JPH09298711A - Device for generating compressed image data for editing for image information recording and reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the degradation of picture quality of B picture just before cut-in point or cut-out point in insert editing by recording bidirectional predictive encoded images again inside the prescribed number of unit encoded image groups just before and just after an editing point. SOLUTION: The image data of respective two frames just before and just after the cut-in point of designated insert editing are reproduced and supplied through an input terminal 2 to an MPEG decoder 14. The bidirectional predictive encoded images in the prescribed number of unit encoded image groups just before the editing point are encoded and recorded again by bidirectional prediction from the pictures of intra-frame encode images in the prescribed number of unit encoded image groups just before the editing point, and pictures in newly prepared image data to be recorded continuously to the editing point. Besides, after the editing point, newly prepared image data to be recorded continuously to the editing point are recorded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention uses a combination of one intraframe coded image and a predetermined number of bidirectional predictive coded images as a group of coded images, and the above group of coded images A compressed image for editing when image data is edited and recorded using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on a recording medium with digital image data corresponding to a predetermined number of It relates to a data generator.

[0002]

2. Description of the Related Art Digital image information has a huge amount of information, while digital image information is transmitted (recorded, reproduced).
Since the communication capacity (storage capacity of the recording medium) of the transmission path used for the transmission is limited, the image information is conventionally compressed when transmitting (recording and reproducing) the digital image information. As a standard for the image information compression, there are JPEG, MPEG1, 2, etc., for example. Among various conventional standards regarding the compression / expansion of the image information described above, the outline of MPEG2 will be described as a typical example. As is well known, MPEG
2 is a discrete cosine transform (DCT), motion compensation prediction,
It is configured by element technologies such as adaptive quantization and variable length coding. Motion-compensated prediction of MPEG2 is performed between frames or between fields.

FIG. 5 is a block diagram showing a configuration example of an MPEG encoder. The image input [video signal composed of a luminance signal (Y) and two color difference signals (Cr, Cb)] supplied to the input terminal 35 is converted into a digital signal by the analog-digital converter 36, and then the format conversion unit. 37. The format conversion unit 37 converts the spatial resolution of the digital image signal supplied thereto into the spatial resolution used for encoding, and then supplies the processed image to the screen rearrangement unit 16. The screen rearranging unit 16 rearranges the screens according to the picture types I, P, and B. Then, the input digital image signal is subjected to discrete cosine transform in the DCT unit 19. The DCT coding coefficient output from the DCT unit 19 is quantized by the quantizing unit 20 and then variable-length coded by the variable-length coding unit 21 together with the motion vector and coding mode information.
2 is stored. Then, the MPEG video bit stream is output from the buffer memory 22 to the output terminal 4.

Since the I picture and the P picture need to be used later as a reference screen for motion compensation prediction, the quantized information is the inverse quantization section 24, the inverse DCT section 25, the adder 32, and the image. Memory 33, motion compensation prediction unit 3
The local decoding operation is performed by the operations such as 4 and the same image as the decoder is restored and stored in the image memory 33. Encoding for each screen is done in macroblock units,
Encoding is performed in order from left to right and top to bottom on the screen. For each macroblock, it is determined whether the motion compensation prediction mode (inter coding mode) or the intra coding mode. In the case of the motion compensation prediction mode, the input macroblock image data and the motion prediction from the reference screen are performed. A prediction error signal is obtained by taking the difference from the obtained macroblock image data. In the switch 31, the switch 31a is turned on and the switch 31b is turned off in the intra coding mode, and the switch 31a is turned off and the switch 31b is turned on in the motion compensation prediction mode.

The prediction error signal is converted into a spatial frequency domain by division cosine transform in block units of 8 pixels × 8 lines, and in the case of intra coding in which motion compensation prediction is not performed, input image data is input. Is DCT encoded as it is. The 8 × 8 DCT coefficient after conversion is quantized according to the target bit rate and visual characteristics, and is sequentially scanned from the low frequency component and converted into one-dimensional information.
The macroblock coding information such as the coding mode and the motion vector and the quantized DCT coefficient are each coded by a variable length code. Therefore, the generated code amount becomes variable, and a mechanism for holding a constant bit rate is required when the fixed bit rate is used. In general, the amount of storage in the output buffer memory is monitored to perform quantization control that matches the target bit rate.

Next, FIG. 4 is a block diagram showing a configuration example of an MPEG decoder. The bit stream of the encoded data supplied to the input terminal 38 is stored in the buffer memory 39. The encoded data read from the buffer memory 39 has macroblock encoded information decoded by the variable length decoding unit 40, and the encoded mode, motion vector,
The quantized information and the quantized DCT coefficient are separated. The decoded 8 × 8 quantized DCT coefficient is restored to a DCT coefficient by the inverse quantization unit 41. Then, the DCT coefficient is converted into pixel space data by the inverse DCT unit 42 and supplied to the adder 43.

The output of the adder 43 is the image memory 4
4 (prediction memory) and the format conversion unit 46. The motion compensation prediction unit 45, to which the pixel space data stored in the image memory 44 is given, performs signal processing using information such as the prediction mode and motion vector supplied from the variable length decoding unit 40, and outputs the signal. Is supplied to the adder 43. Then, in the case of the intra coding mode, the digital data given from the adder 43 to the format conversion unit 46 is
The pixel space data itself output from the inverse DCT unit 42, and the digital data supplied from the adder 43 to the format conversion unit 46 in the motion compensation prediction mode is the motion-compensated block data. It has been added. After all the macroblocks in the screen are decoded, the screen is rearranged in the original input order, the screen output is performed, and the screen size is converted as necessary, and then the analog-to-digital converter 47 converts It is output to the output terminal 48 as an image output signal in a signal form.

[0008]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
In G, in order to realize trick-play reproduction such as fast-forwarding, intermediate reproduction, and reverse reproduction by using a disk, tape, or other storage medium, a GOP structure in which data of a plurality of pictures is collected is introduced. By adding a sequence header to the encoded bit stream, random access in units of GOPs is possible, and a GOP is composed of at least one I picture and an arbitrary number of P and B pictures. By the way, by using a recording / reproducing device such as a video tape recorder (VTR), the intra-frame coded image (I picture) and the forward predictive coded image (P
(Picture) and bidirectional predictive-coded images (B-pictures) and the like, even when recording / reproduction of image data in a state in which image information is compressed is performed in the conventional VTR. It is desirable to be able to easily perform editing work.

However, as described above, the image data adopting the GOP structure composed of at least one I picture and an arbitrary number of P and B pictures is simply recorded and reproduced on the recording medium. Then, the conventional VT
The same editing work as in R cannot be done easily. That is, in the conventional VTR, since the composite video signal of one vertical scanning period is recorded on one recording trace formed on the magnetic tape, the composite video signal of which the video signal of one vertical scanning period is a unit is recorded. Editing operations such as insert recording and assembling recording can be performed relatively easily, but in the case of the image data in the image information compressed state as described above, the data amount for one screen is not constant,
Further, in addition to the intra-frame coded image (I picture), a forward predictive coded image (P picture) and a bidirectional predictive coded image (B picture) are included in order to improve compression efficiency. It is obvious that the editing work similar to that of the conventional VTR cannot be performed.

In order to make it possible to easily perform the editing work even on the image data in the image information compressed state as described above, for example, all the image information to be recorded is intra-frame encoded. It is conceivable to use an image (I-picture), but when all the image information to be recorded is an intra-frame encoded image (I-picture) as described above, the compression rate is naturally low. Since the image information is recorded, when the recording medium having the same recording capacity is used, the recording time of the original image information is shortened. Therefore, one intra-frame coded image (I picture) and a predetermined number of bidirectional predictive coded images (B picture) are combined (for example, I, B or I, B,
B) Then, a predetermined fixed amount of coding is performed on a unit of an encoded image group, and a predetermined number (for example, 3) of recording media is recorded by digital image data corresponding to the unit of the encoded image group. A magnetic recording / reproducing apparatus was conceived, which enables recording / reproduction in a state in which the compression ratio of image information is increased, and also enables editing and special reproduction by forming a recording trace of (book). .

FIG. 3 shows a magnetic recording / reproducing apparatus described above, that is, a magnetic tape (recording medium) T wound around a part of the peripheral surface of a drum composed of an upper drum and a lower drum at a predetermined traveling speed. A so-called helical scan type magnetic recording / reproducing apparatus [Video Tape Recorder (VT
R), or a video cassette recorder (VCR)], the recording trace pattern formed on the magnetic tape according to the rotation locus of the rotating magnetic head. In FIG.
The arrow a indicates the traveling direction of the magnetic tape T, and the arrow b indicates the traveling direction of the rotary magnetic head. Also,
Ci in FIG. 3 indicates a cut-in point when insert editing is performed, and Co indicates a cut-out point. Then, in the sequential recording traces shown in FIG. 3, two frames of data, that is, an I picture frame and a B picture frame are recorded for every three consecutive recording traces.

Here, as described above, two consecutive frames of data, i.e., I-picture frame and B-picture frame, are recorded for every three consecutive recording marks in the sequential recording marks. Two VTRs were prepared and one of them was used as the VTR on the player side,
Considering a case in which the other one is used as a VTR on the recorder side and insert editing is performed in units of data of two frames of the I picture frame and the B picture frame described above, The image information in the recording mark group on the left side of the cut-in point Ci and the image information (image information to be inserted) in the recording mark group on the right side of the cut-in point Ci are separated from each other with the cut-in point indicated by Ci as a boundary. It is a thing.

However, the B picture in the image information data of the two frames immediately before the cut-in point Ci of the magnetic tape to be recorded by the VTR on the recorder side is cut as described above before the insert edit is performed. The backward prediction may also be performed from the I picture in the image information data of the two frames immediately after the in-point Ci. However, since the insert editing is performed at the above-mentioned cut-in point Ci, the above-mentioned cut is performed. 2 immediately after the in point Ci
Since the I picture data in the frame image information data has been lost, the B picture in the 2 frame image information data immediately before the cut-in point Ci of the magnetic tape to be recorded is An incorrect decoding result may be obtained during reproduction.

Further, the image information insert-recorded between the cut-in point Ci and the cut-out point Co is the image information data output from the VTR on the reproducing apparatus side in a state where the image is compressed. Is. Therefore, for the B picture in the image information data of the last two frames in the image information used for the insert recording, the image output from the VTR on the reproducing device side is in an image compressed state. When the information data is used as it is, if it is different from the B picture in the image information data of the next frame recorded on the magnetic tape, it is newly recorded immediately before the cut-out point Co. For the B picture in the image information data of the last two frames, the wrong decoding result may be obtained during reproduction. Therefore, a solution was sought to prevent the above problems from occurring.

[0015]

According to the present invention, a combination of one intraframe coded image and a predetermined number of bidirectional predictive coded images is set as a unit of coded image group, and the coded image of the unit is described. Edit recording of image data for insert recording or assembling recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on a recording medium with digital image data corresponding to a predetermined number of groups At this time, the digital image data of the predetermined number of units of the coded image group just before and just after the editing point on the recording medium are decoded, and both are decoded in the predetermined number of units of the coded image group immediately before the editing point. The predictive coded image is a picture of the intra-frame coded image in the coded image group of a predetermined number of units immediately before the editing point. , Is encoded and re-recorded by bidirectional prediction from the picture in the image data newly prepared to be recorded after the edit point, and is recorded continuously after the edit point. A new editing compressed image data generating device of an image information recording / reproducing device, which records newly prepared image data, and 1
A digital image corresponding to a predetermined number of the coded image group of the unit, where a combination of a single intra-frame coded image and a predetermined number of bidirectional predictive coded images is a unit of the coded image group. When editing and recording image data for performing insert recording or assembling recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on the recording medium by the data, Decode the digital image data of the encoded image group of a predetermined number of units respectively,
A bidirectional predictive coded image in a predetermined number of units of the coded image group immediately before the edit point is coded as a forward predictive coded image in the predetermined number of units of the coded image group immediately before the edit point. Generation of compressed image data for editing of the image information recording / reproducing apparatus, which is adapted to be recorded again after the above-mentioned edit point and is recorded after the above-mentioned edit point. The device and a combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images are set as a unit of the coded image group and correspond to the predetermined number of the unit of the coded image group. Edit recording of image data in which insert recording is performed by using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on a recording medium by digital image data being recorded. In this case, in the coded image group of the unit of the image data recorded on the recording medium of the recorder, which is located immediately after the cutout edit point set for the image data recorded on the recording medium of the recorder. Encoding of the last unit in the reproduction image data prepared for insert recording before the cutout edit point set for the image data recorded in the recording medium of the recorder described above in the frame The code of the last unit in the reproduced image data prepared for insert recording before the cutout edit point by bidirectional prediction by the intra-frame coded image of the coded image group of the image group Generate a bidirectional predictive coded image of the coded image group of the unit to be the coded image group, and insert it before the cutout edit point described above. Image information recording / reproduction in which insert recording is performed by using a bidirectional predictive coded image of a coded image group of a unit which is the last unit coded image group in the reproduced image data prepared for record recording A device for generating compressed image data for editing of the device, and a combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images as a group of coded images, and the coded image of the unit When editing and recording image data for insert recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on a recording medium with digital image data corresponding to a predetermined number of groups. Playback image data prepared for insert recording before the cut-out edit point set for the image data recorded on the recording medium of the machine The bidirectional prediction coded image of the unit coded image group that is the last unit coded image group in is generated with inter-frame prediction limited to forward prediction, Insert recording is performed before the out edit point by using the bidirectional prediction coded image of the coded image group of the unit that is the last unit coded image group in the reproduced image data prepared for insert recording. And a generator of compressed image data for editing of the image information recording / reproducing apparatus.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION The concrete contents of the editing compressed image data generating apparatus of the image information recording / reproducing apparatus of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 1 and FIG. 2 are block diagrams showing an example of the configuration of a generator of compressed image data for editing of an image information recording / reproducing apparatus of the present invention. In the apparatus for generating compressed image data for editing of the image information recording / reproducing apparatus shown in FIGS. 1 and 2, 1 is a dubbing terminal, and this dubbing terminal 1 has one intra-frame coded image (I
(Picture) and a predetermined number (one in the following example) of bidirectional predictive coded images (B pictures) are combined (I, B) as a group of coded images, and the coded image of the unit A predetermined number of recording marks (three recording marks in the following example) on the recording medium (magnetic tape) are formed by digital image data corresponding to a predetermined number of groups (one in the following example). A reproduction signal in the state of encoded data is supplied from the image information recording / reproducing apparatus, and 2 is an output signal (encoding) from a reproducing circuit (not shown) of the image information recording / reproducing apparatus of the present invention. (Reproduction signal in the state of data) is an input terminal, 3 is a video signal input terminal, and the video signal input terminal 3 is supplied with a digital image signal which is not image information compressed.

First, the generator of compressed image data for editing of the image information recording / reproducing apparatus of the present invention shown in FIG. 1 and FIG.
When the signal processing is performed for the digital image signal in the state where the image information supplied to the video signal input terminal 3 is not compressed, the movable contact v of the changeover switch 7 is fixed. While being switched to the a side, the movable contact v of the changeover switch 11 is switched to the fixed contact b side. Then, video signal input terminal 3
The digital image signal that has not been subjected to image information compression is supplied to the screen rearranging unit 16 via the fixed contact a and the movable contact v of the changeover switch 7.

As described above, the generator of the compressed image data for editing of the image information recording / reproducing apparatus of the present invention shown in FIGS. 1 and 2 has one intra-frame coded image (I picture) and a predetermined number of frames. The image information is compressed by using a combination (I, B) as a unit of a combination (I, B) with a bidirectional predictive encoded image (B picture) in the following example. From the above, the screen rearranging unit 16 rearranges the frames coded as B pictures so as to obtain a result in which compression is performed with a structure such as I picture → B picture → I picture → B picture. Stored in the frame memory 17 in the section 16,
The next frame to be coded as an I picture is preceded by the subtractor 18 and the changeover switch via the fixed contact b and the movable contact v of the changeover switch 8 before the frame to be coded as the B picture. The screens are rearranged in accordance with the picture types I and B so that they are supplied to the DCT unit 19 via 9 and 9.

The input digital image signal is DC
In the T section 19, a block of 8 pixels × 8 lines (DCT
Block) units are transformed into the spatial frequency domain by the division cosine transform. The DCT coding coefficient output from the DCT unit 19 is quantized by the quantization unit 20,
It is provided to the variable length coding unit 21 and the inverse quantization unit 24.
The DCT coding coefficient quantized by the quantizing unit 20 is variable-length coded by the variable-length coding unit 21 together with the motion vector and the coding mode information, and the buffer memory 22.
Is accumulated in It is output from the buffer memory 22 as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v of the changeover switch 11 and is supplied to a recording circuit of a VTR (not shown). Further, the rate control unit 23, to which the information on the remaining storage capacity of the buffer memory 22 is supplied, sends a signal for controlling the quantization state in the quantizing unit 20 according to the information on the remaining storage capacity of the buffer memory 22. , To the quantizer 20 via the fixed contact b and the movable contact v of the changeover switch 10.

In the inverse quantizer 24, the quantizer 2 described above is used.
The quantized DCT coding coefficient supplied from 0 is dequantized and given to the inverse DCT unit 25. The calculation result in the inverse DCT unit 25 is so-called locally decoded image data, which is transmitted via the changeover switch 13 to the prediction memory 2
6, 27 are stored in a predetermined one. As mentioned above, I
After the coding of the picture is performed for two frames, the image data of the frame to be the B picture is read from the frame memory 17 of the screen rearranging unit 16 described above. In this state, the prediction memories 26 and 2 described above are
In FIG. 7, image information locally decoded for the frames before and after the frame to be the B picture is stored.

Then, for the image data of the frame to be the B picture read from the frame memory 17 of the screen rearranging unit 16 and the image data stored in the prediction memories 26 and 27, the motion compensation prediction unit 28 and 29 perform motion-compensated prediction, and forward prediction from the previous frame, backward prediction from the subsequent frame,
The averaging unit 3 calculates prediction results from both the forward prediction from the previous frame and the backward prediction from the subsequent frame.
Of the interpolative predictions obtained by averaging at 0, the most efficient prediction signal is selected by the change-over switch 12 and given to the subtractor 18 as a subtraction signal. The subtractor 1
In FIG. 8, the most efficient selected by the changeover switch 12 is selected from the image data of the frame to be the B picture read from the frame memory 17 of the screen rearranging unit 16 and supplied via the changeover switch 8. The prediction signal is subtracted and supplied to the fixed contact a of the changeover switch 9. The fixed contact b of the change-over switch 9 is supplied with the image data of the frame which is the B picture read from the frame memory 17 of the screen rearranging unit 16.

In the DCT unit 19, the amount of information is changed between the state where the movable contact v of the changeover switch 9 is switched to the fixed contact a side and the state where the movable contact v is switched to the fixed contact b side. Which has a smaller number of pixels is selected for each macroblock, the DCT operation is performed, and D obtained as the operation result is selected.
The CT coding coefficient is supplied to the quantizing unit 20 and quantized, and the DCT coding coefficient quantized by the quantizing unit 20 is changed by the variable length coding unit 21 together with the motion vector and the coding mode information. Long coded, buffer memory 22
Is accumulated in It is output from the buffer memory 22 as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v of the changeover switch 11 and is supplied to a recording circuit of a VTR (not shown). B-pictures are not used as reference frames for inter-frame prediction, so local decoding is not performed.

Next, the operation of the editing compressed image data generating apparatus of the image information recording / reproducing apparatus of the present invention shown in FIG. 1 when the insert edit recording is performed will be described. When insert edit recording is performed, first, image data of two frames immediately before the cut-in point Ci (Ci point in FIG. 3) and two frames immediately after the specified insert edit cut-in point (Ci in FIG. 3 (insert recorded portion)). , That is, the cut-in point Ci of the specified insert edit in FIG.
The image data in the three recording marks immediately before (Ci point in FIG. 3) and the image data in the three recording marks immediately after are reproduced. The reproduction signal is a reproduction circuit of the image information recording / reproducing apparatus of the present invention (not shown) (VT of the recorder side not shown).
The output signal (reproduction signal in the state of encoded data) from the R reproduction circuit) is supplied to the MPEG decoder 14 via the input terminal 2, the fixed contact b and the movable contact v of the changeover switch 6. As described above, the reproduction and decoding of the image data of the two frames immediately after the cut-in point is performed by converting the image information of the I picture in the above two frames into the B picture of the immediately preceding two frames. This is because it is used for decoding.

The quantized information used for the decoding operation in the MPEG decoder 14 is the quantized information memory 15
Is stored. Cut-in point Ci of the specified insert edit decoded by the MPEG decoder 14
The image data of the first one frame of the image data of the two frames immediately before (Ci point in FIG. 3) is the fixed contact b and the movable contact v of the changeover switch 7, and the changeover switch 8 of the screen rearranging unit 16. Via the fixed contact b and the movable contact v of the switch 9 and the fixed contact b and the movable contact v of the changeover switch 9
9 and supplied to the DCT unit 19 in the unit of blocks of 8 pixels × 8 lines (DCT block) and converted into the spatial frequency domain by the division cosine transform. DCT
The DCT coding coefficient output from the unit 19 is the quantization unit 2
Although it is quantized by 0, the quantization operation in the quantization unit 20 is performed by the quantization information read from the quantization information memory 15 via the movable contact v and the fixed contact a of the changeover switch 10, That is, it is performed using the quantized information used for the decoding operation in the MPEG decoder 14.

DCT quantized by the quantizer 20
The coding coefficient is provided to the variable length coding unit 21 and the inverse quantization unit 24. Then, D quantized by the quantizer 20
The CT coding coefficient is variable-length coded by the variable-length coding unit 21 together with the motion vector and coding mode information, and is accumulated in the buffer memory 22. In addition, the quantizer 2
The DCT coding coefficient quantized by 0 is the inverse quantization unit 24.
Is supplied to the dequantization section 24 and the dequantization section 24
The inversely quantized DCT coding coefficient supplied from the above is supplied to the inverse DCT unit 25. The inverse DCT unit 2
The calculation result in 5 is so-called locally decoded image data, which is stored in the prediction memory 26 via the movable contact v and the fixed contact a of the changeover switch 13.

The second one frame of the image data of the two frames immediately before the cut-in point Ci (Ci point in FIG. 3) of the specified insert edit decoded by the MPEG decoder 14 is also added. The image data of 1 is stored in the frame memory 17 of the screen rearranging unit 16 via the fixed contact b and the movable contact v of the changeover switch 7. In the state where the above preparation is performed, the changeover switch 6
Of the regenerator side by switching the movable contact v of the
It waits for the reproduction signal in the state of the encoded data output from the TR (the reproduction signal of the data to be insert-recorded) to be supplied to the dubbing terminal 1. As described above, the coded data supplied to the dubbing terminal 1 includes one intra-frame coded image (I picture) and one bidirectional predictive coded image (B picture). A combination (I, B) is used as a unit of encoded image group, and three recording marks on a recording medium (magnetic tape) are formed by digital image data corresponding to the unit of encoded image group. Is a reproduced signal in the state of encoded data.

The reproduction signal in the state of the encoded data supplied to the dubbing terminal 1 is given to the delay memory 5 and also the fixed contact a and the movable contact v of the changeover switch 6 are provided.
It is also supplied to the MPEG decoder 14 via The quantized information used for the decoding operation in the MPEG decoder 14 is stored in the quantized information memory 15. Of the two-frame image data decoded by the MPEG decoder 14, the first one-frame image data (I
(Frame data of a picture) includes fixed contact b and movable contact v of the changeover switch 7, fixed contact b and movable contact v of the changeover switch 8 of the screen rearranging unit 16, and changeover switch 9
Is supplied to the DCT unit 19 via the fixed contact b and the movable contact v of the above, and is converted into a spatial frequency domain by division cosine transform in units of blocks of 8 pixels × 8 lines (DCT block) in the DCT unit 19. .

DCT output from the DCT unit 19
The coding coefficient is quantized in the quantizing unit 20, and the quantizing operation in the quantizing unit 20 is performed by the quantizing information memory 1 via the movable contact v and the fixed contact a of the changeover switch 10.
5 is performed, that is, the quantized information used for the decoding operation in the MPEG decoder 14 is performed. The DCT coding coefficient quantized by the quantizer 20 is not variable-length coded, but is dequantized by the inverse quantizer 24 and then given to the inverse DCT unit 25. The calculation result in the inverse DCT unit 25 is so-called locally decoded image data, which is the changeover switch 13
Is stored in the prediction memory 27 via the movable contact v and the fixed contact b.

Then, the above-mentioned image data stored in the frame memory 17 of the screen rearranging unit 16, that is, the cut-in point Ci of the specified insert edit decoded by the MPEG decoder 14 (FIG. 3). 2 in the image data of the two frames immediately before the (Ci point in)
The image data of the 1st frame is read from the frame memory 17 and given to the motion compensation prediction units 28 and 29 via the fixed contact a and the movable contact v of the changeover switch 7. In the motion compensation prediction units 28 and 29 described above, the cut-in point Ci (C in FIG. 3) of the insert edit read from the frame memory 17 of the screen rearrangement unit 16 described above.
Prediction memories 26 and 2 for the image data of the second one frame of the image data of the two frames immediately before (i point)
Motion-compensated prediction is performed using the image data stored in 7.

The motion-compensated prediction based on the image data stored in the prediction memory 26, the motion-compensated prediction based on the image data stored in the prediction memory 27, and the image data from both of the above-described averaging units 30. The most efficient prediction signal is selected by the change-over switch 12 from among the interpolative predictions based on the data averaged by, and it is given to the subtractor 18 as a subtraction signal. In the subtractor 18 described above,
The most efficient prediction signal selected by the changeover switch 12 is selected from the image data of the frame which becomes the B picture read from the frame memory 17 of the screen rearranging unit 16 and supplied through the changeover switch 8. The value is subtracted and supplied to the fixed contact a of the changeover switch 9. The fixed contact b of the change-over switch 9 is supplied with the image data of the frame which is the B picture read from the frame memory 17 of the screen rearranging unit 16.

In the DCT unit 19, the amount of information is changed between the state where the movable contact v of the changeover switch 9 is switched to the fixed contact a side and the state where the movable contact v is switched to the fixed contact b side. Which has a smaller number of pixels is selected for each macroblock, the DCT operation is performed, and D obtained as the operation result is selected.
The CT coding coefficient is supplied to the quantizing unit 20 and quantized, and the DCT coding coefficient quantized by the quantizing unit 20 is changed by the variable length coding unit 21 together with the motion vector and the coding mode information. Long coded, buffer memory 22
Is accumulated in The image data already stored in the buffer memory 22, that is, the two frames immediately before the cut-in point Ci (Ci point in FIG. 3) of the designated insert edit decoded by the MPEG decoder 14 Of the image data of the first one frame, the DCT coding coefficient quantized by the quantizing unit 20 is the variable length code in the variable length coding unit 21 together with the motion vector and the coding mode information. Along with the encoded data stored in the buffer memory 22 after being converted, it is output as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v of the changeover switch 11 to record a VTR (not shown). Cut-in point Ci of the specified insert edit supplied to the circuit (Ci point in FIG. 3)
The image data of two frames immediately before is newly recorded for the three recording marks. Then, in the subsequent recording traces following the cut-in point Ci of the specified insert edit, the reproduction signal in the state of the encoded data supplied to the dubbing terminal 1 is predetermined by the delay memory 5. It is recorded with a time delay.

In the above description of the embodiments, the quantized information used in the decoding operation in the MPEG decoder 14 is stored in the quantized information memory 15, and the quantized information is used again in the quantized operation. Although the quantized information used for the decoding operation in the MPEG decoder 14 is used, the above method is not a necessary condition for implementing the present invention. However, as described above, the quantization information used for the decoding operation in the MPEG decoder 14 is stored in the quantization information memory 15, and the MPE described above is used when the quantization operation is performed again.
If the quantized information used for the decoding operation in the G decoder 14 is used, it is possible to prevent the deterioration of the image quality when the encoding / decoding operation is repeatedly performed for the frame. It is advantageous. Also, the frame immediately before that frame (frame of B picture) is
The frame may have been previously coded and predictively coded from the locally decoded image, in which case if the new quantization scale code is used, the decoded image will be If the locally decoded image at the time of encoding a frame (frame of a B picture) differs from the locally decoded image only by the difference in the quantization scale code, the image quality of the B picture will deteriorate. If
It is possible to prevent deterioration of image quality of B pictures. Further, in the description of the embodiments, the data input from the reproducing side is the encoded data, but the data input from the reproducing side may be the image information supplied to the input terminal 3, as a matter of course. is there. In that case, it is natural to output all data to the recording circuit after DCT, quantization and variable length coding.

Next, the operation of the generating apparatus for compressed image data for editing of the image information recording / reproducing apparatus of the present invention shown in FIG. 2 when insert edit recording is performed will be described. When insert edit recording is performed, first, image data of two frames immediately before the cut-in point Ci (Ci point in FIG. 3) and two frames immediately after the specified insert edit cut-in point (Ci in FIG. 3 (insert recorded portion)). , That is, the cut-in point Ci of the specified insert edit in FIG.
The image data in the three recording marks immediately before (Ci point in FIG. 3) and the image data in the three recording marks immediately after are reproduced. The reproduction signal is a reproduction circuit of the image information recording / reproducing apparatus of the present invention (not shown) (VT of the recorder side not shown).
The output signal (reproduction signal in the state of encoded data) from the R reproduction circuit) is supplied to the MPEG decoder 14 via the input terminal 2, the fixed contact b and the movable contact v of the changeover switch 6. As described above, the reproduction and decoding of the image data of the two frames immediately after the cut-in point is performed by converting the image information of the I picture in the above two frames into the B picture of the immediately preceding two frames. This is because it is used for decoding.

The quantized information used for the decoding operation in the MPEG decoder 14 is stored in the quantized information memory 15
Is stored. Cut-in point Ci of the specified insert edit decoded by the MPEG decoder 14
The image data of the first one frame of the image data of the two frames immediately before (Ci point in FIG. 3) is the fixed contact b and the movable contact v of the changeover switch 7, and the changeover switch 8 of the screen rearranging unit 16. Via the fixed contact b and the movable contact v of the switch 9 and the fixed contact b and the movable contact v of the changeover switch 9
9 and supplied to the DCT unit 19 in the unit of blocks of 8 pixels × 8 lines (DCT block) and converted into the spatial frequency domain by the division cosine transform. DCT
The DCT coding coefficient output from the unit 19 is the quantization unit 2
Although it is quantized by 0, the quantization operation in the quantization unit 20 is performed by the quantization information read from the quantization information memory 15 via the movable contact v and the fixed contact a of the changeover switch 10, That is, it is performed using the quantized information used for the decoding operation in the MPEG decoder 14.

DCT quantized by the quantizer 20
The coding coefficient is provided to the variable length coding unit 21 and the inverse quantization unit 24. Then, D quantized by the quantizer 20
The CT coding coefficient is variable-length coded by the variable-length coding unit 21 together with the motion vector and coding mode information, and is accumulated in the buffer memory 22. In addition, the quantizer 2
The DCT coding coefficient quantized by 0 is the inverse quantization unit 24.
Is supplied to the dequantization section 24 and the dequantization section 24
The inversely quantized DCT coding coefficient supplied from the above is supplied to the inverse DCT unit 25. The inverse DCT unit 2
The calculation result in 5 is so-called locally decoded image data, which is stored in the prediction memory 26 via the movable contact v and the fixed contact a of the changeover switch 13.

Also, of the image data of the two frames immediately before the designated insert-edit cut-in point Ci (Ci point in FIG. 3) decoded by the MPEG decoder 14, the second one frame Image data passes through the fixed contact b and the movable contact v of the changeover switch 7, the fixed contact b and the movable contact v of the changeover switch 8, and the motion compensation prediction unit 28.
Is supplied to. The motion compensation prediction unit 28 uses the image data recorded in the prediction memory 26 to perform motion compensation prediction.

The prediction signal is supplied as a subtraction signal to the subtractor 18 via the changeover switch 12. In the subtractor 18, the image of the second one frame of the image data of the two frames immediately before the cut-in point Ci supplied from the MPEG decoder 14 through the switch 7 and the changeover switch 8 is displayed. From the data, the changeover switch 1
The prediction signal selected in 2 is subtracted and supplied to the fixed contact a of the changeover switch 9. The fixed contact b of the change-over switch 9 is supplied with the image data of the second one frame of the image data of the two frames immediately before the cut-in point Ci.

In the DCT unit 19, the amount of information is changed between the state where the movable contact v of the changeover switch 9 is switched to the fixed contact a side and the state where the movable contact v is switched to the fixed contact b side. Which has a smaller number of pixels is selected for each macroblock, the DCT operation is performed, and D obtained as the operation result is selected.
The CT coding coefficient is supplied to the quantizing unit 20 and quantized, and the DCT coding coefficient quantized by the quantizing unit 20 is changed by the variable length coding unit 21 together with the motion vector and the coding mode information. Long coded, buffer memory 22
Is accumulated in The image data already stored in the buffer memory 22, that is, the two frames immediately before the cut-in point Ci (Ci point in FIG. 3) of the designated insert edit decoded by the MPEG decoder 14 Of the image data of the first one frame, the DCT coding coefficient quantized by the quantizing unit 20 is the variable length code in the variable length coding unit 21 together with the motion vector and the coding mode information. Along with the encoded data stored in the buffer memory 22 after being converted, it is output as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v of the changeover switch 11 to record a VTR (not shown). Cut-in point Ci of the specified insert edit supplied to the circuit (Ci point in FIG. 3)
The image data of two frames immediately before is newly recorded for the three recording marks. Then, in the sequential recording trace following the cut-in point Ci of the specified insert edit, the reproduction of the state of the encoded data output from the VTR on the reproducing apparatus side and supplied to the dubbing terminal 1 is performed. The signal is supplied from the output terminal 4 to the recording circuit of the VTR of the recorder via the fixed contact a and the movable contact v of the changeover switch 11 and recorded.

In the insert edit recording operation in the editing compressed image data generating apparatus of the image information recording / reproducing apparatus of the present invention shown in FIG. 2, the cut-in point Ci (FIG. The image data of the second frame of the two frames immediately before the (Ci point in the middle) is encoded by performing the inter-frame prediction by limiting the forward prediction to the cut-in of the specified insert edit. The data is re-recorded in the original recording location on the recording medium together with the compressed data of the image data of the first frame of the two frames immediately before the point Ci (Ci in FIG. 3). Therefore, as compared with the case of the insert edit recording operation in the editing compressed image data generator of the image information recording / reproducing apparatus of the present invention shown in FIG. 1, since backward prediction and interpolation prediction are not performed, the code Although the conversion efficiency is deteriorated and the image quality is deteriorated, the delay memory used in the apparatus shown in FIG. 1 is required, and the operation timing control of the image information record data creation unit is facilitated. Further, in the above description of the embodiment, the data input from the reproduction side is the encoded data, but the data input from the reproduction side may be the image information supplied to the input terminal 3. Of course.
In that case, it is natural to output all data to the recording circuit after DCT, quantization and variable length coding.

Next, another mode of insert edit recording by the generator of compressed image data for editing of the image information recording / reproducing apparatus of the present invention shown in FIG. 2 when insert edit recording is performed will be described. First, two frames of image data immediately after the specified insert edit cutout point Co (Co point in FIG. 3), that is, the specified insert edit cutout point Co (in FIG. 3) The image data of the first frame (I picture) in the image data in the three recording marks immediately after the (Co point) is reproduced. The reproduction signal is a reproduction circuit of the image information recording / reproducing apparatus of the present invention (not shown) (VT of the recorder side not shown).
The output signal (reproduction signal in the state of encoded data) from the R reproduction circuit) is supplied to the MPEG decoder 14 via the input terminal 2, the fixed contact b and the movable contact v of the changeover switch 6.

The quantized information used for the decoding operation in the MPEG decoder 14 is the quantized information memory 15
Is stored. Cut-out point C of specified insert edit decoded by the MPEG decoder 14 described above.
Of the two frames of image data immediately after o (point Co in FIG. 3), the first one frame of image data includes the fixed contact b and the movable contact v of the changeover switch 7, and the screen rearranging unit 16
Is supplied to the DCT unit 19 through the fixed contact b and the movable contact v of the changeover switch 8 and the fixed contact b and the movable contact v of the changeover switch 9, and in the DCT unit 19, a block of 8 pixels × 8 lines (DCT). Block) units are transformed into the spatial frequency domain by the division cosine transform. DC above
The DCT coding coefficient output from the T unit 19 is quantized by the quantizing unit 20, and the quantizing operation in the quantizing unit 20 is performed by connecting the fixed contact a and the movable contact v of the changeover switch 10 to each other. This is performed using the quantized information read from the quantized information memory 15 via the quantized information, that is, the quantized information used for the decoding operation in the MPEG decoder 14.

DCT quantized by the quantizer 20
The coding coefficient is not variable-length coded, but is supplied to the dequantization unit 24, which inverts the quantized DCT coding coefficient supplied from the quantization unit 20. It is quantized and given to the inverse DCT unit 25. Inverse D above
The calculation result in the CT unit 25 is so-called locally decoded image data, which is stored in the prediction memory 26 via the movable contact v and the fixed contact a of the changeover switch 13.

The coded data supplied to the dubbing terminal 1 is composed of one intra-frame coded image (I picture) and one bidirectional predictive coded image (B picture) as described above. A combination (I, B) is used as a unit of encoded image group, and three recording marks on a recording medium (magnetic tape) are formed by digital image data corresponding to the unit of encoded image group. It is a reproduction signal in the state of encoded data which enables the fixed contact a and the movable contact v of the changeover switch 11 except for the last two frames of image data in the image information to be insert edited and recorded.
Is sent to the output terminal 4 via and, which is V on the recorder side.
It is supplied to the recording circuit of TR and recorded on the magnetic tape.

In the reproduction signal in the state of the encoded data supplied to the dubbing terminal 1, the image data for the last two frames of the image information to be insert-edited and recorded is the changeover switch 6. Fixed contact a and movable contact v
And is supplied to the MPEG decoder 14 via. Said M
The quantized information used for the decoding operation in the PEG decoder 14 is stored in the quantized information memory 15. Of the two-frame image data decoded by the MPEG decoder 14, the first one-frame image data (I-picture frame data) is the fixed contact b and the movable contact v of the changeover switch 7 and the screen. It is supplied to the DCT unit 19 through the fixed contact b and the movable contact v of the changeover switch 8 of the rearrangement unit 16 and the fixed contact b and the movable contact v of the changeover switch 9, and in the DCT unit 19, 8 pixels × 8 lines are supplied. The block (DCT block) is transformed into the spatial frequency domain by the division cosine transform.

DCT output from the DCT unit 19
The coding coefficient is quantized by the quantizing unit 20, and the quantizing operation in the quantizing unit 20 is performed by the quantizing information memory 1 via the fixed contact a and the movable contact v of the changeover switch 10.
5 is performed, that is, the quantized information used for the decoding operation in the MPEG decoder 14 is performed. The DCT coding coefficient quantized by the quantizing unit 20 is variable-length coded by the variable-length coding unit 21 and stored in the buffer memory 22, and the DCT code quantized by the quantizing unit 20. The dequantization coefficient is dequantized by the dequantization unit 24 and then given to the inverse DCT unit 25. The calculation result in the inverse DCT unit 25 is so-called locally decoded image data, which is stored in the prediction memory 27 via the movable contact v and the fixed contact b of the changeover switch 13.

Next, of the two frames of image data obtained by decoding the last two frames of the coded image data of the image information to be insert-edited and recorded, two of the two frames of image data are obtained. The image data of the 1st frame is
It is given to the motion compensation prediction units 28 and 29 via the fixed contact b and the movable contact v of the changeover switch 8. The motion compensation prediction units 28 and 29 described above perform motion compensation prediction on the image data of the second one frame described above using the image data stored in the prediction memories 26 and 27.

Then, the motion compensation prediction based on the image data stored in the prediction memory 26, the motion compensation prediction based on the image data stored in the prediction memory 27, and the image data from both of them are averaged. Among the interpolative predictions obtained by averaging in the unit 30, the most efficient prediction signal is selected by the change-over switch 12 and given to the subtractor 18 as a subtraction signal. In the subtracter 18, the most efficient selected by the changeover switch 12 from the image data of the second one frame of the image data of the last two frames of the image information to be insert-edited. Of the predicted signal with good
To supply. The fixed contact b of the changeover switch 9 has
The image data of the second one frame of the image data of the last two frames of the image information to be insert-edited is supplied.

In the DCT unit 19, the amount of information is changed between the state where the movable contact v of the changeover switch 9 is switched to the fixed contact a side and the state where the movable contact v is switched to the fixed contact b side. Which has a smaller number of pixels is selected for each macroblock, the DCT operation is performed, and D obtained as the operation result is selected.
The CT coding coefficient is supplied to the quantizing unit 20 and quantized, and the DCT coding coefficient quantized by the quantizing unit 20 is changed by the variable length coding unit 21 together with the motion vector and the coding mode information. Long coded, buffer memory 22
Is accumulated in Then, of the image data already stored in the buffer memory 22, that is, the reproduction signal in the state of the encoded data supplied to the dubbing terminal 1, the last of the image information to be insert-edited and recorded. The DCT coding coefficient quantized by the quantizer 20 for the first one-frame image data (I-picture frame data) in the two-frame image data of
The changeover switch 11 together with the motion vector and the coding mode information, together with the coded data that has been variable-length coded by the variable-length coding unit 21 and accumulated in the buffer memory 22.
Is output as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v, and is supplied to the recording circuit of the VTR (not shown). The image data of the two frames immediately before the Co point) is newly recorded on the three recording traces on which the image data was recorded. Further, in the description of the embodiment, the data input from the reproducing side is the encoded data, but the data input from the reproducing side may be the image information supplied to the input terminal 3, as a matter of course. . In that case, it is natural to output all data to the recording circuit after DCT, quantization and variable length coding.

Next, another embodiment of the insert edit recording by the generator of the compressed image data for editing of the image information recording / reproducing apparatus of the present invention shown in FIG. 2 when the insert edit recording is performed will be described. The fixed contact a and the movable contact v of the changeover switch 11 are excluded except for the last two frames of the encoded data insert-recorded by the encoded data in the image compressed state supplied to the dubbing terminal 1.
It is output as a video bit stream from the output terminal 4 via and is supplied to a recording circuit of a VTR (not shown) for insert recording on a magnetic tape. Further, the image data of the last two frames of the encoded data insert-recorded by the encoded data in the image compressed state supplied to the dubbing terminal 1 is the fixed contact a and the movable contact of the changeover switch 6. It is supplied to the MPEG decoder 14 via v.

The quantization information used for the decoding operation in the MPEG decoder 14 is the quantization information memory 15
Is stored. Of the two-frame image data decoded by the MPEG decoder 14, the first one-frame image data is the fixed contact b and the movable contact v of the changeover switch 7, and the changeover switch of the screen rearranging unit 16. 8 fixed contacts b and movable contacts v, and the fixed contact b and movable contact v of the changeover switch 9 are supplied to the DCT unit 19, and in the DCT unit 19, a block of 8 pixels × 8 lines (DCT block) Then, it is transformed into the spatial frequency domain by the division cosine transform. The DCT coding coefficient output from the DCT unit 19 is quantized by the quantizing unit 20, and the quantizing operation in the quantizing unit 20 is performed by the movable contact v and the fixed contact a of the changeover switch 10. This is performed using the quantized information read from the quantized information memory 15 via, that is, the quantized information used for the decoding operation in the MPEG decoder 14.

DCT quantized by the quantizer 20
The coding coefficient is provided to the variable length coding unit 21 and the inverse quantization unit 24. Then, D quantized by the quantizer 20
The CT coding coefficient is variable-length coded by the variable-length coding unit 21 together with the motion vector and coding mode information, and is accumulated in the buffer memory 22. In addition, the quantizer 2
The DCT coding coefficient quantized by 0 is supplied to the dequantization unit 24, and the dequantization unit 24 dequantizes the quantized DCT coding coefficient supplied from the quantization unit 20 described above, It is given to the inverse DCT unit 25. The inverse DCT section 25
The calculation result at is so-called locally decoded image data, which is stored in the prediction memory 26 via the movable contact v and the fixed contact a of the changeover switch 13.

Of the reproduced signal in the state of the encoded data supplied to the dubbing terminal 1, the second image data of the image data for the last two frames of the image information to be insert edited and recorded. Is the MPEG decoder 14 described above.
Fixed contact b of the changeover switch 8 after being decoded in
And the movable contact v, and is given to the motion compensation prediction unit 28. Then, in the motion compensation predicting unit 28, the last two of the image information to be insert-edited and recorded are added.
The second image data of the image data for the frame,
Motion compensation prediction is performed using the image data stored in the prediction memory 26.

Then, the prediction signal output from the motion compensation prediction unit 28 is given as a subtraction signal to the subtractor 18 via the changeover switch 12, and the subtractor 18 should be subjected to the insert edit recording described above. The second 1 of the image data for the last 2 frames of the image information
The prediction signal supplied via the changeover switch 12 is subtracted from the image data of the frame, and the changeover switch 9
To the fixed contact a. At the fixed contact b of the change-over switch 9, two of the image data for the last two frames of the image information to be insert-edited and recorded are recorded.
The image data of the 1st frame is supplied.

In the DCT section 19, the amount of information is changed between the state where the movable contact v of the changeover switch 9 is switched to the fixed contact a side and the state where the movable contact v is switched to the fixed contact b side. Which has a smaller number of pixels is selected for each macroblock, the DCT operation is performed, and D obtained as the operation result is selected.
The CT coding coefficient is supplied to the quantizing unit 20 and quantized, and the DCT coding coefficient quantized by the quantizing unit 20 is changed by the variable length coding unit 21 together with the motion vector and the coding mode information. Long coded, buffer memory 22
Is accumulated in Then, of the image data already stored in the buffer memory 22, that is, the image data of the last two frames of the image information to be insert-edited and recorded, which is decoded by the MPEG decoder 14, Regarding the image data of the first frame of
The DCT coding coefficient quantized by the quantizing unit 20, together with the motion vector and the coding mode information, together with the coded data that has been variable-length coded by the variable-length coding unit 21 and accumulated in the buffer memory 22, An image of the two frames immediately before the cut-out point Co is output as a video bit stream from the output terminal 4 via the fixed contact b and the movable contact v of the changeover switch 11 and supplied to the VTR recording circuit (not shown). Data is newly recorded for the three recording marks on which the data was recorded. In the insert edit recording operation described above, since the backward prediction and the interpolative prediction are not performed, the coding efficiency is deteriorated and the image quality is deteriorated. However, as in the case of the above-described embodiment, the insert edit recording operation is performed. Prior to the above, it is not necessary to reproduce and decode the image data of the first frame (I picture) of the two frames immediately after the cut-out point Co, and the operation timing control of the image information recording data creation unit is not necessary. The advantage is that it becomes easier. Further, in the description of the embodiment, the data input from the reproducing side is the encoded data, but the data input from the reproducing side may be the image information supplied to the input terminal 3, as a matter of course. . In that case, it is natural to output all data to the recording circuit after DCT, quantization and variable length coding.

[0055]

As is apparent from the above detailed description, the generator of compressed image data for editing of the image information recording / reproducing apparatus of the present invention comprises one intra-frame coded image and a predetermined number of bidirectional images. A combination with a predictive coded image is a coded image group of a unit, and by digital image data corresponding to a predetermined number of the coded image group of the unit,
When editing and recording image data for insert recording or assembling recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on a recording medium, each of the above-mentioned predetermined values immediately before and immediately after the editing point on the recording medium By decoding the digital image data of the coded image group in the unit of the number, the bidirectional predictive coded image in the coded image group of the predetermined number of units immediately before the editing point is set to the predetermined number immediately before the editing point. The picture of the intra-frame coded image in the coded image group of the unit and the picture of the image data newly prepared to be recorded following the edit point are coded by bidirectional prediction, After that, it is possible to record the newly prepared image data so that it will be recorded after the edit point. The bidirectional prediction code in the coded image group of the predetermined number of units immediately before the edit point is decoded by decoding the digital image data of the above-mentioned predetermined number of units of the coded image group immediately before and immediately after the edit point on the medium. The encoded image is encoded as a forward predictive encoded image in the encoded image group of a predetermined number of units immediately before the above-mentioned edit point, and is newly recorded so as to be continuously recorded at the edit point after the above-mentioned edit point. The prepared image data is recorded, or the digital image data of the coded image group of each of the above-mentioned predetermined number of units immediately after the cutout edit point on the recording medium is decoded, and the predetermined number immediately before the edit point is recorded. The bidirectional predictive coded image in the coded image group of each unit is a coded image group of a predetermined number of units immediately after the editing point described above. In the frame, the picture of the intra-frame coded image and the picture in the image data newly prepared to be recorded after the edit point are coded by bidirectional prediction, and the edit point is followed by the edit point. Immediately before the edit point, the image data newly prepared to be recorded is recorded, or the digital image data of a predetermined number of units of coded image groups immediately before the cutout edit point on the recording medium is decoded. The bidirectional predictive encoded image in the predetermined number of units of the encoded image group is encoded as the forward predictive encoded image in the predetermined number of units of the encoded image group immediately before the editing point, and After the edit point, the image data newly prepared to be recorded after the edit point is recorded. According to the editing compressed image data generating apparatus of the image information recording / reproducing apparatus of the present invention, it is possible to prevent the deterioration of the image quality of the B-hitchure immediately before the cut-in point or the cut-out point in the insert editing. According to the invention, the conventional problems described above can be solved well.

[Brief description of drawings]

FIG. 1 is a block diagram of a generator of compressed image data for editing of an image information recording / reproducing apparatus of the present invention.

FIG. 2 is a block diagram of a compressed image data generator for editing of the image information recording / reproducing apparatus of the present invention.

FIG. 3 is an explanatory diagram of a track pattern of a magnetic tape.

FIG. 4 is a block diagram showing a conventional example of a decoder.

FIG. 5 is a block diagram showing a conventional example of an encoder.

[Explanation of symbols]

5 ... Delay memory, 7 to 13 ... Changeover switch, 14 ... M
PEG decoder, 15 ... Quantized information memory, 16 ... Screen rearranging unit, 17 ... Frame memory, 18 ... Subtractor, 1
9 ... DCT unit, 20 ... Quantization unit, 21 ... Variable length coding unit, 22, 39 ... Buffer memory, 23 ... Rate control unit, 24, 41 ... Inverse quantization unit, 25 ... Inverse DCT unit, 2
6, 27 ... Prediction memory, 28, 29, 34 ... Motion compensation prediction unit, 31 ... Switch, 32, 43 ... Adder, 33 ... Image memory, 34 ... Motion compensation prediction unit, 36 ... Analog-digital converter, 37, 46 ... Format conversion unit, 40 ...
Variable length decoding unit, 41 ... Inverse quantization unit, 42 ... Inverse DCT
Part, 47 ... DAC,

Claims (5)

[Claims] 1. A combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images is set as a unit of coded image group, and a predetermined number of the unit of coded image group is used. When performing edit recording of image data for which insert recording or assembling recording is performed using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on the recording medium with corresponding digital image data, edit points on the recording medium Of the predetermined number of units of the encoded image group digital image data immediately before and immediately after, the bidirectional predictive encoded image in the predetermined number of units of the encoded image group immediately before the editing point, The picture of the intra-frame coded image in the coded image group of a predetermined number of units immediately before the above-mentioned edit point and the picture recorded after the edit point are recorded. Image data newly prepared to be encoded and re-recorded by bidirectional prediction from a picture in newly prepared image data, and to be recorded continuously after the edit point after the above-mentioned edit point. An apparatus for generating compressed image data for editing of an image information recording / reproducing apparatus adapted to record. 2. A combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images is a unit of coded image group, and a predetermined number of the unit of coded image group is used. When performing edit recording of image data for which insert recording or assembling recording is performed using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on the recording medium with corresponding digital image data, edit points on the recording medium Of the predetermined number of units of the encoded image group digital image data immediately before and immediately after, the bidirectional predictive encoded image in the predetermined number of units of the encoded image group immediately before the editing point, Coded and recorded again as a forward predictive coded image in a coded image group of a predetermined number of units immediately before the editing point, and A generation device for compressed image data for editing of an image information recording / reproducing device, which is configured to record image data newly prepared to be recorded after the editing point. 3. The image information recording / reproducing according to claim 1 or 2, wherein the intra-frame coded image is subjected to the same quantization as when the coded data before decoding is generated to perform the image information compression. Device for generating compressed image data for device editing. 4. A combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images is set as a unit of coded image group, and a predetermined number of the unit of coded image group is used. When editing and recording image data for insert recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on the recording medium with corresponding digital image data, it is recorded on the recording medium of the recorder. The intra-frame coded image in the coded image group of the unit of the image data recorded on the recording medium of the above-mentioned recorder located immediately after the cut-out edit point set for the image data, and the above-mentioned recorder. The playback image data prepared for insert recording before the cutout edit point set for the image data recorded on the recording medium of A reproduced image prepared for insert recording before the cutout edit point described above by bidirectional prediction with the intra-frame encoded image of the encoded image group of the last unit as the encoded image group of the last unit. Prepared for generating a bidirectional predictive coded image of a unit coded image group which is the last unit coded image group in the data, and insert-recording it before the cutout edit point described above. Of the compressed image data for editing of the image information recording / reproducing apparatus, insert-recording is performed by using the bidirectional predictive encoded image of the encoded image group of the unit that is the last encoded image group in the reproduced image data. Generator. 5. A combination of one intra-frame coded image and a predetermined number of bidirectional predictive coded images is set as a unit of coded image group, and a predetermined number of the unit of coded image group is used. When editing and recording image data for insert recording using an image information recording / reproducing apparatus capable of forming a predetermined number of recording traces on the recording medium with corresponding digital image data, it is recorded on the recording medium of the recorder. Bidirectional predictive coding of the coded image group of the unit that is the last unit of the coded image group in the reproduced image data prepared for insert recording before the cutout edit point set for the image data Used to ensure that the image is generated with interframe prediction limited to forward prediction and insert it before the cutout edit point described above. Editing compression of an image information recording / reproducing apparatus adapted to perform insert recording by using a bidirectional predictive encoded image of a unit of encoded image group which is the last unit of encoded image group in intended reproduced image data. Image data generator.