 divides the data into packets every few bytes, and adds an error correction code to each packet for the data arrangement as shown in FIG. FIG. 32 shows an example in which data of five areas to be continuously input is divided into packets of error correction block units. FIG. 32A is a data sequence before packet division, and FIG. 32B is data after packet division, and FIG. 32 is divided into i-j th packets of the eary of P1 (3).

During high-speed playback, data recorded in a recording medium such as an optical disc in units of 1 GOP is jumped to the head of the GOP by a predetermined time unit, and the data portion of the I picture is read in area units according to the header information, and the demodulation hero 21 is read. In this case, demodulation is performed and input to the memory 22. However, the amount of information in an I picture is large enough to read all I pictures and P pictures within a certain time. If not, jumps to the beginning of the next GOP even during data read for one area. In addition, an error correction process is performed on the data that can be read, and only the data that has been able to perform error correction is input to the buffer memory 22. In this case, the format decoder 23 identifies the addresses of the I picture and the P picture area that can be decoded in the middle, decodes the read data in units of macroblocks, and outputs it as a high speed playback image. In this case, the macroblock that could not be decoded retains the data before one screen and outputs it.

In the sixth embodiment, in the motion compensation prediction of P1 to P4, the search range of the motion vector is closed in each area, but it is not necessarily closed.

Example 7

Next, Example 7 of the present invention will be described with reference to the drawings. 33 is a diagram showing a special reproducing method of the seventh embodiment. Embodiment 6 may execute special reproduction so that a reproduction image as shown in FIG. 31 is output. In this case, as shown in FIG. 33, the format-to-coder 23 synthesizes one screen by reproducing one area at five consecutive frames. In FIG. 33A, reproduction pictures for one screen are synthesized from I pictures and P1 to P4 pictures. In FIG. 33A, a P4 picture in area 1, P3 in area 2, P2 in area 3, P1 in area 4, and I picture in area 5 are reproduced. Also, in FIG. 33, when the temporal attention is paid to the area 5, the video data to be reproduced is the I picture of the nth GOP, P1, P2, P3, P4, the I picture of the n + 1th GOP, P1... It is.

In addition, since the information amount of the I picture and the P picture is large, when all the I and P pictures cannot be read within a predetermined time, the previously reproduced image is synthesized by retaining and outputting the data before one screen. FIG. 34 shows a reproduced image when the areas 1, 4, and 5 of P4 of the nth GOP cannot be read. In this case, as shown in FIG. 30, since the area in the center of one screen is first recorded on the recording medium, even if it is impossible to read all the I and P pictures in time, Playback is made first so that the playback picture is not difficult to see. In addition, even when only data leads of two or more areas are available, it is divided into five areas of one screen, and frames reproduced in each area are different, so it is difficult to know that data is missing in the reproduction picture.

Example 8

Next, Embodiment 8 of the present invention will be described with reference to the drawings. 35 is a diagram showing an arrangement structure of the digital image data of the eighth embodiment. In the sixth embodiment, the data arrangement is written in the order of areas 3, 3, 4, and 1 for I picture and P picture, as shown in FIG. You may also

In FIG. 35, the head area number is scrolled for each frame when data of an I picture and a P picture is recorded at the head of a data string for one GOP. That is, as shown in FIG. 28, in the nth GOP, if I picture data is recorded in the order of I (1), I (2), I (3), I (4), and I (5), the P1 picture is recorded. In this case, P1 (2), P1 (3), P1 (4), P1 (5), and P1 (1) D are in this order. In the P2 picture, P2 (3) is the head, and in P3 and P4 pictures, the head area is scrolled and recorded like P3 (4) and P4 (5) sequentially.

At the head of the GOP, information for discriminating the address stored in the data of each I picture and P picture area and the type of the area of the head of each frame is written as header information. Here, as header information, the number of bytes of the number of bytes of the number of data of each area and the number of five-hundred percent areas recorded at the head of each frame are recorded as header information. This makes it possible to jump to the head address of the GOP by a unit of time during the special playback, and to read data in units of error words from the head of the GOP according to the header information.

In this case, since the five-split I picture and the recorded value of the P picture and the P picture area are scrolled in units of frames, even if only a part of the area of the I picture and the P picture can be decoded during special playback, decoding is possible. Unable area is not concentrated in fixed position on screen.

In high-speed playback, data recorded in 1GOP units on a recording medium such as optical discs is read in area units according to header information, and the data portion of the I picture and P picture is read in area units, and demodulated by the demodulation circuit 21 to buffer the memory. Enter in (22). However, if the amount of information of I and P pictures is large, if all I pictures and P pictures cannot be read within a certain period of time, data is read to the end of the area during reading and then jumps to the next GOP head. Only the data of the area that can be read is input to the buffer merom 22. In this case, the format decoder 23 decodes only the early picture of the I picture and the P picture which can be read and outputs as a high speed playback picture.

FIG. 36 shows a playback image when high-speed playback is performed by reproducing only I pictures and P pictures in one GOP. In this case, since the data amount of the I picture and the P picture is large, if all the I pictures and the P pictures cannot be read within a predetermined time, the high speed ashing image is synthesized by retaining and outputting the data before one screen. In FIG. 36, the data before one screen is maintained as it is when the areas 3, 4, and 5 of P4 of the nth GOP cannot be read.

As shown in FIG. 35, since the order of recording an I picture used for special playback is scrolled in units of GOPs, decoding cannot be performed even when only a part of the I picture can be decoded during random play. The area does not concentrate on a fixed position on the screen.

Example 9

Hereinafter, Example 9 of the present invention will be described with reference to the drawings.

FIG. 37 shows a digital video signal coding processing unit in a digital video recording / reproducing apparatus, which is a block diagram of the recording side in which a DCT block is divided into hierarchies of a low frequency region high frequency region and only the low frequency region is placed at the head of the GOP. In Fig. 37, reference numeral 51 denotes a buffer memory, 52 denotes a subtractor, 53 denotes a DCT circuit, 54 denotes a quantizer, 55 denotes a variable length code, and 56 denotes a dequantizer, 57 is an inverse DCT circuit, 58 is an adder, 59 is a motion compensation prediction circuit, 60 is a calculator for counting the number of events and a code amount, 61 is a format encoder, and 65 is an input terminal. .

Next, the operation will be described. The input video data is, for example, an interlaced image of an oil painting screen size of 704 pixels horizontally and 480 pixels vertically. Here, the subtractor 52, the DCT circuit 53, the alphanumeric encoder 54, the variable length code encoder 55, the inverse quantizer 56, the inverse DCT circuit 57, the adder 58, the negative tone compensation prediction Since the operation of the circuit 59 is the same as that described in the prior art example, the description is omitted.

The operation of the variable length encoder 55 will be described with reference to FIG. 38 is a diagram showing the data arrangement of the DCT coefficients inside the DCT block. In FIG. 38, data of the low frequency component and the high frequency component DCT coefficients are arranged in the upper left portion. Among the data of the DCT coefficients arranged in this DCT block, the DCT data of the low frequency region (for example, the portion hatched in FIG. 38) up to a specific position (the end of the event) is a low frequency region, variable length coded, and the format encoder. The output is 61. Variable length coding is performed on the data of the DCT coefficient after the data of the DCT coefficient at the specific position. That is, coding is performed by dividing data in the spatial frequency domain.

The boundary between the low frequency region and the harmonic region is called a breaking point.

The breaking point is set by the counter 60 so that the code amount of the low frequency region becomes a predetermined code amount accessible to the optical head at the time of special reproduction. The variable length encoder 56 divides the DCT coefficient into low and high frequencies according to the breaking point. And output to the format encoder 61.

In addition, although the encoding region is determined at the end of the event, it is obvious that other methods may be used. For example, the end of the fixed number of Ebernuts may be used, or the data may be divided into quantization data subjected to cost-quantization by the quantizer 54 and a difference value between fine quantization and rough quantization. In the buffer memory, an image obtained by dividing the spatial resolution in half, and an image half of the resolution may be divided by encoding a difference image between the original image and the original resolution image. In other words, it is of course not limited to the division of the frequency domain, and of course, the high-efficiency encoded data of the image may be divided by quantization and spatial resolution.

In this case, the more important data as an image is data of a low frequency region when frequency division is performed, and data that is quantized and hatched when the division by quantization is performed, and when the data divided by spatial resolution is encoded As one piece of data, a decoded image that is easy to be perceived by humans is obtained by decoding only these important data. In this way, one high-efficiency encoding data is divided into more fundamentally important data and other data (this is called layering), and an error correcting code is added and modulated and recorded on the disk.

Since the low-pass components of the I picture and the P picture are divided in this manner, if only the low-pass components are read and reproduced during the special playback, the amount of data to be read during the special playback is greatly reduced. As a result, the data lead time is shortened from the medium, and high speed reproduction of smooth movements can be realized in skip search. If only I pictures and P pictures are continuously arranged, only low-frequency data of I pictures and P pictures can be easily read from the disc and decoded during special playback. In this case, it is more efficient to construct the data by extracting and arranging only low-frequency components than arranging the entire area of the I picture and the P picture at the head of the GOP.

Next, the operation of the format encoder 61 will be described. 39 is a flowchart showing the operation of the format encoder. First, when the encoding is started, it is determined whether or not the encoding mode is a hierarchical mode. If the encoding mode is not the hierarchical mode, information indicating that the encoding mode is non-layered is inserted into the system stream, and the conventional stream is configured. In the hierarchical mode, the sequence header setting is recognized. Specifically, the recognition of the data of the sequence scalable extension is performed. If this is correctly described, the picture header is indexed by the picture header, and the I picture and the four P pictures are separated into low-band data and high-band data to detect respective data lengths.

On the other hand, the data of the B picture detects the data length for each picture.

In addition, a packet is written which records only the address information in the case where the low-band data of the I picture and the P picture follow the head of the GOP. This packet contains the I picture, the low pass component of the P picture, the I picture, the high pass component of the four P pictures, and the address information of the ten B pictures, and the data length of each data is recorded.

Therefore, the head position of each data stream is obtained as the relative address from the head of the GOP header by this data length. Formatting is performed by arranging the packet including the address information, the I picture and the low-pass components of the four P pictures, and the remaining data in order.

Among these, the confirmation of the scalable mode on the sequence extension of the sequence header is the confirmation of the setting of the scalable mode determined by the syntax of MPEG2 in FIG. 40 and the priority break point on the slice header. ) Is a confirmation of the technology. The privacy breakpoint is data indicating the position of the boundary between the low frequency component and the high frequency component divided by the predetermined number of events (corresponding to the breaking point described above) in the drawing.

When the scalable mode is 00, the following bit stream indicates that the data stream is a bead stream, and that the bit stream decomposed into low and high frequency components is continued. The b picture can be dedivided if it is not generated.

FIG. 41 shows an example of a bit stream generated by waxing. Figure 41a shows the bit stream without stratification. If this is layered by the circuit shown in FIG. 37, it is divided into hierarchical layers as shown in FIG. 41B. If this data is arranged in consideration of special reproduction, as shown in Fig. 41C, low-pass components of the I picture and the P picture are arranged at the head of the GOP.

Fig. 41d shows an arrangement of data in the case where these packets are packetized and address information is put into a private packet as shown in the flowchart of Fig. 39. Figs. In this case, the address information may be expanded by the relative address from the head of the GOP header as described above, but the number of bytes of the first packet may be expressed as the head of each picture, or other sector address on the disk. Of course, expressive is also good.

42 shows an example of putting address information in a private packet.

When packetizing elementary stream (hereinafter referred to as PES) packet is a private packet, the stream ID is BF (hexadecimal notation), the packet length is recorded, and all start codes (packet start code, bit stream 1 is set in the MSB of the byte and 0 is set to the next bit so as not to be the same code as the start code, etc., and the remaining 6 bits record the layering mode, the layering type, the type of image used during special playback, the number of start addresses, and the like. do.

Thereafter, 21 bits of address information are recorded so that the GOP data amount can be represented up to 2M bytes, which is the maximum length. However, as described above, 100 (binary display) is inserted into the first three bits of 21 bits so as not to become the same data as the first 24 bits of 00001 (hexadecimal display) of the start code. Here, the start address is a start address of the low pass component of the I picture, a start address of the low pass component of the four P pictures, a high pass component of the I picture, four post pictures of the P picture and ten B pictures. In addition, in order to jump the optical head by special reproduction, the sector address on the disk on which the front and rear GOP data is recorded is additionally recorded.

In addition, when one bit of parity is added to an address of 21 bits, reliability of data is improved. In this case, 10 (binary display) may be appended to the beginning of 21 bits + 1 bit. In addition to the address of the front and rear GOPs in consideration of the double speed of the special playback, and further recording the sector addresses of the front and back GOPs, the amount of change in the double speed of the special playback increases. Although the address information is recorded in the private 2 packet of the PES packet, the information may be recorded in another user area, such as a private descriptor of the program stream map.

The reproduction side of the ninth embodiment will be described with reference to FIGS. 43 and 44. FIG. 43 is a block diagram of a digital video signal decoding processing unit. In FIG. 43, 71 is a program stream header detector, 72 is a PES packet header detector, 73 is a video bit stream generator, and 74 is data. Rearranger, 75 is an address memory, 76 is a mode switch, 77 is a variable length decoder, 78 is a switch, 79 is an inverted magnetizer, 80 is an inverse DCT circuit, Reference numeral 81 denotes an adder, 82 a predictor decoder circuit, 83 a frame memory, and 84 a decodable determiner. 44 is a view showing the operation concept of FIG.

Next, the operation of FIG. 43 will be described with reference to FIG.

45 is a flowchart showing the operation of the format decoder during playback. The bit stream output from the ECC detects the header of the program stream and is separated from the PES packet. Further, the header of the PES packet is detected, and the discrimination between the private packet and the video packet including the address information is performed. In the case of a private packet, address information included in the packet is extracted and stored. On the other hand, in the case of a video packet, a bit stream of video data is extracted. In the case of live reproduction, in the case of the I picture and the P picture, data of the low frequency component and the high frequency component is extracted from the bit stream of the video data, rearrangement of the data is performed, and a reproduced image is output. On the other hand, in the special playback, only the low frequency components of the video data are extracted and reproduced. Here, after reproducing the low pass component, the optical head is jumped to the head of the next GOP.

In this case, when these addresses are described in the video stream, the address information is extracted after being streamed into the video stream, and when the addresses are described in the private descriptor of the program stream map, the address information is stored at the level of program stream header detection. Extraction and memory are lost. It is a matter of course that the address information may be a relative address or an absolute address of the program.

In reality, in Fig. 43, mode signals such as skip search and normal continuous playback are input to the mode switch 76 in the microcomputer or the like. On the other hand, the reproduction signal from the disc or the like is amplified by an amplifier, and signal reproduction is performed by a clock clocked in phase synchronization output from the PLL or the like. Next, after the discrimination operation is executed, the digital demodulation is performed, and the error correction process is executed, the program stream header detector 71 which detects each header of the program stream obtains information of the data following the header. The PES packet header detector 72 detects, for example, the address information of each picture recorded in the private 2 packet of the PES packet and the address information of the special reproduction data and accumulates in the address memory 75. Here, the PES packet for audio, the PES packet for text, and the PES packet for video are distinguished, and only the video packet is output to the video bit stream generator 73.

Here, the video bit stream generator 73 deletes the information added from the PES packet and converts it into a video stream. Specifically, data such as various control codes and timestamps are removed. Thereafter, in accordance with the address information obtained from the address memory 75, the data rearranger 74 executes rearrangement of the bit stream during normal production by the output of the mode switch 76. FIG.

The output (control signal) of the mode switch 76 is supplied to the data reorderer 74 and the decodable determiner 84. The data rearranger 74 obtains a control signal and reconstructs the data before division into the low pass component and the high pass component that are divided into hierarchical components or outputs only the low pass component to the variable length decoder 77. That is, during normal playback, the low frequency components are synthesized with the high frequency components and rearranged in the order of the original pictures. During the special playback, the low frequency components of the I picture or the low frequency components of the I picture and the P picture are output by the double speed. .

In addition, time stamps should not be used for special playback that allows only low-pass components to pass. On the other hand, the variable length decoder 77, together with the decodable determiner 84, extracts the end of the low-band component region indicated by the slice header priority point, decodes the end of the low-order component region, and then decodes the value to 78. ) Although the switch 78 is connected so as not to insert 0 during commutation reproduction, it is controlled by the decodable determiner 84 so that the high frequency unit 0 after the purity break point is inserted during special reproduction.

The above operation will be described with reference to FIG. In FIG. 44, when the partitioning breaking point is E1 to E3, up to E3 is stored in the low frequency component stream and up to E4 to EOB in the high spirit component stream. The low frequency component stream stores data of the low frequency component of the next DCT block following E3.

In normal reproduction, the data rearranger 74 extracts the data of E1 to E3 from the low frequency component stream, the data of E4 to EOB from the high frequency component stream, and extracts data of the next block, respectively. Reconstruct the DCT block. On the other hand, during special playback, the data rearranger 74 extracts data from E1 to E3, decodes the variable length by the variable length decoder 77, and then breaks the priority by the decodable determiner 84. Detection of the point occurs, and the hatched portion of FIG. 44 by the switch 78 inserts 0, and constitutes a DCT block using only the low pass component.

The data converted into the DCT block is decoded according to the motion vector.

Here, since the decoding by the motion vector is the same as in the conventional example, description thereof is omitted. However, the reference used when decoding the P picture during special playback is decoded using the I or P picture decoded only by the low pass component.

Data decoded in units of blocks is input to the frame memory 83.

Here, the frame memory 83 restores the image to the original configuration order of the GOP, executes the conversion from the Brookscan to the raster scan, and outputs it. The frame memory 83 can be shared with the memory built in the prediction data decoding circuit 82.

In addition, although the encoding region is generated at the end of the event, it is of course possible to use other methods. That is, the high efficiency encoding data of the image may be divided by quantization or spatial resolution division without limiting to the division of the frequency domain.

In this case, the more important data is the data of the low frequency region when the frequency division is performed, the data which is encoded by the rough quantization when the division by the quantization is performed, and the data which encodes the image which was chopped when the data is divided by the spatial resolution. . In this case, in a reproduced image decoded using only these data, an area that is easier to be perceived by humans is important data. In other words, one high-rate encoding data may be divided into more fundamentally important data and other data (this is referred to as layering) so that only important data is basically played back during special playback on the disc.

In the ninth embodiment, the recording side and the reproduction side are described in correspondence. However, in the case where recording and reproducing are performed in a manner such as a hard disk, only the reproduction side on the premise that they are recorded according to the assumption as in the current compact disc. Can be considered

Example 10

Next, Example 10 of the present invention will be described. 46 is a block circuit diagram showing a recording system of the digital video signal recording and reproducing apparatus of the tenth embodiment. In Fig. 46, the same reference numerals as in Fig. 37 denote the same portions, respectively, (65) is an input terminal, (51) is a buffer memory, (52) is a subtractor, (53) is a DCT circuit, and (54). ) Is a quantizer, (56) is an inverse quantizer, (57) is an inverse DCT circuit, (58) is an adder, (59) is a motion compensation prediction circuit, (55) is a variable length encoder, and (62) is an area cultivation Open 61 is the format encoder.

FIG. 47 is a block circuit diagram showing a reproduction system of the digital video signal recording and reproducing apparatus of the tenth embodiment. In Fig. 47, the same reference numerals as in Fig. 43 denote the same or corresponding parts, respectively, (71) a program stream header detector, (72) a PES packet header detector, (73) a video bit stream generator, ( 85 is an area rearranger, 75 is an address memory, 76 is a mode switch, 77 is a variable length decoder, 79 is an inverse quantizer, 80 is an inverse DCT circuit, and 81 is an In addition, (82) is prediction data decoding hi, (83) is frame memory.

Next, the operation will be described. The digital video signal is input in line units from the input terminal 65 and supplied to the buffer merom 51. Here, since the operation from the buffer memory 51 to the variable length encoder 55 is the same as in the above-mentioned conventional description, description thereof is omitted.

In the area rearranger 62, rearrangement is performed so that it is stored at the head of the bit stream of the video data of the GOP unit output from the variable length encoder 55. Here, the I picture is divided into three areas as shown in FIG. 48, and the data of the I picture for the areas 1 to 3 are I (1), I (2), and I (3), respectively. . However, each area shown in FIG. 48 is composed of several slice layers of MPEG shown in FIG. 5. In FIG. 48, areas 1 and 3 are composed of 6 slices, and area 2 is composed of 18 slices.

In practice, the area rearranger 62 detects the slice header of the I picture on the bit stream, classifies each slice into three areas shown in FIG. 48, forms a bit stream of each area, and collects each area. ? Rearrange the bit streams. That is, as shown in FIG. 49, rearrangement is carried out in the Eringer unit so that the bit stream is arranged at the head of the GOP at the heads of I (2), I (3), and I (1). The rearranged bit stream is output to the format encoder 61 in GOP units.

Next, the operation of the format encoder 61 will be described with reference to FIG. 50 shows a rhythm of an algorithm for formatting video data into a PES packet in GOP units. In the screen center priority mode, the picture header of the input bit stream is detected to detect the picture information. In the case of an I picture, the area of the screen center portion I (2), I (3) and I (1) shown in FIG. 49 is extracted, the respective data lengths are detected, and each detected area is detected. The address length is created by converting the data length of the data into a 24-bit wide binary number. On the other hand, in the case of P and B pictures, the data length is detected in picture units, and address information is created by converting it into a binary number of 24 bits in width (3 bytes).

In addition, the formatting unit collects the input address information and the bit stream of video data into two kinds of PES packets, respectively. That is, a PES packet of only address information and a PES packet of only video or audio are configured.

Therefore, as shown in Fig. 6, when there are 15 pieces of 1 GOP, there are 17 types of address information, 3 types of I pictures and 10 types of P pictures. In addition, two types of address information (absolute addresses on the disk) of the front and rear GOPs exist as address information during special playback. These address information is collected in one packet and formatted as a PES packet. In reality, it is formatted as a private 2 packet of the PES packet shown in FIG. In Fig. 51, the absolute addresses on the discs of the front and rear GOPs are arranged at the head of the packet data, and the address information of each picture is arranged in the following order. However, since the information amount of 3 bytes (24 bits) is allocated to each address information, the packet length is 57 bits.

On the other hand, a bitstream of one GOP of data other than address data is divided into a plurality of packets and added to header information, such as a synchronization signal, to be formatted into a PES packet (video packet).

The format encoder 61 also decomposes the input bit stream into PES packets to form an MPEG2-PS system stream together with the PES packets of video data. In reality, as shown in FIG. 52, a bit stream of video data for one GOP and a bit stream of audio data corresponding to one GOP period are divided into a number of packets in one pack. In this case, as shown in Fig. 52, a packet indicating the address information is placed in a packet at the head of the system stream, followed by a packet including a bit stream in the center part of the I picture. have.

Next, the operation during reproduction will be described with reference to FIG. In FIG. 47, the operations of the program stream header detector 71, the PES packet header detector 72, the video bit stream generator 73, and the mode switch 76 are the same as in the above embodiment, and description thereof is omitted.

In the bit stream of the decoded video, as shown in FIG. 49, data in the center of the screen of the I picture is arranged at the head of the stream. For this reason, in the area rearranger 85, I picture data is I (1), I for each area according to the data length of the I (2), I (3), and I (1) bit streams output from the address memory 75. (2) and I (3) are rearranged in this order. The rearranged bit stream is input to the variable length decoder 77 and decoded by a block dade, a motion vector, or the like. Since the operation following the variable length decoding during normal reproduction is the same as in the conventional embodiment, description thereof is omitted.

In the high-speed playback, as described above, one GOP of data is allocated to one pack of the system stream. Therefore, when data is read from the disc, the data of the I picture arranged at the head of the system stream is jumped to the head address of each GOP. The way to lead and jump to the beginning of the next GOP is considered. In this case, the disk drive is controlled by detecting the PES packet in which the address information placed at the head of the system stream is recorded and decoding the address on the disk of the next GOP and the address information of the I-package.

In the case of Fig. 6, if the I picture of each GOP can be read in mode within the time of one frame, high-speed reproduction at 15x speed can be realized. In addition, if the I picture of each GOP is read within two frames of time, high-speed reproduction of 7.5x is achieved. As the speed of the high-speed playback increases in this manner, the time for reading data from the disc becomes short.

In addition, when data is read from a recording medium such as an optical disc during high-speed playback, even when the head address of the system stream recorded on the disc is known, the rotation waiting time of the disc occurs when jumping to a place actually recorded on the disc. . When the video signal is encoded with a variable rate, the information amount of the I picture is not constant, and the time required for reading the I picture also changes. For this reason, when the high speed playback speed is increased, the time for reading data on the disc is shortened, and the rotation waiting time of the disc is not constant, so that the I picture data cannot be stably read.

For this reason, in the present embodiment, the data portion of the I picture is read from the disc by jumping to the head of the GOP in a unit of time for data recorded in 1 GOP units on a recording medium such as an optical disc during high speed playback. In this case, even if all I picture data cannot be read within a predetermined time, jump to the head of the next GOP. That is, jumping to the head address of each GOP in units of a predetermined time, reading the data stream as far as possible from the beginning of the recorded system stream, and jumping to the head of the next GOP. In this case, as shown in FIG. 52, a PES packet including an address and the like on the disk of the next GOP and a PES packet including data in the center of the I picture are arranged at the head of the system stream, as shown in FIG. Therefore, even when all the I picture data cannot be read during special playback, at least the address on the disc of the next GOP required for the control of the disc drive and the data in the screen center portion of the I picture can be decoded.

If only the center portion of the screen can be decoded during special playback, the error word rearranger 85 outputs only the decoded data to the variable length decoder 77, and the variable length decoded image data is inverse quantized and inverse DCT. Is executed and input to the frame memory 83. On the other hand, the area rearranger 85 outputs information of areas that could not be decoded into the frame memory 83. As a result, in the frame memory 83, only the area that was decoded during special playback can be reproduced, and the data output in the previous frame is maintained while being output as it is.

FIG. 53 shows an example of a playback picture in the case where high-speed playback is performed by reproducing only I pictures from the nth GOP to the n + 4th GOP. FIG. 53A is a case where all I pictures can be decoded. FIG. 53B is a case where areas 2 and 3 can be decoded. Area 1, which cannot be decoded, maintains the value of the previous frame and outputs it. 53C shows a case where only area 2 can be decoded, and areas 1 and 3 maintain the previous frame values as they are.

In the general video signal recording and reproducing apparatus, I pictures are recorded in units of frames in the data format at the time of recording. On the other hand, in Fig. 52, the area of the screen divided among the three divided I picture data is first placed at the head of the 1GOP, so that only a part of the area of the I picture can be read from the disc within a certain time during the playback. It is also possible to output a reproduced image of at least the middle portion of the screen.

As described above, in this example, as shown in Fig. 52, the I picture used for the special playback is arranged so that the area located at the center of one screen on the head of the 1GOP is preferentially recorded on the recording medium. Even in this case, the playback is performed with priority given to the area 2 located in the center of the screen, so that the contents of the high-speed playback image are easy to understand. In addition, since the special playback which jumps to the head of the GOP by a certain time unit is executed, the output screen can be updated at a constant double speed at all times.

In the above embodiment, all of the areas that can be decoded during special playback are output, and the data of the previous frame is retained for the areas that cannot be decoded. However, only the center portion of the screen may be reproduced during special playback.

In this case, the area rearranger 85 decodes the data of only the area 2 of the I picture read from the disc. The areas 1 and 3 that have not been decoded are mashed with gray data in the frame memory 83, for example. Outputs high speed playback images.

FIG. 54 shows a degenerate image in the case where high speed playback is performed by reproducing only the area 2 of the I picture of the I picture from the nth GOP of the 1GOP to the n + 3th GOP. 1 and 3 are masked with grey data. In addition, even when the information amount of the I picture is small and the disk rotation time is small, and there is a time allowance for reading the data of the areas 1 and 3, the data of the areas 1 and 3 are not decoded.

This is because if the area 1 and 3 data can be read only on the screen, these areas are not updated at regular intervals, and the high speed reproduction image is unnatural. For this reason, when only the screen center part (area 2) of an I picture is reproduced during special playback, the area to be updated is always constant, so that the reproduced image does not become unnatural.

In the above embodiment, only the area of the center portion of the screen of the I picture that was decoded during the special playback is displayed to mask both ends of the screen, but the center portion of the screen may be extended to one screen and output.

In this case, the frame memory 83 expands the decoded area 2 data to the size of one screen as shown in FIG. In the case of Fig. 55, however, the central portion (Fig. 55a) of the area 2 surrounded by the dotted lines is sized twice (Fig. 55b) by linear interpolation in the up, down, left and right directions, respectively. That is, in the case of Fig. 55, the portion enclosed by the dotted line is the size of horizontal 360 pixels x vertical 240 lines, and the dotted line portion is elongated by linear interpolation, for example, to the size of one screen of horizontal 720 pixels x vertical 480 lines. have.

In this way, when the area of the screen center portion is decoded and expanded to one screen size during special playback, the area of the data to be output becomes small, but the mask portion at both ends of the screen that is noticeable when only the screen center portion is output can be eliminated.

In the above embodiment, only the screen center portion of the I picture is preferentially disposed on the bit stream, but the screen center portion of the P picture as well as the I picture may be preferentially arranged. In this case, the data in the screen center portion of the P picture continues after the bit stream of the I picture.

In the above example, the rearrangement of the image data in area units is executed after the bit stream is converted. However, it is not necessary to execute the bit stream after converting the image data to the bit stream, but may be performed before the conversion.

56 shows a flowchart of the reproduction side of this embodiment. Since the flow has been described above, it will be omitted.

In the tenth embodiment, the recording side and the reproduction side are described in correspondence. However, the case where recording and reproducing is performed in combination with a hard disk can be considered. Only the case of side can be considered. The rearrangement of the area units of the screen can be realized in the side data decoding circuit 82 and the frame memory 83 by using the data of the slice vertical position of the lower 8 bits of the slice start code in the slice header. .

Example 11

Next, Embodiment 11 of the present invention will be described. Fig. 57 shows the digital video signal coding processing unit in the digital zero-signal recording / reproducing apparatus, which hierarchizes the DCT block into a low frequency region and a high frequency region. The recording side is divided into a plurality of areas, and a block diagram on the recording side in which the center of the screen in the low frequency region is given priority at the head of the GOP. In FIG. 57, 62 is an area rearranger. In addition, the same code | symbol is attached | subjected to the same or equivalent part in drawing, and description is abbreviate | omitted.

Next, the operation will be described. The input video data is, for example, data of an oil painting screen size of 704 pixels horizontally and 480 pixels vertically. This screen data uses motion compensation and DCT for high efficiency encoding. Here, the operation up to division of data is the same as that of the ninth embodiment, and thus description thereof is omitted.

In addition, the division layering may be divided not only in the frequency domain but also in quantization or spatial resolution in the same manner as in the ninth embodiment. In the eleventh embodiment, the area reordering device 62 divides the important data layered and divided into the areas of the screen as shown in the tenth embodiment, and preferentially arranges the screen center portion at the head of the GOP.

In other words, the data is divided into important data and non-important data and recorded on the disc at predetermined priorities on the screen area.

Since the low center components of the I picture and the P picture are divided in this way, the center portion of the screen is preferentially arranged, and if only the low center component of the screen portion is read and reproduced during the special reproduction, the amount of data to be read during the special reproduction is greatly reduced. As a result, there is a margin in the read speed from the medium, and it is possible to realize skip search and the like which are very high speeds from tens of times to several tens of times.

Here, by arranging the center of the screen of the region of the I picture at the head of the GOP and arranging the P-picture data following the data of the screen periphery of the low end of the I picture, the high-speed playback at the tenth to the tenth speed is performed. This can be achieved by reproducing only the screen center portion of the screen.

In addition, since the data of the screen center portion of the low frequency component of the I picture and the P picture for special reproduction is small, the information can be easily read and decoded on the disc during high-speed reproduction, thereby enabling high-speed reproduction at several times the speed. That is, since the data amount of the low center component screen portion of the I picture and the P picture is smaller than the data amount of all of these low frequency components, the special reproduction which is faster than the ninth embodiment can be realized.

Next, the operation of the area reorder 62 and the format encoder 61 will be described. 58 is the flow chart. First, when encoding is started, low-order component rearrangement is performed, so that the slice header of the I-picture of low-band component division is detected, each slice is classified into three areas shown in FIG. A bit stream is formed to rearrange the bit streams collected for each area. That is, rearrangement is performed in area units so that bit streams are arranged in the order of low band I (2), low band I (3), and low band I (1) at the head of the GOP in the same manner as in FIG. 49. do.

Next, in the screen center priority mode, the picture header of the input bit stream is detected to detect the picture information. In the case of the low pass I picture, the area of the low pass screen center portion I (2), low pass I (3) and low pass I (1) shown in FIG. 49 is extracted, and the respective data lengths are detected. Address information is created from the data length of each area detected as described above. On the other hand, in the case of P and B pictures, the data length is detected in picture units and address information is created. When the screen is not in the center center priority mode, the ninth embodiment is followed.

Next, determination of the hierarchical mode is performed, and when it is not the hierarchical mode, information indicating that the hierarchical structure is not hierarchical is inserted into the system stream, and the conventional stream configuration is followed. In the hierarchical mode, the setting of the sequence header is checked. Specifically, the data of the sequence scalable extension is checked. If this is correctly described, the picture header recognizes the head of the picture, extracts low-order data of the rearranged I pictures and P pictures on the screen area, and detects the data length. On the other hand, the B picture detects the data length for each picture.

In addition, a packet is written that records only the address information in the case where the lower center of the screen of the I picture and the P picture is collected at the head of the GOP. This packet contains the address center of the screen center part, the screen peripheral part of the I, P picture, the high part of the I, P picture, and the B picture, and the data length of each data is recorded. Therefore, the head position of each data stream is obtained as the relative address at the head of the GOP header by this data length.

FIG. 59 shows the created bit stream. As shown in FIG. 59C, the low-order regions of I pictures and P pictures rearranged in area units are arranged at the head of the GOP, so that the packet information is converted into a private 2 packet as shown in the flowchart of FIG. The stream in the case of arrangement in FIG. 59 is shown in 59d. In this case, the address information may be expressed in the relative address at the head of the gop header as described above. However, the number of packets of the first byte may be expressed as the head of each picture, or the like. Of course, it may be expressed in a dress or the like.

FIG. 60 shows an example in which address information is put in a private 2 packet. When the PES packet is a private 2 packet, the stream ID is set, and the layering mode, the layering type, the type of image used during special playback, the number of start addresses, and the like are recorded. Here, the start address is the start address of the lower portion of the screen around the I picture, the start address of the lower portion of the screen around the I picture, the start address of the lower region of the four P pictures, and the start address of the remaining B pictures.

In addition, the sector addresses on the disk of the front and rear GOPs for jumping the optical head by special reproduction are additionally recorded. In this case, in addition to the address of the front and rear GOPs in consideration of the double speed of the special playback, and further recording the sector address of the front and back GOP, the amount of change in the double speed of the special playback is widened. Although the address information is recorded in the private 2 packet of the PES packet, it is of course possible to record the private descriptor of the program stream map, another user area, or the like. The reproduction side of the eleventh embodiment will be described with reference to FIG. 61 is a block diagram of a digital video signal decoding processing unit. In addition, the same code | symbol is attached | subjected to the same or equivalent part in drawing, and description is abbreviate | omitted.

Next, the operation of FIG. 61 will be described with reference to FIG. Fig. 62 is a flowchart showing the operation of the format decoder during playback. The bit stream dispatched in the ECC detects the header of the program stream and separates the PES packet. In addition, the header of the PES packet is detected, and the discrimination between the private packet and the video packet including the address information is performed.

In the case of a private packet, address information contained in the picker is extracted and stored. On the other hand, in the case of a video packet, a bit stream of video data is extracted. In the case of the private packet and the normal reproduction or the video packet, the low frequency component and the high frequency component data are extracted from the bit streams of the video data of the I picture and the P picture to rearrange the data and perform the rearrangement. Outputs

On the other hand, in the case of the private packet and the special playback, first of all, it is judged whether or not there is time to play all the low-order I pictures, and if there is time to restart, there is also time to play the low-order P picture. The judgment is made whether or not there is. When the above two or one judgments are performed and there is time to play low-order I pictures and P pictures, when there is time to play local I pictures and P pictures, low-order I pictures and P pictures are played back. do. If there is a time for reproducing the entire low frequency I picture, but there is no time for reproducing the low frequency P picture, only the low frequency I picture is reproduced. If there is no time to reproduce the entire low picture of the I picture, the center portion of the screen of the low picture I picture is reproduced. In the above three cases, after the mode playback, the over head is jumped to the head of the next GOP.

When the addresses are described in the video stream, the address information is extracted and recorded after the video stream is formed. When the address information is described in the private descriptor of the program stream map, the address information is stored at the program stream header detection level. Extraction and storage are performed. It goes without saying that the address information may be either the relative address or the absolute address of the program.

In practice, as shown in FIG. 61, skip search, normal continuous playback, etc. in the microcomputer. The mode signal is input to the mode switch 76. On the other hand, the reproduced signal from the disco is amplified by an amplifier, digitally demodulated by reproducing the signal by a clock subjected to phase synchronization output by the PLL, and subjected to error correction processing to restore the program stream. In addition, the program stream header detector 71, which detects the header of the program stream, obtains information of the data following the header. In addition, the PES packet header detector 72 stores, for example, the picture unused special reproduction data (low-range data and data arranged by the area of the screen) recorded in the private 2 packet of the PES packet. Address information is detected, and the information is deposited in the address memory 75. Here, it is determined whether the PES packet for audio, PES packets for characters, and the like is a PES packet for video, and only the video packets are output to the video bit stream generator 73. The video bit stream generator 73 outputs the video bit stream by deleting the header removal of the PES packet. Thereafter, in the case of normal reproduction according to the address reduction obtained from the address memory 75, the data rearranger 74 rearranges and outputs the bit stream output from the mode switch 76. FIG.

The output (control signal) of the mode switch 76 is supplied to the data reorderer 74 and the decodable determiner 84. The data rearranger 74 synthesizes and outputs the low pass component and the high pass component which are layered by the control signal in the case of normal reproduction and rearranged for each area. On the other hand, during the special reproduction, data of only the low pass component or only the low pass component in the center portion of the screen is output to variable length decoding (77). That is, during normal playback, the low-pass components of the I and P pictures are operated in the order of the screen area, and during the special playback, the area of the center of the screen of the low-pass component of the I picture or the center of the screen of the low-pass component of the I picture and the P picture by double speed. The area of the output is switched. In addition, the time stamps of PTS and DTS are not used for special playback using only the low pass component.

On the other hand, the variable length decoder 77, together with the decodable determiner 84, extracts the end of the low-pass region event indicated by the slice header's priority breakpoint, decodes it to the end, and switches 78 Will output Although the number position 78 is connected so as not to insert 0 during normal playback, it is controlled by the decodable determiner 84 so that 0 is inserted into the high frequency component after the priority break point during special playback. The operation concept of the low pass decoding is the same as that in FIG. 44, and thus description thereof is omitted. The rearrangement on the area of the screen at this time is the same as that described in the tenth embodiment, and thus description thereof is omitted.

Incidentally, in the above embodiment, the coding region is executed at the end of the event. For example, the data may be divided at the end of a fixed number of events, or the data may be divided into data that is roughly quantized by the quantizer 54 and the difference between fine quantization and rough quantization. The buffer memory may be divided by encoding an image obtained by halving the spatial resolution and a half of the resolution thereof by encoding the difference image between the original image and the original resolution image. That is, it is a matter of course that the high efficiency encoding data of the image may be divided not only by the division of the frequency domain but also by the division of quantization and spatial resolution. In this case, data that is more important as a flower is data of a low frequency region in the case of frequency division, data that is roughly quantized and coded in the case of division by quantization, and data that encodes a picture that has been separated by data that is divided into spatial resolution. . In this case, the important data is an area in which a reproduced image decoded using only these data is more perceptible to humans. In other words, it is possible to divide one high-efficiency encoded data into more fundamentally important data and less important data so that only important data is basically played back during special playback on the disc.

Incidentally, the eleventh embodiment has been described in which the recording side and the reproduction side are associated with each other. However, it is also possible to consider the case where recording and reproducing are performed like a hard disk. In addition, the case of only the reproduction side on the premise that it is recorded according to the assumption like a current compact disc can be considered. It goes without saying that there are screen output methods shown in Figs. 54 and 55 of the tenth embodiment for the components rearranged for each area unit of the screen. Of course, the rearrangement of the area units of the screen can be realized in the surface prediction data decoding circuit 82 or the frame memory 83 using the data of the slice vertical position in the slice header. In this embodiment, only the basic data of the I picture is divided in the area of the screen. For example, the data may be divided in the low range of the P picture or else.

Example 12

Hereinafter, Example 12 of this invention is demonstrated also referring a surface. FIG. 63 is a block diagram showing a digital image signal coding processing unit in the digital aspect recording and playback apparatus. In FIG. 63, numerals 101 and 104 are preprocessors, and numerals 102 and 105 are motions. Vector detector, 103 is resolution converter, 106, 107 is subtractor, 108, 109 is DCT circuit, 110, 111 is quantizer, 112, 113 is Variable length encoder, 114, 115 are inverse quantizers, 116, 117 are inverse DCT circuits, 118, 119 are adders, 120, 121 are image melodies, Reference numerals 122 and 123 denote rate controllers, 124 denote resolution inverse transformers, and 125 denote data reconstruction as data arranging means. 63 shows the first encoding means and the second encoding means as one example. In particular, the subtractor 106 here outputs the difference component in both encoding sections by the first encoding means and the second encoding means.

Next, the operation will be described. Video data is input to the resolution converter 103 in the order of restless scans. The input video data is filtered out by the resolution converter 103 so as to prevent the return noise of the high frequency region. 64 is an explanatory diagram which expresses the concept of this resolution conversion on an image. For example, the data of horizontal 704 pixels and vertical 480 pixels is, for example, petered out, and is decimated into images of horizontal 352 pixels and vertical 240 pixels of 1/2 times horizontally and vertically, and are converted into low resolution screen data.

The low resolution screen data is input to the preprocessor 104 to convert from raster scan to block scan. Note that block scanning here means that data is sent in the order of DCT blocks. The I picture is encoded without performing inter-frame operations that use the output of the frame memory to perform intra-frame encoding.

In the case of an I picture, since the input image memory 121 of the subtractor 107 outputs nothing, the video signal passes through the subtractor 107. This data is orthogonally transformed into a frequency component in the DCT circuit 109, and the orthogonal transform is input to the quantizer 111 in a zigzag scan in the low frequency region and quantized. Further, the quantized image data is skin coded into the entry via the variable length encoder 113, and output to the data generator 125. FIG.

On the other hand, the image data quantized by the quantizer 111 is inversely quantized by the inverse quantizer 115 and inversely transformed from the frequency component data to the spatial component data by the inverse DCT circuit 117. The I picture is decoded without performing interframe calculation using the output of frame memory which is the intraframe encoding. Therefore, in the case of an I picture, since there is no input from the image memory 121 of the adder 119, the data passes. The output of the adder 119 is used as data accumulated in the image memory 121. At least I pictures or data of I pictures and P pictures must be stored in the image memory. This is because in MPEG1 and MPEG2, reference data is required for I picture and P picture in order to decode B picture.

In addition, the image memory 120 averages the image decoding data by weighting the output of the decoded data adder 118 and the result of the pixel number being returned to the original by interpolating the pixels by the resolution inverse converter 124. To accumulate. Hereinafter, for the sake of simplicity, the case where the output of the resolution inverse converter 124 is used as 1 and the output of the adder 118 is used as 0 load is recorded.

The input video data is buffered by the preprocessor 101 to perform scan conversion from raster scan to brook scan and accumulate the data and subtractor in the image memory 120 accumulating the signals subjected to the above-described low resolution processing. 106) (this is referred to as resolution residual difference component). The resolution residual difference component is orthogonally transformed into the frequency domain by the DCT circuit 108, and then converted into a scan from the low frequency domain to obtain proper quantization by the magnetizer 110. This data is entropy encoded via the variable length encoder 112 and output to the data reconstructor 125.

On the other hand, the data quantized by the quantizer 110 is inversely quantized by the inverse quantizer 114, and inversely transformed by the inverse DCT circuit 116 into data in the spatial domain. The adder 118 performs inverse conversion data of the low resolution processing input from the image memory 120 and addition of the output of the inverse DCT circuit 116, thereby resolving at least one of the low resolution data and the data other than the low resolution data. As an example, the result of decoding the data in two layers by the data of the previous filter component is obtained and stored in the image memory 120. This layer is determined by the number of resolution conversions to be performed, and it is also possible to perform two resolution conversions to make three series, and a detia of any layer can be created by the same approach.

In order to encode B pictures by performing bidirectional prediction from I pictures and P pictures for encoding such as MPEG, the I pictures and P pictures are decoded and accumulated as decoding data. In this manner, the I picture and the P picture are encoded, and then the B picture processing is executed.

The coding processing of the I picture, the P picture, and the B picture as described above is performed for both the low resolution component and the resolution residual difference component. In this way, a sequence in which the low resolution component R (hereinafter referred to as R component) and the resolution residual component S are mutually arranged can be configured. The data reconstructor 125 operates to collect and arrange the GOPs in a place suitable for accessing the optical head such as the head of the GOP, and so on.

When rearranged in this way and the first half of the region occupied by the l component is read, the low resolution component is reproducible. The resolution residual component has less data than the comparative residual component, so that efficient hierarchization can be performed. In other words, here, the first encoding means for encoding according to a predetermined condition and the remaining information by encoding using the first encoding means as one example of the image information other than that encoded by the first encoding means among the image information. Second encoding means for encoding the difference component is provided to perform a favorable hierarchical layering.

65 is a diagram showing one example of the result of the data structure. In FIG. 65, sequence a is a sequence generated by the encoding process of the present embodiment, sequence b is a sequence generated by the encoding process of another embodiment, The sequence c is a sequence generated by the encoding process of another embodiment. In the sequence b, L denotes a low frequency component, H denotes a high frequency component, C denotes a component encoded by coarse quantization, and A denotes a residual component obtained by coarse quantization, respectively. In addition, as shown in the sequence a of FIG. 70, only the I picture and the P picture may be executed, and only the components may be arranged at the head of the GOP.

In this way, if only low-resolution components are collected and arranged at the head of the GOP, the ratio of the L component to the whole can be greatly reduced, and the read speed from the medium can be freed, thereby making it possible to easily implement skip search. In addition, as shown in the sequence a of FIG. 70, when only the R component of the I picture and the P picture are arranged at the head of the GOP, the arrangement is performed so as to decode only the low resolution data of the I picture and the P picture. Incidentally, in the above-described embodiment, the case where the thinning pot is 1/2 times horizontal and 1/2 times vertical has been described. However, this ratio may be other than this and may be applied to any ratio.

Also, there are MPEG1, MPEG2, JPEG and the like for the encoding method, but it is not necessary to use a common encoding method in the resolution layer. This is because, in the case of encoding at a lower resolution, even the MPEG1 system without mentioning the interlace can sufficiently cope with the encoding. In addition, since the stacking for each frame is a moving image even in the JPEG system corresponding to the stationary state, it is possible to occupy any specific position of the GOP and to decode it normally. In addition, although this description was demonstrated in two resolution layers, it is a matter of course that even more hierarchical layers may be sufficient. The data of the low resolution component is encoded by the first encoding means in FIG. 63, and the difference component with the image before the interpolation is performed on the output from the first encoding means and the pixel subtraction is subtracted. Obtained by (106), a differential component encoding means for performing encoding on the differential component may be provided to perform encoding on the differential component.

The frame to be read from the image memory is usually taken from the predictive reference frame. However, since there are low resolution frames, it is necessary to accumulate and read the memory in a well-aligned time axis (by addressing the memory). It is a matter of course that an information adding means may be provided to add the bar value information such as an audio signal, a header, and an error correcting code to data of the residual difference component.

Example 13

Description of Embodiment 13 of the present invention will be described with reference to FIG. In this embodiment, the DCT block is divided into layers of the low frequency region and the high frequency region, and only the low frequency region is arranged at the head of the GOP. FIG. 66 is a block diagram of the digital video signal encoding processing unit. In FIG. 66, reference numerals 126 and 127 denote first variable length encoders and second variable length encoders, respectively. In addition, in FIG. 63, the same code | symbol is attached | subjected to the same or equivalent part, and description is abbreviate | omitted.

Next, the operation will be described. The video data of this interlace is, for example, data of 704 pixels vertically and 480 pixels vertically. Since the I picture is decoded without performing interframe operations using the output of the frame memory to be encoded in the frame, video data is passed through and output. This video data is orthogonally transformed into a frequency component by the DCT circuit 108, and then converted into a scan from a low frequency region, so that the quantizer 110 makes proper quantization.

Fig. 67 shows the data values of the DCT coefficients inside the DCT block. In FIG. 67, data of the DCT coefficients of the high frequency component are arranged in the upper left portion for the upper left portion and in the lower right portion for the lower right portion. Among the data of the DCT coefficient arranged in the DCT block, the data of the DCT coefficient in the low frequency region (for example, the hatched portion in FIG. 67) up to the data of the DCT coefficient at a specific position is used as the low frequency region extraction means. Entropy coded through the variable length encoder 126 of 1 is output to the data reconstructor 125. Further, the second variable length encoder 127 performs variable length coding on the data of the DCT coefficient after the data of the DCT coefficient at the specific position. In other words, the data is divided and encoded in the frequency domain.

As for the encoding of the motion vector or the DC component, only the first variable length encoder 126 is sufficient, and the second variable length encoder 127 is not necessary. This is because in normal playback, the output data of the first variable length encoder 126 and the output data of the second variable length encoder 127 may be synthesized and decoded.

In addition, although the encoding region performs the determination at a fixed position of the DCT coefficient, other methods may be used, for example, it may be determined as a fixed number of event numbers. In other words, the unit to which Huffman protection, which is a variable length code, is given is an event, and the coding region may be set to a predetermined number of events such as three units. As an example of the output bit stream of the data reconstructor 125, if the low frequency region of the first half is read as shown in the sequence b of FIG. 65, the image of the low frequency region can be reproduced. In addition, it may be performed with anomalies in the same arrangement as the squeeze b of FIG.

On the other hand, the data quantized by the quantizer 110 is inversely quantized by the inverse quantizer 114 and inversely transformed into data in the spatial domain by the inverse DCT circuit 116. The I picture is decoded without performing interframe calculation using the output of the frame memory to be encoded in the frame. Therefore, in the case of an I picture, since there is no input from the image melomi 120 of the adder 118, the data passes. The output of the adder 118 is used as data accumulated in the image memory 120.

At least I pictures and P pictures must be stored in the image memory. This is because even in MPEG1, even for MPEG2, the I picture and the P picture are required as reference data for decoding the B picture.

In such a configuration, the proportion of the L component as a whole is greatly reduced, and the lead speed from the medium is freed, thereby making it impossible to realize skip search or the like. If only the I picture and the P picture are collected and arranged as described later, it becomes possible to operate to easily decode only the data of the low frequency components of the I picture and the P picture. Since the data in the high region in the frequency domain has less data than the data in all the regions, it is possible to efficiently organize the data than extracting the data in the low frequency region and storing it in front of the data in all regions.

When the I picture ends encoding by encoding through the subtractor 106, the B picture is encoded by bidirectional prediction with the last P picture of one GOP in time. By comparing the output of the preprocessor 101 with data from the reference frame memory (omitted on the screen), the detection of the motion vector and the determination of the prediction mode, the frame structure, and the like are carried out. 101) and the data of the memory of the reference frame to which the data from the memory of the reference frame is most suitable are read in the image memory 120 as data of the forward and backward directions, and the read data and the B picture preprocessor. (The output result of 101_ is subtracted by the subtractor 106. (These results are referred to as time residual components for both P and B pictures.) The DCT operation is performed on the residual residual components, and the results are quantized. And the variable length code.

Example 14

A fourteenth embodiment of the present invention will be described with reference to FIG. In this embodiment, as an example of data other than the roughly quantized component and the roughly quantized component of the DCT coefficient, the roughly quantized component is divided into hierarchical layers of roughly quantized components so that the roughly quantized component is placed at the head of the GOP. FIG. 68 is a block diagram showing a digital aspect signal coding processing section. In FIG. 68, 128 is a subtractor, and 129 is an adder. In Fig. 63 in the drawing, reference numeral 128 denotes a subtractor and 129 an adder. In addition, the same code | symbol is attached | subjected to FIG. 63 in FIG. 63, or an equivalent part, and abbreviate | omits description.

Next, the operation will be described. The input image data of this interlace is, for example, data of an effective screen scene of 704 pixels horizontally and 480 pixels vertically. One picture is decoded without performing interframe calculation using the output of the frame memory to be encoded in the frame. Therefore, in the case of one picture, decoding which is an input of the subtractor 106 is performed. Therefore, in the case of one picture, the image memory 120 which is an input of the subtractor 106 outputs nothing, so that the video signal passes through the subtractor 106. This data is orthogonally transformed into a frequency component by the DCT 108, and then converted into block scans from the low frequency domain, and the quantizer 110 makes the appropriate approximate quantization in which the amount of encoded data is less than half. The quantized data is antropy encoded via the variable length encoder 112 and output to the data reconstructor 125.

On the other hand, the data quantized by the quantizer 110 is dequantized by the inverse quantizer 114 (the result is roughly quantized), and the dequantized data is converted into another encoding processor (68). In the figure, the data is sent to the dotted line, and is inversely converted into data in the spatial domain by the inverse DCT circuit 116. Since the coding of one picture is described here, there is no output from the image memory 121, but since the decoding result of this code processing unit is normally stored in the image memory 121, the motion vector detector is based on the data. The motion vector detection and prediction mode are determined by 102, and the DCT block mode decision is made, and data of a position suitable for the determined mode is referred to and entered on the subtraction input side of the subtractor 107.

DCT is applied to the output of the subtractor 107, and the result of roughly quantizing and the residual difference is calculated by the subtractor 128 (this is called roughly quantized residual). The roughly quantized residual difference is finely quantized by the quantizer 111 (fine quantization as much as almost normal coding considering code rate control), variable length coding, inverse quantization, inverse DCT, decoding, and image Accumulated in the memory 121. The encoding result and the rough quantization encoding result are determined by the data reconstructor 125, and the header and the like are added at the same time.

As an example of this output data, as shown in the sequence c of FIG. 65, the decoding result of the roughly quantized image can be obtained simply by reading the first half of the GOP. In addition, since the data of the roughly quantized residual difference is smaller than that of the data point of fine quantization, the data can be more efficiently constructed than the roughly quantized data is extracted and accumulated before the finely quantized data.

As another example, anomalous processing such as arrangement of the sequence c in FIG. 70 may be performed. In this way, the proportion of the C component (component encoded by rough quantization) is greatly reduced, and the read speed from the medium is increased to allow skip search and the like. As described later, when only the I picture and the P picture are collected and arranged, it operates so as to decode only data obtained by roughly quantizing the I picture and the P picture.

When the I picture finishes encoding by encoding through the subtractor 106, the B picture is encoded by bidirectional prediction with the last P picture of one GOP in time. The output of the preprocessor 101 and the data from the reference frame memory (arrows in the drawing) are compared to detect the motion vector, determine the prediction mode, the frame structure, and the like, and according to the determination result, the preprocessor 101 ) And the data of the reference frame with which data from the memory of the reference frame is most suitable are read in the image memory 120 as data of the forward and backward directions, and the read data and the preprocessor 101 of the B picture. The result of subtraction is subtracted by the subtractor 106 (these results are referred to as time residual components for both P and B pictures). Data in the forward direction and the backward direction is read from the image memory 121, and the data and the output of the preprocessor 101 are subtracted by the subtractor 107 to encode orthogonal transform entropy. Also, the P picture is encoded using the same process.

FIG. 69 shows an example of the statistical amount of encoded data, but shows the deviation of the code amount when the number of frames in the GOP: N = 15, the period of the picture and the P picture: M = 3. In this figure, it is understood that the I picture and the P picture are about 50% of the total. Therefore, if only this part or only the I picture is divided as in the above-described embodiment by dividing the layer by resolution, frequency, and quantization, the amount of code to be reproduced is further reduced. Therefore, the travel time of the optical head can be shortened, and functions such as skip search can be easily realized.

The processing sequence at that time is shown in FIG. 70 shows an arrangement of I pictures, P pictures, and B pictures of an original picture, and executes the above-described processes of Embodiments 12, 13, and 14 for only I pictures and P pictures therein, and divides the hierarchy for B pictures. Encode without. According to the process of Example 12, the sequence which processed the I picture and the P picture is sequence b, and according to the process of Example 14, the sequence which processed the I picture and the P picture is called sequence c.

Each sequence mode collects data such that the low resolution component (R), low frequency component (L), and roughly quantized component (I) and P picture components are collected and arranged at the head of the GOP by the respective data reconstructor 125. Configure. As for the sequence a, the low resolution image of the I picture and the P picture can be decoded by only the low resolution components of the I picture and the P picture (the core portion of the 70 degree sequence a after data reconstruction). It becomes possible. Naturally, the data other than the core area need not be arranged as shown, and of course, the data may be arranged in the frame number order at the time of encoding of FIG.

As for the sequence b, a low frequency image of the I picture and the P picture is generated only in the low frequency components of the I picture and the P picture (the core area portion of the 70 degree sequence b after the data reconstruction). Since the sequence c can be decoded only the roughly quantized component of the I picture and the P picture (the core area of the 70th sequence c after the data reconstruction), the roughly quantized image of the I picture and the P picture can be decoded. It is easy to cope with skip search.

For example, in the configuration shown in FIG. 63, the B picture may be operated to be encoded only by the encoding loop including the preprocessor 101 without using the encoding loop including the preprocessor 104. In the structure shown in the figure, all frequency components may be coded in the first variable length encoder 126. In the configuration shown in FIG. 68, fine quantization may be performed in the quantizer 110 and encoded.

The most basic data such as data on the low frequency side is ideally collected at the head of the GOP, but it may be offset slightly so as to overlap the head of the unit constituting the error correcting code. Incidentally, the basic data arrangement in accordance with the unit of error correction can be similarly implemented in other embodiments.

Example 15

A fifteenth embodiment according to the present invention will be described with reference to FIGS. 71 and 72. FIG. FIG. 71 shows an example of an arrangement of DCT blocks and an arrangement scheme of frequency components in one block of bit streams. FIG. 71A shows a DCT block of a macroblock header and a luminance signal with respect to the entire arrangement of DCT blocks. One macro block is formed by Y1 to Y4, the DCT block U1 of the color difference signal BY, and the DCT block V1 of the color difference signal RY. Fig. 71B shows macroblock headers and DCT blocks Y1L to Y4L of the luminance signal, DCT blocks U1L of the color difference signal BY and DCT blocks V1L of the color difference signal RY for the arrangement of the DCT blocks of the low frequency component. It expresses what forms one low-pass macroblock.

Fig. 71C shows a macro of one high frequency component in accordance with the DCT blocks Y1H to Y4H of the luminance signal, the DCT block U1H of the color difference signal BY and the DCT block V1H of the color difference signal RY with respect to the arrangement of the DCT blocks of the high frequency component. It represents what forms a block. 71D shows the concept of arrangement of frequency component data of one block of bit streams. 72A and 72B are explanatory diagrams showing a block diagram of a digital video signal decoding processing unit and an operation concept thereof. In Fig. 72A, reference numeral 130 denotes a mode switch as a mode switching means, and 131 denotes a data rearrangement means. The data rearranger as 132 is a decodable determiner, 133 is a variable length decoder, and 134 is a switch as a switch and the decodable determiner 132 and a switch 134 constitute data manipulation means. Reference numeral 135 denotes an inverse quantizer, 136 denotes an inverse DCT circuit, 137 denotes an image memory, 138 denotes an adder, and 139 denotes an inverse scan converter.

Next, the operation will be described. The data in Fig. 71 is, for example, a code string assembled into eight bits (one byte) in the longitudinal direction, and information on the macroblock even if it is a macroblock is recorded in each macroblock. This information is, for example, information such as an increment address, a quantization scale code, a motion vector, a marker bit, a macro block pattern, and the like.

After this macroblock header, the encoded data of each DCT block continues. In this data embedding method, the byte streams are first arranged in the bit stream, and each byte is arranged in order. However, since each DCT block has a variable code length, the block boundary or the header and data boundary are completed in byte units. In most cases, the boundary exists in the middle of the byte unit. By the way, the data of each block is the data of the low frequency area as it is variable length and is closer to the macro block header side.

This data is an event (a unit of giving a variable length code, and in the case of a DC component, a single component is composed of a DC component. The AC component is a combination of a nonzero DCT coefficient and a run length in order to encode a run length. The low frequency component (L) is formed as shown in Figs. 71B and C, with the maximum fixed length code amount irrelevant to the end of the block. The encoding data is divided into and high frequency components (H).

Next, operation of FIG. 72 will be described. First, a mode signal such as skip search or normal continuous playback is input to the mode switch 130. On the other hand, the reproduced signal from the disc or the like is amplified by an amplifier, and the digital demodulation is performed by performing the discrimination operation on the output data after the signal reproduction is performed by the phase locked clock output from the PLL or the like, and the error correction process is executed. After that, the audio signal is separated from the layer of the system consisting of the data of the video signal and the data of the audio signal, and the bit stream of the video is extracted and input to the data rearranger 131.

The output (control signal) of the mode switcher 130 is supplied to the data rearranger 131 and the decodable determiner 132. The data rearranger 131 obtains a control signal and outputs the control signal to the variable length decoder 133 by reconnecting the data before division in the L component and the H component of FIG. The variable length decoder 133, together with the decodable determiner 132, extracts the end of the event of the L component region, decodes it to the end, and outputs it to the switch 134. This switch 134 is connected so as not to insert 0 during normal playback. The switch 134 controlled by the output of the decodable determiner 132 performs input from the decoded low frequency component to the DCT block, and configures the entire DCT block such that 0 is inserted into the high frequency side of the DCT block. .

In decoding, the DCT block constructed as described above is corrected by the motion vector of the reference I picture or P picture in the case of P picture so that the data is reversed and the output of the adder 138 is passed in case of the I picture. In the case of the B picture, the read of the image memory 137 is controlled in accordance with the case of each picture so that the motion vector is corrected and added in both the I picture and the P picture as the reference, and the adder 138 Is added by.

The DCT mode and prediction mode motion vectors at this time are controlled according to the information obtained by decoding the encoding of the header. In accordance with the above-described process, the motion compensated predictive data is decoded and accumulated in the image memory 137, and the image is restored in the original configuration order of the GOP. The inverse scan converter 139 performs buffering and block scanning to raster scanning in the order of output of the images.

In addition, although this embodiment expresses the fixed length even if the skip of the macroblock or the predetermined fixed length data is shorter, the L component can be reliably extracted by sequentially detecting the EOB even if the data is shorter than the fixed length data. It goes without saying that connecting the L component data to the next block does not cause any problem. In addition, of course, when the predetermined data length is exceeded, the EOB may be added as L component data even to the event partition. Although not particularly shown in the description of the above embodiments, it is a matter of course that an information adding means for adding an additional information such as an audio signal, a header, and an error correction code may be further provided and added to the data for the high frequency region.

Example 16

Next, a sixteenth embodiment of the present invention will be described with reference to FIGS. 73 and 74. 73A and 73B are explanatory diagrams showing a block diagram of the digital video signal coding processing unit and the operation concept thereof. In FIG. 73A, reference numeral 140 denotes a rate controller. Here, the first variable length encoder 126 and the second variable length encoder 127 are provided as encoding means. In addition, in FIG. 63, the same code | symbol is attached | subjected to the same or equivalent part, and description is abbreviate | omitted.

Next, the operation will be described. The input image data of the interlace is buffered by the preprocessor 101 to convert the last scan into a block scan. The I picture is decoded without performing interframe calculation using the output of the frame memory to be encoded in the frame. Therefore, in the case of an I picture, the image memory 120, which is an input of the subtractor 106, outputs nothing, so that the video signal passes through the subtractor 106.

This data is orthogonally transformed into a frequency component by the DCT circuit 108, and then converted into block scans from the low frequency region and quantized appropriately by the quantizer 110. The data of the low frequency region up to the data of the DCT coefficient at any particular position among the quantized data are entropy coded through the first variable length encoder 126 and output to the data reconstructor 125.

The second variable length encoder 127 also variable length encodes the DCT coefficient after the data at the specific position. As for the encoding of the motion vector and the DC component, only the first variable length encoder 126 may be used at least. At this time, in order for the weight of the L component to be converted into a rate without code limitation, it is necessary to add an EOB to both the L component and the H component even though it is one block. In this way, by temporarily placing the EOB code in the partition portions of the L component and the H component, the boundary between the L component can be changed by the rate. On the other hand, the data quantized by the quantizer 110 is inversely quantized by the inverse quantizer 114 and inversely transformed into spatial data by the inverse DCT circuit 116.

The I picture is decoded without performing interframe calculation using the output of the frame memory to be encoded in the frame. Therefore, in the case of an I picture, since there is no input from the image memory 120 of the adder 118, the data passes. The output of the adder 118 is used as data accumulated in the image memory 120. At least I pictures or data of I pictures and P pictures must be stored in the image memory. This is because, normally, MPEG-1 and MPEG2 are required as reference data for I picture and P picture for decoding of B picture.

When encoding of the I picture is finished, encoding of the B picture is performed next by bidirectional prediction with the last P picture of all the GOPs. Motion vector detection and prediction mode by comparing the output of preprocessor 101 with data from the memory of the reference frame (arrows in the figure). Determination of the frame structure and the like is executed, and according to the determination result, the image memory 120 forwards and backwards the data of the reference frame memory in which the output of the preprocessor 101 and the data from the reference frame memory are most suitable. The data is read together with the data of minutes, and the data and the output result of the preprocessor 101 of the B picture are subtracted by the subtractor 106 (this result is referred to as a time residual component of both the P picture and the B picture). At this time, the DCT is quantized and variable length coded. When divided into the low frequency region and the high frequency region, the rate is not constant with the frequency components included in the figure. Therefore, the data rate of the low frequency region also does not become constant, so that the possible range of actuator control of the head cannot be completely compensated. Here, the rate controller 140 makes the region of the low frequency component variable according to the rate, and controls the size of the low frequency region to be variable as shown in FIG. 73B for the target rate.

That is, while the output of the first variable length encoder 126 is monitored, if the amount is larger than the target rate set by the application, the occupied area of the data in the low frequency region is reduced, and when the code amount of the first variable length encoder 9926 is low, the low frequency Increase the area of the area. In this way, the setting of the occupied area of the low frequency region is appropriately changed for the first variable length encoder 126 and the second variable length encoder 9227 while monitoring the code amount.

In addition to this, for example, the target rate may be set by determining the reference for setting the occupied area of the low frequency region based on the result of whether or not there is a large number of codes in the precoding.

74 is a block diagram of a digital video signal decoding processing unit, and is a block diagram of decoding processing for decoding encoded data encoded as described above. In Fig. 74, reference numeral 141 denotes an EOB searcher. In addition, in FIG. 72, the same code | symbol is attached | subjected to the same or equivalent part as FIG. 72a, and description is abbreviate | omitted. A mode signal, such as whether a skip search is performed by a microcomputer or the like, is normally input to the mode switcher 130. On the other hand, the reproduced signal from the disc or the like is amplified by an amplifier, the reproduced signal is discriminated by a clock subjected to the PLP, digital demodulated, and error correction processing is performed to separate the audio signal from the layer of the system and to remove the video bit stream. The data is extracted and input to the data rearranger 131. The output of the mode switch 130 as the mode switching means is supplied to the data reorderer 131 and the decodable determiner 132. The data reordering unit 131 obtains this control signal and operates to reconnect the data before division in the L component and the H component of FIG. 71 or the variable length decoder 133 which is a decoding means using only the L component without connecting the H component. Will print

It is not ideal that the L component is cut off in the middle of the event, but the variable length decoder 133 and the decodable determiner 132 may be used in case of decoding the signal having poor signal quality such as skip search. By checking the end of the event of the L component region, decoding to the end is output to the switch 134. This switch 134 operates to always be turned on in the case of playback data having good signal quality as in normal playback. Here, the data operation means is constituted by the decodable determiner 132 and the switch 134.

The switch 134 is controlled by the output of the decodable determiner 132 so that 0 is inserted into the high frequency side of the block in the low frequency component that has been successfully decoded to form a DCT block, and inversely DCTs the data to perform an I picture. The output of the adder 138 is passed through, and the P picture is corrected and added by the motion vector in the I picture of the reference, and the B picture is corrected and added by the motion vector in the I picture and the P picture. The leads of the image memory 137 are controlled and added by the adder 138. The DCT mode or prediction mode motion vector at this time is controlled by decoding the code of the header. In this way, the motion compensation predicted data is decoded and stored in the image memory 137, and the image is made into the original configuration order of the GOP. The inverse scanning converter 139 buffers and converts from block scan to raster scan in the order of image output.

In addition, although the above description demonstrated the example which controls the magnitude of the area | region of DCT coefficient, you may make it control the number of events. In this case, although the EOB may be given without reaching the predetermined number of events of the L component, since the EOB searcher 141 monitors the appearance of the EOB, the L component can be reliably identified. In particular, the data rearranger 131 and the EOB searcher 141 reconstruct the data according to the data of the low frequency region, the data of the high frequency region, and the EOB. That is, the data rearranger 131 and the EOB searcher 141 constitute data reconstruction means.

As a matter of course, since the energy after DCT is small, the same handling of L component and H component is preferable for the non-coded block which is not encoded. For the H component, it is ideal to run-length code the data from which the L component has been removed. However, the H component may be encoded with the L component set to zero. Since it can cope with the structure similar to the conventional variable-length coder of MPEG, it can simplify this circuit.

Example 17

A seventeenth embodiment of the present invention will be described with reference to FIG. FIG. 75 is a block diagram showing a digital video signal decoding processing unit, in FIG. 75, (142) is a multiplexer, (143) is a switch, (144) is a first variable length decoder, and (145) is Variable length decoder 2, 146 is a first inverse quantizer, 147 is a second inverse quantizer, 148, 149 is an adder, 150, 151 is an image memory, 152 is a resolution inverse converter. 75 shows a low resolution decoding section as a decoding means. In addition, in FIG. 72, the same code | symbol is attached | subjected to the same or equivalent part as FIG. 72a, and description is abbreviate | omitted.

Next, the operation will be described. FIG. 75 may be considered to correspond to a processing block for video data of a reproduction signal from a disc when the encoded data as described with respect to FIG. 68 is recorded on an optical disc or the like. A mode signal such as whether skip search or normal continuous playback is performed by a microcomputer or the like is input to the mode switch 130 as the mode switching means. On the other hand, the reproduced signal from the disc or the like is amplified by an amplifier, the reproduced signal is discriminated by a clock applied to the PLL, digital demodulated, an error correction process is performed, and the audio signal is separated from the layer of the system. Extract

The bit stream of the extracted video is input to the multiplexer 142. The multiplexer 142 sends data of the low resolution component to the second variable length decoder 145, and sends other data to the first variable length decoder 144 via the switch 143. The switch 143 is controlled by the mode switch 130, and although the mode requires only the reproduction image output of the low resolution component by skip search or the like, it is redundant when the residual resolution difference component is reproduced in the middle. It stops the transmission of the data. In normal playback, this switch 143 operates to remain connected.

The second variable length decoder 145 decodes the Huffman code and the run length code, is dequantized by the second inverse quantizer 147, and is decoded from the frequency component to the spatial component by the inverse DCT circuit 136. Is converted.

In the case of an I picture in the converted data, the adder 149 is passed through and accumulated in the image memory. In the case of the P picture, the first P picture is read from the I picture stored in the image memory, and after the second P picture is referred to the P picture immediately before being stored in the image memory, the position is corrected for motion vectors, and read. The adder 149 performs motion compensation prediction. In the case of the B picture, the same operation is performed in accordance with the I picture and the P picture. In the figure, motion vectors, quantization parameters for inverse quantization, prediction modes, and the like are output from a variable length decoder and are the same as those shown in FIG. The loop shown in FIG. 75 as a dotted block is a component block for decoding the low resolution component, but the decoding result is interpolated by the resolution inverse transformer 152 as the interpolation image generating means and is used as the residual resolution component. It is input to the image memory 150 to supplement the decoding result.

During normal playback, decoding of the residual residual component is output as an image by the inverse scan changer 139 in combination with the decoding result of the low resolution component (in accordance with the division processing when this is time-divided). The first variable length decoder 144 can decode the frequency components via the switch 143, dequantize it by the first inverse quantizer 146, and inverse the DCT circuit 136. Is decoded into the resolution residual component data of the spatial domain.

The image memory 150 is referred to the pixel interpolation data of the low resolution component, the P picture is an I picture, the B picture is an I picture, and the P vector is read out from the image memory 150 by the position correction for the motion vector. The motion compensation prediction is decoded by 148.

In the case of a steep search or the like, unnecessary data is output to the output of the inverse DCT circuit 136 by stopping the output of the resolution residual component in order to prevent the resolution residual component from being reproduced halfway through the switch 143. Prevent output. Therefore, only the data with pixel interpolation of the low resolution component is passed through the image memory 150 (even if a switch is provided at the input portion of the image memory 150), and the image output is performed through the inverse scan converter 139. do.

Example 18

Next, Example 18 of the present invention will be described with reference to FIGS. 76 and 77. FIG. FIG. 76 is a block diagram showing a GOP address generating unit and a disk control unit, and particularly shows a processing block in the case where the rate information is recorded in a sequence header or the like. In Fig. 76, reference numeral 153 denotes a register, reference numeral 154 denotes a GOP address calculator, and reference numeral 155 denotes an optical head / disc rotation control converter as head position converting means and disk rotation control converting means. 77 is a block diagram showing a GOP address generation unit and a disc control unit including a reproduction process, in particular, a structure for executing reproduction of a GOP on a disc in which the rate information is collected and recorded at several locations on the disc. It is shown. In FIG. 77, reference numeral 156 denotes a reproduction amplifier, 157 denotes a digital demodulator, 158 denotes an error corrector, 159 denotes a system layer processor, 160 denotes a rate information memory, and a system layer processor 159 and The rate information memory 160 constitutes data rate information extraction means. Reference numeral 161 denotes a GOP number counter, which constitutes position information calculating means by the GOP address calculator 154 and the GOP number counter 161.

Next, the operation will be described. As described in the prior art, by changing the rate per 1 GOP, the overall rate of one program can be optimized, so that the image quality is considerably improved. However, if the head of the 1GOP does not look at the contents of the data, it is impossible to know whether the head of the GOP is the head of the GOP. In this case, you have no choice but to search the data on disk and search for its location.

Therefore, in such a case, first, the rate control of the variable rate is set as discrete rate targets, for example, 1 M bit, 1.5 M bit, 2 M bit, 2.5 M bit, 3 M bit, etc. per GOP. Each rate information is recorded on the disc. In particular, TOC (Table Of Contents: a recording area is provided in the first part of the disc, and information such as a title and a recording time is recorded) as a predetermined area on a recording medium such as a disc. It is most efficient to record the rate information for each GOP on the back and the like. The rate information for the GOP may be incorporated into a sequence header of a video bit stream. For example, a two-hour software is composed of 14.4k GOPs. At this time, the rate information can be expressed in three bits because it can be distinguished for five types of rates in the above case.

Therefore, it is possible to record the rate of all GOPs on the disc with information of 5.4k bytes (14.4k pieces x 3 bits ÷ 8 bits / byte).

For example, the rate information of each GOP is stored in the rate information memory 160 shown in FIG. 77, and the information length corresponding to the value is accumulated to enable fast access to the desired GOP.

A description with reference to FIG. 76 is as follows. The variable length decoder 133 decodes the Huffman code and the run length code, and decodes the header to determine the motion vector, the type of the picture, and the like.

Meanwhile, the sequence header and the like are also decoded, and the rate information is input to the GOP address operator 154. In addition, the address information of the currently accessed GOP is stored in the register 153, the GOP head address address of the next access destination is calculated, stored in the register 153, and the optical head is stored to the head of the GOP of the next access destination. The position of the optical head is obtained from the address using the / disc rotation control converter 155. Next, a control signal for the next access is calculated from the difference between the GOP currently being accessed and the head address of the GOP of the access destination. In accordance with this control signal, position control and disc rotation control of the optical head actuator are executed.

The reproduction processing will be described with reference to FIG. The rotation of the optical head and the optical disc is controlled so as to read directly from the TOC area or the area corresponding thereto, either directly or indirectly (the rate information base address is specified and then accessed to this address portion to read the rate information), and the reproduction signal from the optical head is controlled. The signal is amplified by the reproduction amplifier 156, digital signal is detected by the digital demodulator 157, discriminated as a digital signal, and digital demodulation is performed.

The digital demodulation is performed and digital reproduction data is input to the error corrector 158 to correct an error contained in the reproduction signal. The data after performing this error correction is separated by the system layer processor 159 into an audio bit stream, a video bit stream, and other data.

For example, the stream path is classified by determining which data type (binary data such as AV (video and audio) data, text data, program, etc.) is suitable for this signal. Among them, the rate information as described above is stored in the rate information memory 160.

On the other hand, information on the number of GOPs to be processed is generated using the GOP number counter 161, and the information according to the address calculated by the GOP address calculator 154 is generated. Controls the rotational speed of the head actuator and disc.

In the above description, the GOP number counter 161 shows an example of receiving a signal from the system layer processor 159, but a part taking a user I / F such as a microcomputer substitutes processing or moves to skip search or the like during playback. In some cases, it is better to input address data from a variable length decoder or the like which processes the video bit stream.

Example 19

Next, operation of the nineteenth embodiment of the present invention will be described with reference to Figs. 78, 79, and 80. FIG. 78 is a diagram showing a signal processing unit in the case of performing division by frequency and quantization in the digital signal reproducing unit, especially for reproducing processing when rate information is collected and recorded in several places on the disc. A block diagram of the configuration. In addition, the same code | symbol is attached | subjected to the same or equivalent part as the above-mentioned example in drawing, and description is abbreviate | omitted.

At the beginning of disc playback, rotation of the optical head and the optical disc is controlled to read directly or indirectly from the TOC area or the area corresponding thereto, and the reproduction signal from the optical head is reproduced by the reproduction amplifier 156. This signal is amplified and detected by the digital demodulator 157, discriminated as a digital signal, and digital demodulated. As a result, the reproduction signal made of digital data is input to the error corrector 158, and the error contained in the reproduction signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 159, and other data is also processed.

For example, the stream path is classified by determining whether the signal is binary data such as AV data, text data, or program. Among them, the rate information as described above is stored in the rate information memory 160. On the other hand, the GOP number counter 161 generates information on the number of GOPs to be processed, and calculates an address by the GOP address calculator 154 to control the rotation speed of the optical head actuator and the disk. .

In this manner, even during skip playback, the address of the GOP to be accessed is obtained and the disc is skipped. When the desired GOP is accessed, the next address is calculated in the same manner and the configuration described in Example 16 is used. The data of the obtained low frequency region is reproduced and screened.

Mode signals such as whether skip search or normal continuous playback are input from the microcomputer or the like are input to the mode switcher 130. As described above, the bitstream of the video is extracted and input to the data rearranger 131. The output of the mode converter 130 is supplied to the data rearranger 131 and the decodable determiner 132. The data rearranger 131 obtains this control signal and operates to reconnect the data before division from the L component and the H component of FIG. 71 or outputs it to the variable length decoder 133 using only the L component without connecting the H component. do.

In the present embodiment, it is not ideal that the L component is cut off in the middle of the event, but in consideration of decoding a signal having poor signal quality such as skip search or the like, the variable length decoder 133 and the decodable plate may be used. The end 132 confirms the end of the event of the L component region, and decodes the end of the event to the switch 134.

The switch 134 is controlled by the output of the decodable determiner 132 so that 0 is inserted into the high frequency side of the block in the low frequency component that has been successfully decoded to form a DCT block. The output of the adder 138 is passed through and added, and in the case of the P picture, the motion vector in the I picture of the reference is corrected and added. In the case of the B picture, the motion vector is corrected and added from the I picture and the P picture. The lead of the image memory 137 is controlled so as to be added by the adder 138 as much as possible.

The DCT mode or prediction mode motion vector at this time is controlled by decoding the code of the header. In this way, the motion-predicted data is decoded and accumulated in the image memory 137, and the image is in the original configuration order of the GOP. The inverse scanning converter 139 buffers and converts from the block scan to the raster scan in the order of output of the images.

In addition, the switch 134 is not connected to insert 0 during normal playback, and controls only the playback data to perform the playback operation.

In this case, when the encoding is divided into the low frequency region and the high frequency region, the quantization table that focuses on the low frequency side and the quantization table that focuses on the high frequency side, or evenly, a fine quantization table for one quantization table irrespective of the frequency domain is generated. Although it can be considered easily, such a case can be realized by having two variable length decoders and two dequantizers as in the local decoder of FIG. At that time, the data rearranger 131 must be a multiplexer.

Next, the operation will be described according to FIG. FIG. 79 is a diagram showing a signal processing unit in the case of performing division by bit length of a digital signal reproducing unit, and specifically illustrates an embodiment of the reproducing process in the case where the rate information is recorded in several places on a disc. This is a block diagram. The digital demodulator 157 controls the rotation of the optical head and the optical disc so as to lead directly or indirectly (designating the rate information description address) from the TOC area or the area corresponding thereto at the beginning of the disc reproduction, which is a predetermined area on the recording medium. This signal is detected by the following method, and it is determined as a digital signal to perform digital demodulation.

As a result, the reproduction signal made of digital data is input to the error corrector 158, and the error contained in the reproduction signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 159, and other data is also processed.

For example, the stream path is divided by determining whether the signal is video audio data, text data, or binary data such as a program. Among them, the rate information as described above is stored in the rate information memory 160. On the other hand, the GOP number counter 161 generates information on the number of GOPs to be processed, and calculates an address by the GOP address operator 154 to control the rotation speed of the optical head actuator or the disk.

In this way, if the address of the GOP to be accessed can be obtained even during the playback of the scaffold, and the desired GOP can be accessed by scanning the disk, the following address is calculated in the same manner, and the configuration described in Example 15 is used. The data in the low frequency region obtained by reproducing is reproduced and screened.

Mode signals, such as whether skip search or normal continuous playback are input from the microcomputer or the like, are input to the mode switch 130. The bit stream of the video is extracted and input to the data rearranger 131. The output of the mode converter 130 is supplied to the data rearranger 131 and the debottle determiner 132. The data rearranger 131 obtains this control signal and outputs the control signal to the variable length decoder 133 using only the L component without connecting the H component or the H component of FIG. .

The variable length decoder 133, together with the decodable determiner 132, extracts the end (boundary) of the event of the L component region, decodes up to the end, and outputs it to the switch 134. The switch 134 is controlled by the output of the decodable determiner 132 so that a zero is inserted into the high frequency side of the block in the low frequency component that has been successfully decoded to form a DCT block, and inversely CTCTs the data to perform an I picture. The output of the adder 138 is passed through and added, and in the case of the P picture, the image is corrected and added by the motion vector in the I picture of the reference, and in the case of the B picture, the motion vector is corrected and added in the I picture and the P picture. Reads of the memory 137 are controlled and added by the adder 138.

The DCT mode or prediction mode motion vector at this time is controlled by decoding the code of the header. The data thus predicted for motion compensation is decoded and accumulated in the image memory 137, and the image is in the original configuration order of the GOP. The inverse scanning converter 139 buffers and converts the image from the block scan to the raster scan in the order of output of the images. In addition, the switch 134 is not connected to insert 0 during normal playback, but is connected to play only the playback data.

Next, the operation of FIG. 80 will be described. FIG. 80 is a diagram showing a signal processing block in the case of performing division by the resolution of the digital signal reproducing unit, and specifically, an embodiment of the reproducing process in the case where the rate information is recorded by gathering at several places on the disc. This is a block diagram. Initially, the rotation of the optical head and the optical disc is controlled to lead directly or indirectly from the TOC area or an area corresponding thereto to designate the rate information description address, and amplify the reproduction signal from the optical head by the reproduction amplifier 156. The signal is detected by the digital demodulator 157, discriminated as a digital signal, and digital demodulated.

As a result, the reproduction signal made of digital data is input to the error corrector 158, and the error contained in the reproduction signal is corrected. The error free data is separated into an audio bit stream and a video bit stream by the system layer processor 159, and other data is also processed. For example, the stream path is discriminated by determining whether the signal is binary data such as AV data, text data, or a program. Among them, the rate information as described above is stored in the rate information memory 160.

On the other hand, the GOP number counter 161 produces | generates the information which wants to process several GOPs, calculates an address with the GOP address calculator 154, and controls the rotation speed of an optical head actuator and a disk. In this way, the rotational speeds of the actuator and the disk are controlled. In this way, if the address of the GOP to be accessed is also obtained during skip playback, the skip on the disc, and the desired GOP can be accessed, the following address is calculated in the same manner and the configuration described in Example 15, for example. This is to reproduce and display the data in the low frequency region obtained by using.

Mode signals such as whether skip search or normal continuous playback are input from the microcomputer or the like are input to the mode switcher 130. The bit stream of the video is extracted and input to the multiplexer 142. The multiplexer 142 sends the data of the low resolution component to the second variable length decoder 145 and the data of the low resolution component to the first variable length decoder 144 via the switch 143. . The switch 143 is controlled by the mode switch 130, and the mode is turned OFF when the resolution residual difference component is reproduced in the middle, even though only the low resolution component output is required for the skip search or the like. To work. In addition, the switch 143 is turned on in the case of reproducing with good signal transmission quality as in normal reproducing.

The second variable length decoder 145 decodes the Huffman code and the run length code, is dequantized by the second inverse quantizer 147, and is inversed by the inverse DCT circuit 136 from the frequency domain to the spatial domain. Is converted. If the data is an I picture, the adder 149 passes and accumulates in the picture memory. If the P picture is a P picture, the P picture is referred to from the picture memory, and the motion vector is corrected and read. The decoder 149 decodes the motion compensation prediction. do. In the case of the B picture, the same operation is performed for the I picture and the P picture.

In the figure, the motion vector, the quantization parameter for inverse quantization, the prediction mode, and the like are output from the variable length decoder, but the information flow is the same as that in FIG. The lower loop of the figure is decoding of the low resolution component, but the decoding result is input to the image memory 150 to interpolate the pixels by the resolution inverse transformer 152, and to supplement the decoding result as the residual resolution component. .

Example 20

Next, Example 20 of the present invention will be described with reference to FIGS. 81, 82, and 83. FIG. FIG. 81 is a block diagram for encoding processing, and FIG. 82 is a block diagram for decoding processing. 81 and 82, reference numeral 162 denotes a video signal encoder as encoding means, 163 denotes an audio signal encoder, 164, 167 denotes a memory, 165, and 168 denotes a memory controller. The data replenishing means is constituted by the memory 164 and the memory controller 165.

The video signal encoder 162 and the memory controller 165 constitute a code amount comparison means. Denoted at 166 is a system layer bit stream generator, 169 is a variable length decoder, 170 is a decoded signal processor after the variable length decoder, and a variable length decoder 169 and a decoded signal processor 170 are data. It functions as a decoding means, and the data rearranger 131 in FIG. 82 functions as data reconstruction means.

First, the operation of the configuration shown in FIG. 81 will be described. The memory layer 164 is disposed between the video signal encoder 162 and the system layer bit stream generator 166, and the system layer bit stream generator 166 is executed after the data embedding operation is performed between each GOP of the encoded video signal. While inputting each GOP to the audio signal, the audio signal is encoded by the audio signal encoder 163 and then input to the system layer bit stream generator 166 together with the video signal, and an operation such as adding a header or the like is performed.

Here, the embedding operation of data in the memory 164 will be described below. The memory controller 165 functions as a control circuit of the memory 164, and executes control to refill the encoded video signal data without spaces between the respective GOPs. Hereinafter, signal processing will be described with reference to FIG. 83. FIG. FIG. 83 is an explanatory diagram showing the concept of a process of a digital signal recording / reproducing apparatus. For example, when nGOP generates redundant data for an access position of an optical head and a control unit of error control (n + 1), FIG. In the case where a blank space occurs in the data area as an intermediate termination method for the access position of the optical head or the control unit of error control, as shown in FIG. 83A, the nGOP data over the blank portion after the (n + 1) GOP as shown in FIG. 83A. And embeds the remaining data of (n + 1) GOP and the data of (n + 2) GOP in the empty portion of (n + 3) GOP, which is very small to fill nGOP. Direction is controlled as shown in FIG.

As another control method, the remaining data is not conveyed as described above, and as shown in FIG. 83B, the (n + 2) GOP slightly exceeds the optical head's access position and error control unit ( n + 3) When the GOP is an intermediate termination method for the access position of the optical head and the control unit of error control, the data of (n + 3) GOP overfilling the blank after the (n + 2) GOP is embedded. Similarly, since the (n + 3) GOP is embedded, the remaining data of the (n + 2) GOP, which is only very small, is embedded in the blank portion of the (n + 1) GOP, and the data of the (n + 1) GOP Control is performed such that the data which has not been completely filled is embedded in the blank portion of the nGOP (the embedding direction is from left to right in the drawing).

Next, the operation of the configuration shown in FIG. 82 will be described. The memory 167 is controlled by the memory controller 168 to restore the data rearranged by the rules as described with respect to the above-described 83rd and 83rd degrees. For example, in the case of restoring data as shown in FIG. 83a, the portion of the nGOP continued after the (n + 1) GOP is concatenated after the nGOP data on the left side of the drawing, and then the (n + 1) GOP data is concatenated. It connects and operates to return to the state of the original GOP data, such as concatenating data of (n + 1) GOP which is continued after data of (n + 2) GOP.

In addition, this rearrangement rule needs to be determined in advance as a formatting rule of the medium, for example, recorded as flag information in an area collected after the TOC area, and if it is not determined, it must be specified somewhere on the medium. do.

Example 21

Next, Example 21 of the present invention will be described with reference to FIGS. 84, 85, and 86. FIG. FIG. 84 is a block diagram showing a signal processing unit in the case of performing division by frequency or quantization in the digital signal reproducing unit, and FIG. 85 is performing division or division by quantization in the bit length of the digital signal reproducing unit. A block diagram showing a signal processing unit in the case. FIG. 86 is a block diagram showing a signal processing unit in the case of performing division by resolution or quantization by the digital signal reproducing unit, where 171 is an IP selection indicator and 172 is a decodable plate. Periodically, 173 is a switch. Moreover, in FIG. 86, the said part is shown as an example as a 1st decoding means, 2nd, 3rd decoding means. In addition, the same code | symbol is attached | subjected to the same or equivalent part in drawing, and description is abbreviate | omitted.

Next, the operation will be described. 84 and 85, first, the rotation of the optical head and the optical disc is controlled so as to lead directly or indirectly (designate the rate information description address) from the TOC area or the area corresponding thereto at the beginning of disc playback. The reproduction signal is amplified by the reproduction amplifier 156, the digital demodulator 157 detects this signal, and discriminated as a digital signal to perform digital demodulation. As a result, the reproduction signal made of digital data is input to the error corrector 158, and the error contained in the reproduction signal is corrected. The error free data is processed by the system layer processor 159 into an audio bit stream, a video bit stream, or other data. 84, 85 and 86 output control signals to the IP selection indicator 171 from the mode switch 130 as the mode switching means for switching the decoding means in accordance with the special order reproduction speed. The control signal and the skip search speed are controlled to switch between displaying only the I picture or displaying the I picture and the P picture.

In the case of a speed such as 1000 times the skip search speed, if both the I picture and the P picture are displayed on the screen, the GOP must be removed and outputted very unnaturally. In this case, in order to eliminate the unnaturalness, it is necessary to switch to the reproduction mode of only the I picture, and the decoding unit 132 ((172) in Fig. 86) is instructed to decode not only the B picture but also the P picture. The image memory 137 (150 and 151 in FIG. 86) is controlled to display only the I picture at the same time as the stop (switch 173 plays this role in FIG. 86).

Normally, up to 5 times the screen display of the I picture and the P picture is good, but if it is 15 times or more, the screen display of only the I picture is better. This is because when I and P pictures are displayed at 15 times the speed, the continuity of movement is extremely reduced for the next 5 GOP destination even if the frame screen is updated. When the number of frames in the GOP: N = 15, the period of the I picture, and the P picture: N = 3, for example, at 7x speed playback, etc., decoding of the P picture is performed. When outputted in every P picture (3rd and 9th in a GOP), a much finer skip search can be performed.

As described above, when recording is performed by dividing the frequency domain at a predetermined position of each GOP data read on the recording medium, when recording is performed by dividing by the resolution, when the recording is performed by dividing by the quantization level, etc. When the data state is divided according to a predetermined condition and recording is performed, when the first data, which is the basic data in the reproduction data, and the second data except the basic first data are reproduced from the data in which data is arranged intensively. In the case of performing decoding on all of the first and second data, when the first data is decoded (at least) to obtain a reproduced image corresponding to the low-frequency region of the I picture or the number of pixels squeezed out, the first Decoding the data to obtain at least an I picture, a low frequency region of the P picture, or a reproduced image corresponding to the number of pixels to be subtracted. A has a decoding means for obtaining a reproduced image, it may be switched according to the special reproduction rate whether to use either the decoding means in the special reproduction.

It is a matter of course that the display method setting of the I picture and the P picture may be changed during playback in the forward direction and in the reverse direction. Since the decoding of the P picture can only be performed in the positive direction of time, in the reverse playback, it is necessary to accumulate a picture existing before the P picture to be decoded, and to use that much memory.

Therefore, in order to facilitate the use of this memory without using extra memory, for example, the I skip and P picture may be played in the forward skip search, and the reverse skip search may be played in such a manner that only the I pictures are played.

Example 22

A twenty-second embodiment of the present invention will be described with reference to FIGS. 87, 88, 89, and 90. FIG. FIG. 87 is a block diagram showing a signal processing unit in the case of performing division by frequency or quantization in the digital signal reproduction unit, and FIG. 88 is a signal processing block in the case of performing division by bit length in the digital signal reproduction unit. Figure is a diagram. 89 is an explanatory diagram showing the concept of processing in skip search. In FIG. 87 and FIG. 88, reference numeral 174 denotes a field display controller.

The system layer processor 159 also functions as image data extraction means. 87 and 88, corresponding portions are shown as an example of the video data decoding / reproducing means. Reference numeral 130 denotes a mode changer as mode change means. In addition, the same code | symbol is attached | subjected to the same or equivalent part as the above-mentioned example in drawing, and description is abbreviate | omitted.

Next, the operation will be described. 87 and 88, first, the rotation of the optical head and the optical disc is controlled so as to read directly or indirectly from the TOC area or the area corresponding thereto at the beginning of the disc reproduction (direct rate description address designation), and from the optical head. The reproduction signal is amplified by the reproduction amplifier 156, the digital demodulator 157 detects this signal, and discriminated as a digital signal to perform digital demodulation. As a result, the reproduction signal made of digital data is input to the error corrector 158, and the error contained in the reproduction signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 159, and other data is also processed.

In the case of skip search, for example, if the I picture and the P picture are outputted in succession, the screen output order is the same as the arrow of FIG. At this time, the interval between the four fields is empty in the encoded data even though the skip search time is continued between the even field of the I picture and the odd field of the P picture. In other words, the playback speed is five times higher than that between the odd and even fields of the I picture. Therefore, an image of an unnatural movement direction, which travels between one and five times the reproduction speed for each field, is obtained.

The best field of I picture is replaced with the same screen as the radix field, or the interlace is supplemented with the average of the upper and lower scanning lines. The screen is created and outputted. The image memory 137 in FIG. Is controlled using the field display controller 174. FIG. In this case, since the skip amount between the fields, which are the intervals of the encoded fields, is obtained almost evenly when each field of the reproduced image is recorded, the jerks (unnatural movement) become inconspicuous.

FIG. 90 is an explanatory diagram showing the concept of the process at the time of reverse playback, and particularly the field order of reverse playback. The field order of reverse reproduction will now be described based on FIG. When reverse playback is performed, reverse playback is normally performed in units of frames in which the odd field and the even field are paired. Specifically, when moving from the radix field to the even field, playback is performed in the same direction as the time on the image (reproduction operation for tracking the progress of a in FIG. 90A), and when moving from the even field to the radix field. Two fields are sent in the opposite direction to the image phase (the skip operation of tracking the progress of b in FIG. 90A).

However, if the reproduction path as described above is taken, the image is sent at the reverse speed of 3 times, resulting in a visually poor movement in which the movement sequence cannot be felt purely. As shown in FIG. 90B, the screen is tilted one by one in units of fields. When the image memory 137 is controlled by the field display controller 174 in Figs. 87 and 88 to display in the reverse order in the order of, since the skip amount between the fields is almost evenly obtained, jerkyness is not noticeable. Will not. However, for the synchronization signal at this time, it is necessary not to reverse the field synchronization signal but to maintain a normal relation of the slope field.

In addition, in order to adopt an independent display method for each field in this manner, the image memory 137 is independently taken out of the image memory 137 without taking the output of the adder 138 as it is. In addition, in order to perform such an operation, a buffer may be separately provided and sequence switching may be performed, or a three-port memory in which address control can be set independently may be used. The above operation can be realized even by multiplexing and reading into a memory having a very high operation speed. In addition, since the inverse scanning converter 139 has at least one field plus one slice of memory, the inverse scanning converter 139 may of course have such a buffering function.

Also, for slow playback during special playback, low-keyness is not noticeable because the same frame is continuously output several times. Therefore, the frames are reconstructed and output so that the playback time is equally spaced. For example, in the case of slow playback at 1 / 3x speed, for example, one decoded frame is output three times, and then the decoded B frame is not output three times, and the first one frame is an odd field of one frame. Compose one frame. The storm field side may take the average of the upper and lower lines.

In this way, in the interlaced image, there is no image shift in the screen vertical direction due to line shift, resulting in a stable image. The next one frame outputs the original one frame, and the next one frame is the even field of one frame. In the radix field, one frame is constructed (the radix field side may take an average of upper and lower lines), and the next 1 frame is composed of 1 frame in the radix field of a B frame (the superior field side may take an average of upper and lower lines). Then, the next one frame outputs the original B frame, and the next one frame constitutes one frame from the even field of the B frame (the radix field side may take an upper and lower average). Can slow the interval.

As mentioned above, although this invention was divided into several Example and described, this invention is not limited to the said Example, Of course, various changes are possible in the range which does not deviate from the summary.

Claims (46)

A digital video signal recording and reproducing apparatus for recording and reproducing a digital video scene holol recording medium encoded by using motion compensation prediction and orthogonal transformation, wherein at least one frame of image data is recorded for an I picture that performs intra-frame encoding at the time of recording. means for dividing by n, and n-divided I-picture frames preferentially record the area on the screen centered in each area unit, and at the same time, indicate a recording position on the recording medium in each of 1 to n divided areas. Means for simultaneously recording information, means for reading only the area in the center of the I picture at the time of special reproduction from the recording medium, a buffer memory for storing data of the read area, and means for outputting only the data in the center area read. Digital video signal recording and reproducing apparatus. The digital video signal recording arch according to claim 1, wherein special reproduction is executed by expanding and outputting data of the center area read during special reproduction to one screen. A digital video signal reproducing apparatus for reproducing a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein at least one frame is divided into n by one frame for an I picture that performs intra-frame encoding (n1). For an I picture frame divided by n, the area in the middle of the screen is preferentially recorded in each area unit, and at the same time, the position information indicating the recording position in the recording medium of each of the divided areas 1 to n is also recorded. A digital video signal reproducing apparatus comprising means for reading only the area in the center of an I picture from a recording medium during special reproduction on a medium, a buffer memory for storing data of the read area, and means for outputting only data in the center area of the read area. . 4. The digital video signal reproducing apparatus according to claim 3, wherein special reproduction is executed by expanding and outputting data of the center area read during special reproduction to one screen. A digital video signal recording and reproducing apparatus for recording and reproducing a digital video signal encoded using motion compensation prediction and an orthogonal transformation onto a recording medium, wherein at least one frame of image data is recorded for an I picture which performs intra-frame encoding at the time of recording. means for n-dividing (n1), for an n-divided I picture frame, preferentially recording the area on the screen centered in each area unit, and at the same time, indicating a recording position on the recording medium of each divided area of 1 to n. Means for simultaneously recording information, means for reading at least I pictures from the recording medium during special playback, a buffer memory for storing the data of the read I pictures, and an area that could not be read if all the I picture areas could not be read. A digital video signal recording and reproducing apparatus comprising interpolation means for interpolating data before one screen. A digital video signal recording and reproducing apparatus for recording and reproducing a digital video signal encoded using motion compensation prediction and an orthogonal transformation onto a recording medium, wherein at least one frame of image data is recorded for an I picture which performs intra-frame encoding at the time of recording. means for n-dividing (n1), for an n-divided I picture frame, preferentially recording the area on the screen centered in each area unit, and at the same time, indicating a recording position on the recording medium of each divided area of 1 to n. Means for simultaneously recording information, means for reading at least I pictures from the recording medium during special reproduction, areas 1, 2,... means for outputting a special playback image for one screen in accordance with the data of the n area, and interpolation means for interpolating an area that could not be read if all the I picture areas could not be read from data before one screen. Digital video signal recording and reproducing apparatus. A digital video signal recording and reproducing apparatus for recording and reproducing a digital video scene holol recording medium encoded by using motion compensation prediction and orthogonal transformation, wherein at least one frame of image data is recorded for an I picture that performs intra-frame encoding at the time of recording. When recording the n-divided (n1) unit and the n-divided I picture in area units, the sequence number of the area where recording is started is scrolled and recorded in the GOP unit of motion compensation prediction, and the value of 1 to n value in the GOP is recorded. Means for simultaneously recording position information indicating a recording position on the medium, means for reading at least I pictures from the recording medium during special reproduction, a buffer memory for storing data of the read I pictures, and special data according to the data of the read I pictures. Data of one screen before the means for outputting the reproduced image and the area that could not be read if all the I picture areas could not be read. Di jitreo standing interpolation video signal recording and reproducing apparatus comprising: means for interpolation. I-frames that perform intra-frame encoding, P-pictures that perform forward motion compensation prediction, I-pictures that are located before and after temporally, and B-pictures that perform motion-compensation prediction by using P-pictures as reference pictures A digital video signal recording and reproducing apparatus for recording and reproducing a digital video signal encoded by using a recording frame in units of several frames, wherein at least I-frame and at least one-frame video data are divided (n1) for an I picture and a P picture during recording. Means for encoding in each n-divided area unit, I-picture and P-picture frame divided by n, priority is given to I-picture, and an area of 1 to n divided by recording the area that comes to the center on the screen first in area units. Means for simultaneously recording position information indicating a recording position on the recording medium of the area. At the time of special playback, at least I pictures and P pictures are read from the recording medium. However, the buffer memory for storing the data of the read I picture and the P picture, the means for outputting the data of the read I picture and the P picture as a special playback picture, frame by frame, and all I pictures or the P picture area can be read. A digital video signal recording and reproducing apparatus comprising interpolation means for interpolating an area that could not be read if there was no data based on data before one screen. A number of frames in which an I picture that performs intra-frame encoding, a P picture that performs motion compensation prediction in the forward direction, and an I picture that is located before and after temporally and a B picture that performs motion compensation prediction by using the P picture as a reference picture are mixed. A digital video signal recording and reproducing apparatus for recording and reproducing a video signal encoded in units on a recording medium in units of several frames, wherein at least I-picture and P-picture are divided by n-frame video data (n1). In this case, for each n-divided area coding unit, n-divided I pictures and P-picture frames, I pictures are given priority, and areas that come to the center on the screen are preferentially recorded in area units and divided into 1 to n. Means for simultaneously recording position information indicating a recording position on the recording medium of each area of the recording medium, and continuous n when decoding a reproduced image for one screen during special playback. Area 1, 2,... In each of I and P pictures. means for sequentially reading the data of the n area, means for outputting a special playback image for one screen according to the data of the I picture and the P picture, and an area for which all the I pictures or areas of the P picture cannot be read. A digital video signal recording and reproducing apparatus comprising interpolation means for interpolating data before one screen. A number of frames in which an I picture that performs intra-frame encoding, a P picture that performs motion compensation prediction in the forward direction, and an I picture that is located before and after temporally and a B picture that performs motion compensation prediction by using the P picture as a reference picture are mixed. A digital video signal recording and reproducing apparatus for recording and reproducing a digital video signal encoded by a unit on a recording medium in units of several frames, wherein at least I-picture and P-picture are divided by n by at least one frame. means for encoding in each n-divided area unit, and for the n-divided I picture and P picture frame, an I picture is given priority, and when recording n-divided I pictures and P pictures in area units, In addition to scrolling and recording the position in units of I and P picture frames, it is also possible to simultaneously record position information indicating the recording position on the recording medium of each of the areas 1 to n in the GOP. Means for reading at least an I picture or a P picture from a recording medium, a buffer memory for storing data of a read I picture or a P picture, and data of a read I picture or P picture as a special playback picture in a special playback picture. And interpolation means for interpolating an area that could not be read if the area of all I pictures or P pictures could not be read from the data before one screen. A digital video signal reproducing apparatus which encodes a digital video signal by using motion compensation prediction and orthogonal transformation and reads and reproduces data recorded on a recording medium, wherein at least an I picture for performing intra-frame encoding is performed on a frequency domain, a quantization level or Among the data divided by the spatial resolution and divided at least for an I picture, a bitstream of video data in which data more important as an image is placed at the head is constructed, and the address information of the divided data is used as header information as the header information. Data re-arrangement means for arranging the packet at the head of the bit stream, rearranging and outputting the data recorded on the recording medium in order of data before division according to the header information in the packet during normal reproduction, and during the special reproduction. Decode and output the data arranged at the head of the recording medium Digital video signal reproducing apparatus comprising special reproduction data output means for executing special reproduction by the. A digital video signal reproduction method for reproducing a digital video signal recorded by coding using motion compensation prediction and orthogonal transformation, wherein at least I pictures for performing intra-frame coding are divided by frequency domain, quantization level, or spatial resolution, and at least I Of the data divided for a picture, a bitstream of video data in which data more important as an image is placed at the head is constituted, and when data recorded on a recording medium is normally reproduced, data is divided before data according to the header information in the packet. And rearranging and outputting the data, and performing special playback by decoding and outputting data arranged at the head of the recording medium during special playback. A digital video signal recording / reproducing apparatus for recording a digital video signal encoded by using motion compensation prediction and an orthogonal transformation on a recording medium and reproducing data on the recording medium, wherein at least an I picture for performing intra-frame encoding is frequency domain and quantized. Means for dividing by level or spatial resolution, means for constructing a bitstream of video data in which data more important as an image is placed first among data divided for at least I pictures, and address information of the divided data as header information. Means for constructing a packet at the head of the bit stream of the video data; means for recording the configured data on the recording medium; in normal playback, data is rearranged in order of data before division according to header information in the packet. Data rearrangement means to be output, and to the top when special playback Digital video signal recording and reproducing apparatus comprising special reproduction data output means for executing a special playback by decoding and outputting the data. A digital video signal recording and reproducing method for reproducing a digital video signal recorded by encoding using motion compensation prediction and orthogonal transform, comprising: dividing at least by an I-picture corresponding frequency domain, quantization level, or spatial resolution to perform intra-frame coding; Constituting a bitstream of the video data in which data which is more important as an image is placed at the beginning of the data divided for at least the I picture, and address information of the divided data as the header information at the beginning of the boot stream of the video data. Arrange the packets by arranging them, and record them on the recording medium. During normal playback, the data is rearranged and outputted in order of data before division according to the header information in the packet. During special playback, the data arranged at the head is decoded and output. Digital containing steps for performing special playback by Video signal recording and playback method A digital video signal recording / reproducing apparatus for recording a digital video signal encoded using motion compensation prediction and an orthogonal transformation on a recording medium, and reproducing data on the recording medium, wherein at least I pictures for performing intra-frame encoding at the time of recording are recorded. means for dividing the n-divided I-picture data into area units to form n-bit (n1) to form a bitstream of video data having an area that comes to the center on the screen first; header information of the address information of the divided area Means for forming a packet at the head of the bit stream of the video data and recording the packet on the recording medium. In normal playback, the I picture data is rearranged in area units according to the header information in which the packet is placed at the head. Data re-arrangement means to be outputted and outputted at the time of special reproduction can be read from the beginning of the packet within a certain time Digital video signal recording and reproducing apparatus comprising special reproduction data output means for executing a special playback mode by only the I picture data output. A digital video signal recording and reproducing method for recording a digital video signal encoded by using motion compensation prediction and an orthogonal transformation on a recording medium, and reproducing data on the recording medium, comprising: at least I pictures for executing at least intra-frame encoding at the time of recording Is divided by n, and the n divided I picture data are rearranged in area units to form a bitstream of video data in which the area that comes to the center on the screen is first, and the address information of the divided area is header. In the step of forming a packet at the head of the bit stream of the video data as the information and recording it on the recording medium, and during normal playback, the data of the I picture is reconstructed in area units according to the header information arranged at the head of the packet. I pictures that have been arranged and output and can be read from the beginning of the packet within a certain time during special playback Only the recording and reproducing method comprising the digital image signal and executing special reproduction by the output data. A digital video signal reproducing apparatus which encodes a digital video signal by using motion compensation prediction and orthogonal transformation and reads and reproduces data recorded on a recording medium, by dividing at least n pictures (n1) of I pictures that perform intra-frame coding. Rearrange the n-divided I picture data in area units to form a bitstream of video data having an area that is centered on the screen first, and use the address information of the divided area as header information to determine the bitstream of the video data. In the case of normal reproduction of the data recorded on the recording medium by arranging the packet at the head, rearranged and outputted data of the I pictures rearranged by area according to the header information arranged at the head of the packet. In the case of data rearrangement means and special reproduction, a lead can be read within a predetermined time from the head of the packet. And a special reproduction data output means for executing special reproduction by outputting only data that has been stored. A digital video signal reproducing method of encoding and reproducing digital video signals using motion compensation prediction and orthogonal transformation to read and record data recorded on a recording medium, by dividing at least n pictures (n1) of I pictures for performing intra-frame encoding. Rearrange the n-divided I picture data in area units to form a bitstream of video data having an area that is centered on the screen at the head, and use the address information of the divided area as header information at the head of the bitstream of the video data. To reconstruct and output data of the I picture rearranged by area according to the header information arranged at the head of the packet when the data recorded on the recording medium is normally reproduced. During the step and special playback, it was possible to read from the head of the packet within a certain time. Only the digital video signal reproducing method comprising the special reproduction data output means for executing a special playback by the output data. A digital video signal recording and reproducing apparatus which encodes a digital video signal by using motion compensation prediction and orthogonal transformation to read and record data recorded on a recording medium. And a bitstream of video data in which the basic data of the divided I-picture data is rearranged in units of areas on the screen, at least by means of dividing by high-frequency region, quantization level, or spatial resolution. Means for constructing a packet by arranging the divided area, data division, and address information of the picture as header information at the head of the bitstream of the video data, and recording the packet on the recording medium. Area data according to header information arranged in the part Data reordering means for rearranging and outputting the data, means for rearranging in order of data before division, and special playback data for performing special playback by outputting only data that can be read within a predetermined time from the head of the packet during special playback. A digital video signal recording and reproducing apparatus comprising output means. A digital video signal recording and reproducing apparatus which encodes a digital video signal by using motion compensation prediction and orthogonal transformation to read and record data recorded on a recording medium. And high frequency range. A step of dividing by the quantization level or the spatial resolution, the divided area, the data division, and the address information of the picture as header information at the head of the bit stream of the video data to form a packet, and recording on the recording medium; In reproducing, the data is rearranged and output in area units according to the header information arranged at the head of the packet, and the divided data is rearranged in the original data order. And reproducing special playback by outputting only data that can be read in. A digital video signal reproducing apparatus which encodes a digital video signal by using motion compensation prediction and orthogonal transformation and reads and reproduces data recorded on a recording medium. High frequency range. A packet is formed by dividing the quantization level or spatial resolution by dividing the divided area, data division, and address information of the picture as header information at the head of the bitstream of the video data, and storing the data recorded on the recording medium. The data rearrangement means for rearranging and outputting data in area units according to the header information arranged at the head of the packet at the time of reproduction, the means for rearranging data in order before division, and at the beginning of the packet during special playback. A digital video signal reproducing apparatus comprising special reproduction data output means for performing special reproduction by outputting only data that can be read in time. A digital video signal reproducing apparatus which reads and reproduces data recorded on a recording medium by encoding a digital video signal by using motion compensation prediction and orthogonal transformation. At least an I picture for performing intra-frame encoding includes a low frequency region and a high frequency region. Areas of the divided I-picture data divided by quantization level or spatial resolution are rearranged in units of areas on the screen to form a bit stream of video data in which the area that comes to the center on the screen is placed first, and the divided area, A packet is formed by arranging data division and address information of a picture as header information at the head of the bit stream of the video data. According to the header information arranged at the head of the packet when data recorded on the recording medium is normally reproduced. Rearranging and outputting the data in area units, rearranging the divided data in the original data order, and special playback by outputting only data that can be read within a predetermined time from the beginning of the packet during special playback. And a step of executing the digital video signal. A digital video signal recording apparatus that encodes a digital video signal using motion compensation prediction and orthogonal transformation and reads and reproduces the data recorded on the recording medium. The digital video signal is encoded by using motion compensation prediction and orthogonal transformation. First encoding means for encoding a video signal comprising an encoded picture including at least an intra-frame coded picture among the signals, and encoding for residual residual components by encoding using the first encoding means among the video signals. And encoding means for arranging respective output data outputted from the first encoding means and the second encoding means in each image group data string for each image group data. The digital video signal recording apparatus according to claim 23, wherein said first encoding means encodes the video information subtracted from the video signal comprising at least a coded picture including an intra-frame coded picture. 24. The digital video signal recording apparatus according to claim 23, wherein the first encoding means encodes only a low frequency region orthogonally transformed. A digital video signal recording apparatus according to claim 23, wherein said first encoding means quantizes and coarsely coarser than a specific quantization level. A digital video signal recording apparatus for recording a digital video signal encoded using a motion compensation prediction and an orthogonal transformation onto a recording medium, the digital video signal recording device comprising at least an intra-frame coded image of the digital video signal encoded using the motion compensation prediction and an orthogonal transformation. And a low frequency region extracting means for extracting data of a low frequency region from each partitioned data string by dividing a video signal composed of an encoded image. A digital video signal reproduction apparatus for reproducing on a recording medium a digital video signal encoded by using motion compensation prediction and an orthogonal transformation, and data of a low frequency region and data of a high frequency region divided by regions, wherein the low frequency region Data rearrangement means for rearranging data and data in the high frequency region; and mode switching means for selecting one of a mode for decoding the rearranged data or a mode for selectively decoding the data in the low frequency region. Digital video signal playback device. 29. The method of claim 28, when decoding in the mode of decoding only the data of the low frequency region, decoding only the decodable data, discarding the undecodable data, and performing inverse orthogonal transformation by substituting a fixed value for the obtained resident wave region data. And a digital video signal reproducing apparatus. A digital video signal recording apparatus for recording a digital video signal encoded using a motion compensation prediction and an orthogonal transformation onto a recording medium, the encoding comprising at least an intra-frame coded image of the digital video signal encoded using the motion compensation prediction and an orthogonal transformation. Means for attaching a block terminating code to the coded data of each block of the video signal of an image as the data of the low frequency region and the encoded data exceeding the number of bits to which the block terminating code is added. A digital video signal recording apparatus comprising encoding means for performing encoding as data. A digital video signal reproducing apparatus which reads data of a low frequency region encoded by a motion compensation prediction and an orthogonal transformation and data of a high frequency region partitioned by a block terminating code from a recording medium. Data reconstruction means for reconstructing the data in accordance with each of the data and the block terminating code, a mode switching means for selecting any one of a mode for decoding jacked data or a mode for selectively decoding only data in a low frequency region, the mode And a data processing means for decoding the reconstructed coded data according to the output of the switching means, and data processing means for performing inverse orthogonal transformation by replacing the high frequency region with a fixed value. Records a digital video signal which has been encoded according to motion compensation prediction and orthogonal transformation, and has data of low resolution components from which pixels are extracted, and data of difference components between an image before the pixels are extracted and an image interpolated after the pixels are extracted. A digital video signal reproducing apparatus read from a medium, comprising: means for synthesizing the low resolution component data and the difference component data, and decoding means for decoding the synthesized data. 33. The digital video signal reproducing apparatus according to claim 32, further comprising mode switching means for switching a mode for synthesizing and decoding data of a low resolution component and data of a difference component and a mode for decoding only data of a low resolution component. 33. The apparatus of claim 32, further comprising interpolation image generating means for generating an interpolated image after decoding when decoding a low resolution image. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation is executed is determined. A judging means, adaptive encoding means for adaptively changing and encoding a data rate in accordance with the judging output from the judging means, and information adding means for adding an audio signal, additional information, and an error correcting code; And a digital video signal recording apparatus multiplying the additional information to record on a recording medium. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation is executed is determined. An information adding means for adding the determining means, the audio signal, the additional information and the error correcting code, the first encoding means for encoding the video signal which has been subtracted from the video signal comprising at least the coded image including the intra-frame coded image and A second encoding means for performing encoding by the residual difference component by encoding using the first encoding means in the video signal, wherein the first and second encoding means in accordance with a determination output from the determining means. Code by varying the data rate in at least one of the encoding means Digital video signal recording device. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation is executed is determined. An information adding means for adding an audio signal, additional information, and an error correcting code, first encoding means for encoding only a low frequency region orthogonally transformed to a video signal composed of at least a coded image including an intra-frame coded image, and A second encoding means for performing encoding on the residual difference component of the video signal by the encoding using the first encoding means, wherein the first and the second encoding means according to the determination output from the determination means. Adaptively vary the data rate in at least one of the encoding means Digital video signal recording apparatus for encoding. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation is executed is determined. A means for quantizing and encoding a video signal composed of a coded image including at least an intra-frame coded image at a level roughly quantized than a specific quantization level, for determining a means, an audio signal, additional information, and an error correcting code. First encoding means and second encoding means for performing encoding on the residual difference component of the video signal by encoding using the first encoding means, wherein the first encoding means comprises the first encoding means and the first encoding means. Data in at least one of the second encoding means. Digital video signal recording apparatus for adaptively changing the data. A digital video signal reproducing apparatus for reading data in which the data rate of encoded data encoded according to motion compensation prediction and orthogonal change is adaptively variable according to the drawing, which performs switching between the normal reproduction mode and the special reproduction mode. Calculating the position on the recording medium in which the data for special reproduction is present according to the mode switching means, the data rate information extracting means for extracting the data rate information, and the data rate information output from the data rate information extracting means in the special reproduction mode. A digital video signal reproducing apparatus comprising a position calculating means. A digital video signal reproducing apparatus according to claim 39, further comprising head position converting means for controlling the head position to a position on a recording medium in accordance with an output from said position calculating means and a special reproduction speed. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and an orthogonal transformation onto a recording medium, wherein the digital video signal recording apparatus is assigned to one group of images formed by the digital video signal encoded according to the motion compensation prediction and orthogonal transformation. Extra data is blanked in accordance with the encoding means for executing and controlling the code amount corresponding to the area, the code amount comparing means for comparing the output from the encoding means with a specific code amount, and the output from the sound code amount comparing means. A digital video signal recording apparatus comprising data supplementing means embedded in a blank area of an image group having an area. A digital video signal reproducing apparatus which reads from a recording medium a digital video signal data encoded according to a motion compensation prediction and an orthogonal transformation, and embedded with data of an image group corresponding to a blank area of an image group formed by recording the encoded data. And a data decoding means for reconstructing the embedded video signal coded data into an original picture group, and data decoding means for decoding the data reconstructed by the data reconstruction means. First and second corresponding to the first and second encoded data in data in which encoding is performed in accordance with the motion injury prediction and orthogonal transformation and an array of the first and second encoded data is formed in each image group data string. A digital video signal reproducing apparatus for generating decoded data, comprising: first decoding means for decoding the first and second coded data to obtain a reproduced image, and low frequency of an intra-frame coded image by decoding the first coded data. Second decoding means for obtaining a reproduced image corresponding to a quantization that is rougher than a region, a depressed hydrogen, or a specific quantization level, and a low frequency region of at least an intra-frame coded image and an inter-frame predictive coded image by decoding the first coded data. Decoding means for obtaining a reproduced image corresponding to the number of pixels squeezed out or the quantization coarser than a specific quantization level and during special reproduction. Digital video signal reproducing apparatus comprising a first, a second and a mode switching means for switching in accordance with the special playback speed whether to use any decoding means of the decoding means of the three. A digital video signal reproducing apparatus for reproducing video information encoded according to a motion compensation prediction and an orthogonal transformation on a recording medium, comprising: video data extraction means for extracting data corresponding to a video signal from a reproduction code, and output from the video data extraction means. Data decoding and reproducing means for decoding and reproducing the video data and the mode for reproducing and displaying any one of the normal playback mode, the odd field or even field at the time of reproducing the video data, or the field configuration of the odd or even field is reversed. A digital video signal reproducing apparatus comprising mode switching means for switching a mode for displaying playback. The encoding is performed in accordance with the motion compensation prediction and the orthogonal transformation, and the array of the first and second encoded data is formed in each image group data string. A digital video signal reproduction apparatus for generating first and second decoded data corresponding to first and second encoded data, comprising: first decoding means for decoding the first and second encoded data to obtain a reproduced image; Second decoding means for decoding the first coded data to obtain a reproduction image corresponding to a low frequency region of the intra-frame coded image, the number of pixels to be subtracted, or a quantization that is coarser than a specific quantization level, and the first coded data. To decode at least the low-frequency region of the inter-frame coded picture, the inter-frame predictive coded picture, the number of pixels to be subtracted, or the specified quantization level. A third decoding means for obtaining a reproduction image corresponding to proton quantization, and a mode switching means for switching which decoding means of the first, second and third decoding means to be used in the special reproduction according to the special reproduction speed. Digital video signal playback device. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the degree of image degradation when encoding and decoding according to motion compensation prediction and orthogonal transformation is executed is determined. A judging means, adaptive encoding means for adaptively changing and encoding a data rate in accordance with the judgment output from the judging means, and information adding means for adding an audio signal, additional information, and an error correction code, and including data rate information. Digital video signal recording device to write to the recording medium.