JP3821714B2 - Moving picture encoding apparatus and moving picture recording / reproducing apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a moving image encoding apparatus having a function of encoding a moving image signal including a part of continuous reproduction of the same image, such as slow reproduction, and a moving image recording / reproducing apparatus using the same.
[0002]
[Prior art]
In recent years, a moving image signal is encoded using a well-known moving image compression technique such as MPEG, and the moving image encoded data is recorded on a recording medium such as a magnetic recording tape, a hard disk, or a DVD-RAM. And a system for performing decoding processing on moving image encoded data recorded on a recording medium and reproducing it as a moving image signal have been developed.
[0003]
In such a system, an apparatus having a function of reproducing moving image encoded data recorded on a recording medium and transmitting the encoded data to another device has been developed. On the device side, it is necessary to convert the data into a moving image encoded data format that can be handled on the receiving device side, and to send it out.
[0004]
For example, a signal recorded in an MPEG program stream (hereinafter referred to as MPEG_PS) format on a DVD is converted into an MPEG transport stream (hereinafter referred to as MPEG_TS) format, and set top box (hereinafter referred to as STB) via IEEE1394. In the case of normal reproduction, it is relatively easy to convert from the MPEG_PS format to the MPEG_TS format in an encoded state using a technique called transcoding. However, at the time of slow reproduction or the like, it is difficult to convert to an MPEG_TS format that can be decoded by an STB decoder having no special mechanism.
[0005]
Therefore, it is easy to add a moving image encoding device that can encode a decoded moving image in the MPEG_TS format so as to be compatible with any reproduction mode. However, since a motion-compensated predictive coding circuit that requires a large-scale circuit is mounted, there has been a problem that the cost is greatly increased.
As an example of a method for solving such a problem, in a moving image re-encoding device that re-encodes a moving image signal obtained by decoding encoded data, a code included in the compressed moving image data before decoding A moving image re-encoding device that can realize high-quality re-encoding while omitting a motion vector detection circuit necessary for motion compensation prediction by encoding using an encoding parameter is disclosed in JP 2000- No. 244929.
[0006]
Here, the technique disclosed in Japanese Patent Application Laid-Open No. 2000-244929 will be specifically described with reference to FIGS.
FIG. 8 is a block diagram showing a configuration of a system including a moving image re-encoding device, FIG. 9 is a block diagram showing a configuration example of a video encoder in the moving image re-encoding device, and FIG. 10 is a moving image re-encoding. It is a schematic diagram which shows the example of operation | movement timing of an optimization apparatus.
[0007]
8, 110 is a data separation unit, 111 is an audio data buffer memory, 112 is a moving image data buffer memory, 113 is a sub-picture data buffer memory, 114 is a navigation data buffer memory, 115 is an audio decoder, 116 is a video decoder, 117 is a sub-picture decoder, 118 is a video overlay processor, 119 is an adder, 120 is a GUI image signal generator, 121 is an audio encoder, 122 is a video encoder, 123 is a frame memory, and 124 is a re-encoding unit audio data buffer Memory, 125 is a re-encoding unit moving image data buffer memory, 126 is a multiplexing unit, 127 is an IEEE 1394 interface, 128 is a re-encoding unit, 129 is an audio signal D / A converter, and 130 is a video signal D / A A converter, 131 is data Mori, 150 is an optical disk device.
[0008]
In FIG. 9, 170 is a motion compensation unit, 172 is a motion compensation mode and DCT type determination unit, 173 is a rate control unit, 174 is a DCT processing unit, 175 is a quantization unit, 176 is an inverse quantization unit, and 177 is an inverse DCT. A processing unit 178 is a variable length coding unit.
[0009]
10, (a) is the decoding timing of the moving image encoded data, (b) is the decoding timing at the time of 1 / 3-speed slow playback by the video decoder 116, and (c) is the 1 / 3-speed slow playback. (D) is a video re-encoding processing timing, and (e) is a re-decoded signal.
[0010]
The operation of the system including the moving image re-encoding device having the above-described configuration will be briefly described.
The optical disk device 150 is of the DVD standard, for example, and audio data, encoded moving image data, sub-picture data, navigation data, etc. are multiplexed and recorded.
A signal reproduced from the optical disk device 150 is input to the data separation unit 110 and separated into audio data 151, moving image data 152, sub-video data 153, and navigation data 154. The separated data 151, 152, 153, 154 are decoded by the audio decoder 115, the video decoder 116, and the sub-picture decoder 117 via the buffer memories 111, 112, 113, 114, respectively.
[0011]
For the moving image signal 156 obtained by decoding the moving image data by the video decoder 116, the sub-picture signal 157 obtained by decoding the sub-picture data by the sub-picture decoder 117, and the reproduction apparatus uses a GUI (Graphical User). The GUI image signal 159 that is independently generated by the image signal generator 120 in response to the user operation 160 is subjected to overlay (combination) processing by the video overlay processor (OSD) 118 via the adder 119.
[0012]
The audio signal 155 obtained by decoding the audio data by the audio decoder 115 and the moving image signal 158 overlay-processed by the video overlay processing unit 118 are converted into analog signals 161, 130 by D / A converters 129 and 130, respectively. It is converted to 162 and output to the outside. Further, the audio signal 155 and the moving image signal 158 after the overlay processing are re-encoded by the audio encoder 121 and the video encoder 122, respectively. The audio data and moving image data obtained by this re-encoding are multiplexed into a transport stream by the multiplexing unit 126 via the buffer memories 124 and 125, respectively, and further digitally output from the IEEE 1394 interface 127.
[0013]
The video decoder 116 extracts the encoding parameters 164 included in the moving image data 152 before decoding such as the motion vector for each macroblock, the encoding mode, and the quantization step simultaneously with the decoding. The parameter 164 is temporarily stored in the data memory 131. The video encoder 122 reads out encoding parameters corresponding to an image frame and a macroblock to be re-encoded from the encoding parameter data temporarily stored in the data memory 131, and predetermined motion vector information given from the outside. It is input together with 165.
[0014]
FIG. 9 shows a configuration example of the video encoder 122 in the moving image re-encoding device. In FIG. 9, the same components as those shown in FIG. 8 are denoted by the same reference numerals.
The video encoder 122 is based on a motion compensated predictive orthogonal transform coding method represented by MPEG2 and the like, and as shown in FIG. 9, a motion compensation unit 170, a motion compensation mode (coding mode), and a DCT type determination unit 172. , A DCT processing unit 174, a quantization unit 175, an inverse quantization unit 176, an inverse DCT processing unit 177, a frame memory 123, a rate control unit 173, and a variable length coding unit 178.
[0015]
The video encoder 122 is characterized in that it does not have a motion vector detection circuit necessary for a moving picture coding apparatus such as a normal MPEG2 encoder. In a normal moving image encoding apparatus, a huge block matching operation is required to detect a motion vector for each macroblock. In general, the hardware resources that the motion vector detection circuit occupies in the entire encoding apparatus occupy a very high ratio, but the video encoder 122 does not need such a motion vector detection unit with a large hardware scale. That is, in this moving image re-encoding device, motion compensation is performed using both the motion vector information of the encoding parameter 164 extracted from the moving image data 152 before decoding and the predetermined motion vector information 165 given from the outside. I do.
[0016]
Next, the motion compensation mode and DCT type determination unit 172 determines the motion vector information or motion of the encoding parameter 164 from the power or magnitude of the prediction error signal by each motion compensation and the power or magnitude of the encoding target image signal. Whether to use vector information 165 or to perform intra coding is determined for each macroblock.
Next, for a macroblock for which predictive coding based on motion vector information of the coding parameter 164 extracted from the video data 152 before decoding is selected, a DCT type described later is extracted from the video data 152 before decoding. In other cases, the optimal DCT type is detected.
[0017]
The rate control unit 173 determines a quantization step for each macroblock used in the quantization unit 175. That is, for a macroblock for which predictive coding based on motion vector information of the coding parameter 164 extracted from the moving image data 152 before decoding is selected, the moving image data 162 is not subject to output bit rate restrictions. When the quantization step extracted from the above is used as it is and the output bit rate is restricted, the quantization step is determined with reference to the generated code amount 180 so that the desired bit rate is obtained. If the macroblock for which the predetermined motion vector information 165 from the outside is selected or the macroblock for which intra coding is selected is not subject to the output bit rate restriction, the quantization extracted from the moving image data 152 is used. If the output bit rate is constrained using a step or a predetermined predetermined quantization step, the quantization step is determined with reference to the generated code amount 180 so that the desired bit rate is obtained.
[0018]
In the moving image re-encoding apparatus configured as described above, a method of performing decoding by performing slow reproduction processing on moving image encoded data input to the video decoder 116, and decoding performed by performing the slow reproduction processing A re-encoding procedure for the encoded data will be described in detail. For convenience of explanation, the slow playback speed is 1/3.
[0019]
Here, the features of the MPEG encoding and decoding methods will be briefly described. MPEG encoding consists of motion compensation prediction, DCT processing, and quantization processing, and is classified into three types of I picture, P picture, and B picture according to the prediction direction of the time axis, one I picture and 0 picture. This is performed based on the GOP structure composed of the above P picture and B picture. An I picture is an intra-frame encoded image, a P picture is a temporally preceding I picture or an inter-frame predictive encoded image predicted from a P picture, a B picture is temporally backward and forward I and P pictures, or It is an inter-frame predictive encoded image predicted from one of them. Therefore, an I picture that is a prediction source is required to decode a P picture, and an I picture and a P picture that are prediction sources are required to reproduce a B picture.
[0020]
In FIG. 10, I, B, and P indicate the type of each picture, and the numbers that follow them indicate the display order. For example, the I2 picture is an intra-frame encoded image, and the P5 picture is an inter-frame predicted from the I2 picture. The predictive encoded image, the B0 picture, and the B1 picture are interframe predictive encoded images that are predicted from the I2 picture, and the B3 and B4 pictures are interframe predictive encoded images that are predicted from the I2 and P5 pictures. Therefore, decoding of P5 picture requires I2 picture, decoding of B0 picture and B1 picture requires I2 picture, and decoding of B3 picture and B4 picture requires I2 picture and P6 picture.
[0021]
When decoding is performed according to the decoding method described above, the general MPEG moving image decoding timing is as shown in FIG. This is because the B0 picture and the B1 picture cannot be referred to until the processing of the I2 picture is performed, and the B3 picture and the B4 picture cannot be referred to until the processing of the I2 picture and the P5 picture is performed. When performing slow playback at 1/3 speed, it is only necessary to display the same frame data three times at a time. However, as shown in FIG. A device equipped with an MPEG decoder that does not have a device cannot be displayed well.
[0022]
Accordingly, when decoding is performed by performing slow reproduction processing on the encoded moving image data, decoding is performed at a timing that is an integral multiple of the frame period (3 in the case of realizing slow reproduction at 1/3 speed). (Times), when decoding is not performed, slow playback is realized by repeatedly displaying the same display frame, and the decoded moving image signal is as shown in FIG. In FIG. 10 (c), the part with only the number is the part in which data of the same frame as the I, P, B picture with the same number is repeatedly displayed.
[0023]
On the other hand, when the re-encoding is performed on the signal shown in FIG. 10C, which is the decoded moving image signal, as the slow reproduction at 1/3 speed, the re-encoded output is between the encoding procedure and the frame. Considering the prediction structure, the re-encoding process is performed at the timing shown in the interval 10 (d). This is because the B0 picture and the B1 picture cannot be processed until the I2 picture is processed because re-encoding is performed using the encoding parameter extracted when the encoded data is decoded.
[0024]
Here, the frames of B0 picture and B1 picture are re-encoded as the same moving image frame as a B picture, and the P2 ′ picture is re-encoded as the same moving image frame as the I2 picture as a P picture. At that time, since the reference image and the predicted image coincide with each other, by performing motion compensation predictive coding of the zero vector, the prediction error signal is always zero, and coding with very high coding efficiency can be performed. ing.
[0025]
[Problems to be solved by the invention]
However, in the above-described conventional technique, when the same image frame as the I picture is re-encoded as a P ′ picture using motion compensated prediction encoding using a zero vector, the difference between the P ′ pictures Is obtained as a difference between a moving picture signal of an I picture that has already been re-decoded and a P picture that is a moving picture signal (that is, an I picture before re-encoding), so that a motion-compensated prediction code using a zero vector is used. Even if the conversion is performed, an error may occur because the processing by the DCT processing unit 174 and the quantization unit 175 shown in FIG. 9 is performed, and the difference is not always zero. Therefore, the re-decoded P ′ picture and the re-decoded I picture are not the same image, and the re-decoded video signal is as shown in FIG.
[0026]
In addition, decoding of moving image data is performed at a timing that is an integral multiple of the frame period (3 times when slow playback at 1/3 speed is realized), and when not decoded, the same display frame is repeatedly displayed. Therefore, the number of pictures constituting the GOP at the time of re-encoding is increased by an integral multiple of the frame period (3 times in the case of re-encoding for slow reproduction at 1/3 speed).
[0027]
In this case, there is no problem if the re-encoded moving image compressed data is output via IEEE 1394 and decoded and displayed on the receiving side, but the re-encoded moving image compressed data is recorded on a randomly accessible medium. In the case, the following problems occur.
That is, when random access such as reproduction from the middle of the GOP is performed on the re-encoded moving image compressed data, the decoding process of the I picture and the P picture necessary for decoding the B and P pictures is performed. , The P2 ′ picture and the P5 ′ picture increase, and there arises a problem that it becomes more noticeable in reverse reproduction, reverse slow reproduction, and the like. In addition, when performing GOP unit linkage editing work, the time required for one GOP becomes longer, which causes a problem that the degree of freedom decreases.
[0028]
In addition, since the re-encoding method in the conventional technique described above is based on the premise that the encoding parameter extracted when the encoded data is decoded is used, the moving image encoded data other than MPEG such as the DV system is used. There was a problem that could not be adapted in some cases.
[0029]
As described above, in the conventional technique described above, it is possible to add a moving image re-encoding device at a low cost, but there are various problems. On the other hand, some commercially available MPEG data reproducing apparatuses include a circuit for recording an analog input signal from the outside by MPEG encoding. In such an apparatus, a moving image is used only for re-encoding. Since it is not necessary to add an image encoding device, it is possible to provide an encoding device equipped with a motion vector detection circuit necessary for motion compensation prediction.
[0030]
The present invention has been proposed in view of the above-described circumstances, and is based on the premise that an encoding device equipped with a motion vector detection circuit necessary for motion compensation prediction is used, and in an encoding unit for processing of consecutive identical image frames. , When the encoded video data is decoded, it is encoded so that the same image display of the image can be maintained, and the GOP configuration is controlled according to the switching of the image, thereby displaying the screen at the time of random access reproduction. A moving picture coding apparatus capable of preventing the increase in the number of processed pictures up to the present time, increasing the degree of freedom in connecting and editing, and capable of performing the same coding for a compression format other than MPEG. It is an object of the present invention to provide a moving image recording / reproducing apparatus.
[0031]
[Means for Solving the Problems]
The moving picture coding apparatus according to the present invention is a moving picture coding apparatus for coding a moving picture signal, and whether or not the moving picture signal to be coded is the same as the moving picture signal of the immediately preceding or following frame. Switching detection means for detecting a moving picture, moving picture encoding means for generating moving picture encoded data by encoding a moving picture signal, and difference for generating encoded data of a B picture of difference 0 Encoded data output means for selecting encoded data output from the moving picture encoding means and encoded data output from the differential 0 encoded data generation means; , The encoded data selection means comprises: When the moving image signal to be encoded is the same as the image of the immediately following frame, the B picture encoded data of the backward prediction of the difference 0 output from the difference 0 encoded data generating means is selected and encoded. If the moving image signal to be played is different from the image of the immediately following frame, the encoded data output from the moving image encoding means is It is characterized by selecting.
[0033]
In addition, the moving picture coding apparatus according to the present invention sets an upper limit for the number of pictures constituting the GOP, and sets an upper limit for setting the number of pictures constituting the GOP according to the slow playback speed of the input moving picture signal. It is characterized by being variable in the following.
[0034]
In the video encoding device according to the present invention, in the video encoding device for encoding the video signal, whether the video signal to be encoded is the same as the video signal of the immediately preceding or subsequent frame. Image switching detection means for detecting whether or not, moving picture encoding means for generating moving picture encoded data by encoding a moving picture signal, and generating encoded data of B and P pictures with a difference of 0 Difference 0 encoded data generating means, encoded data output from the moving picture encoding means, and encoded data selection for selecting encoded data output from the difference 0 encoded data generating means Means and
The encoded data selection means is based on the information detected by the image switching detection means, and the encoded data output from the moving picture encoding means and the encoded data output from the difference 0 encoded data generation means One of the data is selected.
[0035]
In the video encoding device according to the present invention, in the video encoding device for encoding the video signal, whether the video signal to be encoded is the same as the video signal of the immediately preceding or subsequent frame. Image switching detection means for detecting whether or not, moving picture encoding means for generating moving picture encoded data by encoding a moving picture signal, and generating encoded data of B and P pictures with a difference of 0 Difference 0 encoded data generating means, encoded data output from the moving picture encoding means, and encoded data selection for selecting encoded data output from the difference 0 encoded data generating means And the encoded data selection means comprises: When the moving image signal to be encoded is the same as the image of the immediately preceding or subsequent frame and is encoded as a B picture or a P picture, the backward prediction of the difference 0 output from the difference 0 encoded data generation means is performed. Select one of B-picture encoded data, B-picture encoded data for forward prediction with difference 0, and P-picture encoded data for forward prediction with difference 0, and the moving picture signal to be encoded is the next frame The encoded data output from the moving image encoding means is It is characterized by selecting.
[0037]
In the moving image encoding apparatus according to the present invention, the image switching detection unit refers to switching information of a moving image signal supplied from the outside, and the moving image of the frame immediately before or immediately after the input moving image signal is input. It is characterized by detecting whether or not it is the same as the image signal.
[0038]
The moving image recording / reproducing apparatus according to the present invention is a moving image recording / reproducing apparatus for recording / reproducing encoded data obtained by encoding a moving image signal.
The moving image encoding apparatus, a recording medium for recording and reproducing moving image encoded data, and a moving image decoding means for decoding the encoded data,
When performing slow playback or still playback of the signal recorded on the recording medium, whether or not the video signal decoded by the video decoding means is the same as the video signal of the immediately preceding or subsequent frame Switching information is generated, and it is detected by the image switching detection means in the moving image encoding device whether or not it is the same as the moving image signal of the immediately preceding or immediately following frame with reference to the switching information. To do.
[0039]
The moving image recording / reproducing apparatus according to the present invention is a moving image recording / reproducing apparatus for recording / reproducing encoded data obtained by encoding a moving image signal.
The moving image encoding device, a recording medium for recording and reproducing moving image encoded data, and a moving image decoding means for decoding moving image encoded data supplied from an external device,
When an external device performs slow playback or still playback, the moving image signal decoded based on the screen switching information added to the moving image encoded data by the moving image decoding means is the moving image of the frame immediately before or after Switching information as to whether or not it is the same as the image signal is generated, and the same as the moving image signal of the frame immediately before or after referring to the switching information by the image switching detecting means in the moving image encoding device It is characterized by detecting whether or not.
[0040]
With the above-described configuration, when processing an image in which the same moving image frame continues, such as slow playback, B-picture encoded data for backward prediction with a difference of 0 or forward prediction with a difference of 0 is used. Even when the compressed video data is decoded by selecting and using the B picture encoded data or the P picture encoded data of the forward prediction with the difference 0, the same image display can be realized. it can. Furthermore, by controlling the GOP configuration in accordance with the user's intention, it is possible to prevent an increase in the number of processed pictures until the screen is displayed at the time of random access reproduction, and it is possible to increase the degree of freedom during connection editing. .
The same encoding method can be applied to compression methods other than MPEG.
GOP means Group Of Pictures.
[0041]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of a moving picture coding apparatus according to the present invention will be described with reference to the drawings.
[0042]
<Moving picture encoding apparatus>
1 to 5 show a moving picture coding apparatus according to an embodiment of the present invention. FIG. 1 is a block diagram showing a system configuration including the moving picture coding apparatus. FIG. FIG. 3 is a schematic diagram showing an example of a moving image encoding operation timing for slow reproduction, FIG. 3 is a schematic diagram showing an example of a moving image encoding operation timing for 1 / 5-speed slow reproduction, and FIG. 4 is a 1 / 3-speed slow from normal reproduction. FIG. 5 is a schematic diagram showing an example of a moving image encoding operation timing for a moving image signal including a portion that shifts to reproduction. FIG. 5 is a moving image encoding for a moving image signal including a portion that shifts from 1 / 3-speed slow reproduction to normal reproduction. It is a schematic diagram which shows the example of operation timing.
[0043]
In FIG. 1, 1 is a moving image signal input terminal, 2 is a moving image signal storage buffer, 3 is a screen switching detection circuit, 4 is a difference 0 B and P picture encoded data generation unit, 5 is an encoding control unit, 6 is a DCT processing unit, 7 is a quantization unit, 8 and 9 are prediction memories “1” and “2” for storing reference images, and 10 is for inter-frame predictive encoding of B pictures. 1 is an inverse quantization unit, 12 is an inverse DCT processing unit, and 13 is a moving image data output terminal.
[0044]
2 to 5, (a) in each figure is an input moving image signal, (b) in each figure is encoded data in the encoding order when encoding is performed, and (c) in each figure. ) Is encoded data in display order at the time of decoding when encoding is performed. Also, numbers such as “0”, “1”, “2”, etc. in (a) of each figure are numbers that are updated each time the image is switched. That is, in (a) of each figure, the same number indicates that the images are the same. Also, I0, P1, B2, etc. in (b) and (c) in each figure identify the type of picture at the time of encoding, that is, the identification of the I picture, P picture, and B picture and the corresponding input image. The last F and B of B0F and B0B indicate the prediction direction of the B picture, that is, the forward direction (Forward) and the reverse direction (Backward).
[0045]
A method for encoding the moving image signal input from the moving image signal input terminal 1 in the moving image encoding apparatus configured as described above will be specifically described.
First, the moving image signal input from the moving image signal input terminal 1 is input to the moving image signal storage buffer 2 and the screen switching detection circuit 3. In the screen switching detection circuit 3, the input frame image is displayed immediately before. Whether it is the same as the frame image is detected.
[0046]
When the moving picture signal shown in FIG. 2 (a) is input, the first picture 0 is determined to be the same picture as the second picture 0, and the second picture 0 is the same picture as the third picture 0. Instead, it is counted that two B pictures for backward prediction with a difference of 0 are necessary.
It is determined that the third picture 0 is not the same image as the next picture 1, and is set as an I picture by the encoding control unit 5, and is encoded data I0 that has been subjected to encoding processing by the DCT processing unit 6 and the quantization unit 7. Is output to the moving image data output terminal 13, and the encoded data I 0 is decoded by the inverse quantization unit 11 and the inverse DCT processing unit 12 and stored in the prediction memory “1” 8.
[0047]
Thereafter, B0B, which is B picture encoded data, is read from the B and P picture encoded data generating unit 4 with a difference of 0 for the two times counted previously and output to the moving image data output terminal 13.
Subsequently, it is determined that the first picture 1 is the same as the second picture 1 and the second picture 1 is the same picture as the third picture 1, and the backward prediction of the difference 0 between the two pictures is not performed. A B picture is counted as needed.
[0048]
It is determined that the third picture 1 is not the same picture as the next picture 2, and is set as a P picture by the encoding control unit 5, and motion prediction compensation with the decoded data of I 0 stored in the prediction memory “1” 8 Encoded data P1 subjected to the encoding process in the DCT processing unit 6 and the quantizing unit 7 for the subsequent difference is output to the moving image data output terminal 13, and the encoded data P1 is reversed. Processing is performed by the quantization unit 11 and the inverse DCT processing unit 12, and the result is stored in the prediction memory “2” 9.
Thereafter, B1B, which is B picture encoded data, is read from the B and P picture encoded data generating unit 4 with a difference of 0 for the two times counted previously and output to the moving image data output terminal 13.
[0049]
Thereafter, the encoded data shown in FIG. 2 (b) can be output by repeating and encoding the processing after picture 2 in the same procedure.
Note that the GOP configuration of the encoded data shown in FIG. 2B is configured to circulate in 15 frames as an example, so picture 5 is encoded as an I picture.
[0050]
When the general MPEG decoding apparatus performs decoding processing on the encoded data shown in FIG. 2B generated as described above, the display order is as shown in FIG. In this case, since all B pictures such as B0B, B1B, and B2B are encoded data for backward prediction with a difference of 0, the same image can be displayed for three frame periods.
[0051]
Next, a GOP configuration setting method will be described.
The first method is a method for determining the upper limit of the number of pictures constituting the GOP, and the second method is a method for making the number of pictures constituting the GOP and the positional relationship between the I, P, and B pictures constant.
[0052]
First, the first method will be described in detail. As an example, the upper limit of the number of pictures constituting a GOP is 15 and the slow playback speed of the input video signal is 1/3 times speed, 1/4 times speed, 1/5 times speed, 1/6 times speed, 1/7 The device is capable of double speed. At this time, since the upper limit of the number of pictures is 15, a GOP configuration with 15 pictures at 1/3 times speed and 1/5 times speed, a GOP structure with 12 pictures at 1/4 times speed and 1/6 times speed, By using a GOP configuration with 14 pictures at / 7 × speed, the above-described encoding can be easily applied.
In this way, when the slow playback speed is known, the number of pictures constituting the GOP can be made variable below the set upper limit value according to the slow playback speed, thereby enabling random access and GOP unit linkage editing. An advantageous GOP configuration can be obtained.
[0053]
Next, the second method will be described in detail. As an example, the number of pictures constituting a GOP is 15, the period in which an I picture or P picture appears is 3, that is, the B picture inserted between them is 2, and the slow playback speed of the input video signal is 1 The device is capable of 3/3 speed and 1/5 speed. In 1/3 speed, if the encoding method used in this embodiment is applied as it is, the number of pictures constituting a GOP may be set to 15 and the period in which an I picture or P picture appears may be “3”. it can.
Next, the operation at 1/5 speed will be described with reference to FIG. At 1/5 times the speed, the cycle of the I picture or P picture and the switching of the screen are not synchronized, and thus a contrivance is required.
[0054]
When the moving picture signal shown in FIG. 3A is input, it is determined that the first picture 0 is the same picture as the second picture 0 immediately after, and the second picture 0 is the same picture as the third picture 0 immediately after. Then, it is counted that two B pictures for backward prediction with difference 0 are necessary without encoding.
The third picture 0 is the same image as the fourth picture 0. However, in order to comply with the GOP configuration, the encoding control unit 5 forcibly sets the I picture, and the DCT processing unit 6 and the quantization unit 7 The encoded data I0 having been subjected to the encoding process is output to the moving image data output terminal 13, and the encoded data I0 is decoded by the inverse quantization unit 11 and the inverse DCT processing unit 12. And stored in the prediction memory “1” 8.
[0055]
After that, B0B, which is the B picture encoded data of the backward prediction with the difference 0, is read out from the B and P picture encoded data generation unit 4 with the difference 0 for the counted twice, and is output to the moving image data output terminal 13. To do.
Subsequently, it is determined that the fourth picture 0 is the same as the immediately preceding third picture 0, and the fifth picture 0 is the same as the immediately preceding fourth picture 0. It counts that a forward-predicted B picture is needed.
[0056]
In order to comply with the GOP configuration, the next picture 1 is forcibly set to a P picture by the encoding control unit 5 and after motion prediction compensation with the decoded data of I0 stored in the prediction memory “1” 8. The encoded data P1 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 is output to the moving image data output terminal 13 and the inverse quantum is applied to the encoded data P1. The processing in the conversion unit 11 and the inverse DCT processing unit 12 is performed and stored in the prediction memory “2” 9.
[0057]
After that, B0F, which is the B picture encoded data for forward prediction with difference 0, is read out from the B and P picture encoded data generation unit 4 with difference 0 for the counted twice, and output to the moving picture data output terminal 13. To do.
Subsequently, it is determined that the second picture 1 is the same picture as the immediately preceding first picture 1 and the third picture 1 is the same as the immediately preceding second picture 1, and the difference between the two pictures is 0 without encoding. It counts that a forward-predicted B picture is needed.
[0058]
The fourth picture 1 is forcibly set to a P picture by the encoding control unit 5 in order to comply with the GOP configuration. However, since the fourth picture 1 is the same image as the first picture 1 encoded as the P picture immediately before, the predictive encoding is not performed, and the B and P picture encoded data generation unit 4 with the difference 0 is used. The P picture encoded data P1 ′ having a difference of 0 is read out and output to the moving image data output terminal 13.
[0059]
After that, B1F, which is B picture encoded data of forward prediction with a difference of 0, is read twice and output to the moving image data output terminal 13.
Subsequently, it is determined that the fifth picture 1 is the same image as the immediately preceding fourth picture 1, and it is counted that one B picture for forward prediction with a difference of 0 is necessary without encoding.
[0060]
The first picture 2 is determined to be the same image as the second picture 2 immediately after it, and it is counted that one B picture for backward prediction with a difference 0 is necessary without encoding.
The second picture 2 is compulsorily set to a P picture by the encoding control unit 5 in order to comply with the GOP configuration, and motion prediction compensation with the decoded data of P1 stored in the prediction memory “2” 9 Encoded data P2 obtained by performing encoding processing in the DCT processing unit 6 and the quantizing unit 7 for the subsequent difference is output to the moving image data output terminal 13, and the encoded data P1 is inverted. Processing is performed by the quantization unit 11 and the inverse DCT processing unit 12, and the result is stored in the prediction memory “1” 8.
[0061]
Thereafter, B1F, which is the forward prediction B picture encoded data, is counted once from the B and P picture encoded data generation unit 4 with the difference 0, and the backward prediction B picture encoded data with the difference 0 is counted once. A certain B2B is read out for one time counted previously and output to the moving image data output terminal 13.
Subsequently, it is determined that the third picture 2 is the same as the second picture 2 immediately before, and the fourth picture 2 is the same picture as the third picture 2 immediately before. It counts that a forward-predicted B picture is needed.
[0062]
The fifth picture 2 is forcibly set to a P picture by the encoding control unit 5 in order to comply with the GOP configuration. However, since the fifth picture 2 is the same image as the second picture 2 coded as the P picture immediately before, the predictive coding is not performed, and the B and P picture coded data generation unit 4 with the difference 0 is used. Then, the P picture encoded data P2 ′ of the forward prediction with the difference 0 is read and output to the moving image data output terminal 13.
Thereafter, B2F, which is B picture encoded data, is read twice as counted previously and is output to the moving image data output terminal 13.
[0063]
Encoding data shown in FIG. 3B can be generated by encoding in the above procedure. Also, when decoding processing is performed on the encoded data shown in FIG. 3B by a general MPEG decoding apparatus, the display order is as shown in FIG. In this case, all B pictures such as B0B, B0F, B1F, B2B, and B2F are coded data for forward prediction or backward prediction with a difference of 0, and P1 ′ and P2 ′ pictures are codes for forward prediction with a difference of 0. Since it is digitized data, the same image can be displayed for a period of 5 frames.
[0064]
In the present embodiment, the prediction direction is determined for B pictures included between I or P pictures of the same image, but the prediction direction may be any as long as the decoding results are the same. For example, the B picture between the P1 picture and the P1 ′ picture is the B1F picture of the forward prediction, but it is clear that the decoding result is the same even if it is the B1B picture of the backward prediction.
[0065]
As described above, even when the cycle of the I picture or P picture and the switching of the screen are not synchronized, if the input moving image signals are the same, the corresponding decoded images are the same. In addition, the number of pictures constituting the GOP and the positional relationship between the I, P, and B pictures can be encoded.
[0066]
Next, a method for encoding a moving image signal having a transition between normal reproduction and slow reproduction in accordance with the second method of the GOP configuration described above will be specifically described with reference to FIGS.
[0067]
In FIG. 4, (a) is a moving image signal including a portion that shifts from normal reproduction to 1 / 3-speed slow reproduction, and (b) is an encoding order when encoding is performed by the second method of the GOP configuration. Encoded data (c) is encoded data in the display order when encoding is performed by the second method of the GOP configuration.
In FIG. 5, (a) is a moving image signal including a portion that shifts from 1 / 3-speed slow reproduction to normal reproduction, and (b) is an encoding when encoding is performed by the second method of the GOP configuration. The encoded data in the order, (c) is the encoded data in the display order when encoding is performed by the second method of the GOP configuration.
[0068]
First, a moving image encoding method for a moving image signal including a portion that shifts from normal reproduction to 1 / 3-speed slow reproduction will be described.
When the moving image signal shown in FIG. 4A is input, since picture 0 is not the same picture as picture 1 immediately after and picture 1 is not the same picture as picture 2 immediately after, a code is used to comply with the GOP configuration. The B control is forcibly set to the B picture.
[0069]
In order to comply with the GOP configuration, the picture 2 is forcibly set to an I picture by the coding control unit 5, and the coded data I2 that has been subjected to coding processing by the DCT processing unit 6 and the quantization unit 7 In addition to being output to the image data output terminal 13, the inverse quantization unit 11 and the inverse DCT processing unit 12 perform decoding processing on the encoded data I <b> 2 and store it in the prediction memory “1” 8.
[0070]
Thereafter, with respect to the difference after motion prediction compensation between the picture 0 and the picture 1 set as the B picture and the decoded data of I2 stored in the prediction memory “1” 8, the DCT processing unit 6, the quantum The encoded data B 0 and B 1 subjected to the encoding process in the encoding unit 7 are output to the moving image data output terminal 13.
Subsequently, since the picture 3 is not the same picture as the immediately following picture 4 and the picture 4 is immediately following the picture 5, the encoding control unit 5 forcibly sets the picture to the B picture in order to comply with the GOP configuration.
[0071]
In order to comply with the GOP configuration, the picture 5 is forcibly set to a P picture by the encoding control unit 5 and the difference after motion prediction compensation with the decoded data of I2 stored in the prediction memory “1” 8 On the other hand, the encoded data P5 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 is output to the moving image data output terminal 13, and the inverse quantization unit is applied to the encoded data P5. 11. The inverse DCT processing unit 12 performs the decoding process and stores it in the prediction memory “2” 9.
After that, for the pictures 3 and 4 set as the B picture, the decoded data of I2 stored in the prediction memory “1” 8, the decoded data of P5 stored in the prediction memory “2” 9 and the average thereof The encoded data B3 and B4 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 are output to the moving image data output terminal 13 for the difference after the motion prediction compensation.
[0072]
The first picture 6 is judged to be the same as the second picture 6 immediately after it, and the second picture 6 is judged to be the same picture as the third picture 6 immediately after it. Are counted as necessary.
In order to comply with the GOP configuration, the third picture 6 is forcibly set as a P picture by the encoding control unit 5 and motion prediction compensation with the decoded data of P5 stored in the prediction memory “2” 9 Encoded data P6 subjected to encoding processing in the DCT processing unit 6 and the quantizing unit 7 for the subsequent difference is output to the moving image data output terminal 13, and the encoded data P6 is inverted. The quantization unit 11 and the inverse DCT processing unit 12 perform the decoding process and store the result in the prediction memory “1” 8.
[0073]
After that, B6B, which is the B picture encoded data for backward prediction with a difference of 0, is read twice from the B and P picture encoded data generation unit 4 with a difference of 0 and output to the moving image data output terminal 13. To do.
Subsequently, it is determined that the first picture 7 is the same as the second picture 7 immediately after it, and the second picture 7 is the same as the third picture 7 immediately after it. It is counted that a B picture for backward prediction is required.
[0074]
The third picture 7 is compulsorily set as a P picture by the encoding control unit 5 in order to comply with the GOP configuration, and motion prediction compensation with the decoded data of P6 stored in the prediction memory “1” 8 Encoded data P7 subjected to the encoding process in the DCT processing unit 6 and the quantizing unit 7 for the subsequent difference is output to the moving image data output terminal 13, and the encoded data P7 is reversed. Decoding processing is performed by the quantization unit 11 and the inverse DCT processing unit 12 and stored in the prediction memory “2” 9.
Thereafter, B7B, which is the B picture encoded data for backward prediction with a difference of 0, is read from the B and P picture encoded data generation unit 4 with a difference of 0 for the two times counted previously and output to the moving picture data output terminal 13. To do.
[0075]
Encoding data shown in FIG. 4B can be generated by encoding according to the above procedure. Further, when the general MPEG decoding apparatus performs decoding processing on the encoded data shown in FIG. 4B, the display order is as shown in FIG. 4C.
[0076]
Next, a description will be given of a moving image encoding means for a moving image signal including a portion that shifts from 1 / 3-speed slow reproduction to normal reproduction.
When the moving picture signal shown in FIG. 5A is input, the first picture 0 is the same as the second picture 0 immediately after, and the second picture 0 is the same picture as the third picture 0 immediately after. Judgment is made and it is counted that two B pictures for backward prediction with difference 0 are necessary without encoding.
[0077]
In order to comply with the GOP configuration, the third picture 0 is forcibly set to an I picture by the encoding control unit 5 and encoded data I0 subjected to encoding processing by the DCT processing unit 6 and the quantization unit 7 Is output to the moving image data output terminal 13, and the encoded data I 0 is decoded by the inverse quantization unit 11 and the inverse DCT processing unit 12 and stored in the prediction memory “1” 8.
[0078]
After that, B0B, which is the B picture encoded data of the backward prediction with the difference 0, is read out from the B and P picture encoded data generation unit 4 with the difference 0 for the counted twice, and is output to the moving image data output terminal 13. To do.
Subsequently, it is determined that the first picture 1 is the same as the second picture 1 immediately after and the second picture 1 is the same image as the third picture 1 immediately after, and the difference between the two pictures is 0 without encoding. It is counted that a B picture for backward prediction is required.
[0079]
In order to comply with the GOP configuration, the third picture 1 is forcibly set to a P picture by the encoding control unit 5 and motion prediction compensation with the decoded data of I0 stored in the prediction memory “1” 8 Encoded data P1 subjected to the encoding process in the DCT processing unit 6 and the quantizing unit 7 for the subsequent difference is output to the moving image data output terminal 13, and the encoded data P1 is reversed. Decoding processing is performed by the quantization unit 11 and the inverse DCT processing unit 12 and stored in the prediction memory “2” 9.
[0080]
Thereafter, B1B, which is the B picture encoded data for backward prediction of difference 0, is read from the B and P picture encoded data generation unit 4 with difference 0 for the counted twice, and output to the moving picture data output terminal 13. To do.
Subsequently, it is determined that the first picture 2 is the same as the second picture 2 immediately after it, and the second picture 2 is the same image as the third picture 2 immediately after it. It is counted that a B picture for backward prediction is required.
[0081]
The third picture 2 is compulsorily set to a P picture by the encoding control unit 5 in order to comply with the GOP configuration, and motion prediction compensation with the decoded data of P1 stored in the prediction memory “2” 9 Encoded data P2 obtained by performing encoding processing in the DCT processing unit 6 and the quantizing unit 7 on the subsequent difference is output to the moving image data output terminal 13, and the encoded data P2 is reversed. The quantization unit 11 and the inverse DCT processing unit 12 perform the decoding process and store the result in the prediction memory “1” 8.
After that, B2B, which is the B picture encoded data for backward prediction of difference 0, is read twice from the B and P picture encoded data generation unit 4 with difference 0 and output to the moving image data output terminal 13. To do.
[0082]
Since picture 3 is not the same picture as immediately following picture 4 and picture 4 is immediately following picture 5, in order to comply with the GOP configuration, encoding control unit 5 forcibly sets the picture to B picture.
In order to comply with the GOP configuration, the picture 5 is forcibly set as a P picture by the encoding control unit 5 and the difference after motion prediction compensation with the decoded data of P2 stored in the prediction memory “1” 8 On the other hand, the encoded data P5 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 is output to the moving image data output terminal 13, and the inverse quantization unit is applied to the encoded data P5. 11. The inverse DCT processing unit 12 performs the decoding process and stores it in the prediction memory “2” 9.
[0083]
Thereafter, with respect to pictures 3 and 4 set as B pictures, the decoded data stored in the prediction memory “1” 8, the decoded data of P5 stored in the prediction memory “2” 9 and their average motion Encoded data B3 and B4 obtained by performing encoding processing in the DCT processing unit 6 and the quantization unit 7 on the difference after the prediction compensation are output to the moving image data output terminal 13.
Subsequently, since the picture 6 is not the same picture as the immediately following picture 7 and the picture 7 is not the same picture as the immediately following picture 8, the encoding control unit 5 forcibly sets the picture to the B picture in order to comply with the GOP configuration.
[0084]
In order to comply with the GOP configuration, the picture 8 is forcibly set to a P picture by the encoding control unit 5 and the difference after motion prediction compensation with the decoded data of P5 stored in the prediction memory “2” 9 On the other hand, the encoded data P8 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 is output to the moving image data output terminal 13, and the inverse quantization unit is applied to the encoded data P8. 11. The inverse DCT processing unit 12 performs the decoding process and stores it in the prediction memory “1” 8.
[0085]
Thereafter, for the pictures 6 and 7 set as the B picture, the decoded data of P8 stored in the prediction memory “1” 8, the decoded data of P5 stored in the prediction memory “2” 9 and the average thereof, The encoded data B6 and B7 subjected to the encoding process in the DCT processing unit 6 and the quantization unit 7 are output to the moving image data output terminal 13 for the difference after the motion prediction compensation.
[0086]
Encoding data shown in FIG. 5B can be generated by encoding according to the above procedure. Further, when the general MPEG decoding apparatus performs decoding processing on the encoded data shown in FIG. 5B, the display order is as shown in FIG. 5C.
In this way, even when the playback mode changes in the middle of the GOP structure, it can be applied to the moving picture encoding apparatus according to the present embodiment.
[0087]
<Moving image recording / playback device>
Next, a moving picture recording / reproducing apparatus using the moving picture coding apparatus according to the present invention will be described with reference to FIGS.
6 and 7 show a moving picture recording / reproducing apparatus using the moving picture encoding apparatus according to the present invention. FIG. 6 shows an example applied to the slow reproduction of the disc recording / reproducing apparatus, and FIG. A configuration example in the case where an input signal is decoded, MPEG-encoded again, and recorded and reproduced is shown.
[0088]
In FIG. 6, 21 is a moving image signal input terminal, 22 is a moving image input signal selector, 23 is an MPEG encoding unit described in the above embodiment, 24 is an MPEG decoding unit, and 25 is a recording / playback signal process for media. , 26 is an IEEE 1394 terminal for transmitting and receiving encoded data, 27 is a recording medium, 28 is a system controller that controls the entire recording / reproducing apparatus, and 29 is a moving image data output terminal.
In FIG. 7, the same blocks as those in FIG. 6 are denoted by the same reference numerals. Therefore, only the added blocks will be described. 30 decodes a DV compressed signal input via the IEEE 1394 terminal 26. A DV decoding unit 31 is a moving image output signal selector.
[0089]
In the moving image recording / reproducing apparatus configured as described above, a method of recording a moving image signal input from the moving image signal input terminal 21, slow reproduction of encoded data recorded on the recording medium 27, and MPEG encoding again. A method of outputting from the IEEE 1394 terminal 26 and a method of decoding the DV compressed signal input from the IEEE 1394 terminal 26, re-encoding it to MPEG, and recording it on the recording medium 27 will be specifically described.
The DV compressed signal is a compressed signal defined by “HD DIGITAL VCR CONFERENCE” and mainly used in a consumer digital VCR.
[0090]
First, a method for recording a moving image signal input from the moving image signal input terminal 21 will be described assuming that the recording medium 27 is a DVD.
The moving image signal input from the moving image signal input terminal 21 is supplied to the MPEG encoding unit 23 via the moving image input signal selector 22. The MPEG encoding unit 23 encodes the input moving image signal in the MPEG_PS format, which is a DVD recording format, and supplies it to the recording / playback signal processing unit 25. The recording / reproducing signal processing unit 25 performs error correction encoding and digital modulation processing for adding a parity for correcting an error that may occur at the time of reproduction on the encoded data, and records it on the recording medium 27. To do.
[0091]
Next, a description will be given of a method of performing MPEG encoding again and outputting from the IEEE 1394 terminal 26 when the encoded data recorded on the recording medium 27 is played back slowly.
The signal reproduced from the recording medium 27 is digitally demodulated by the recording / reproduction signal processing unit 25, and if there is an error during reproduction, correction processing is performed to reconstruct the original encoded data. The encoded data reproduced in this way is supplied to the MPEG decoding unit 24 based on an instruction from the system controller 28, and an MPEG decoding process is performed to restore a moving image signal.
[0092]
At the time of slow playback at 1/3 speed, the MPEG decoding unit 24 outputs the same moving image signal three times, thereby outputting the moving image signal as shown in FIG. The timing is transmitted to the system controller 28. The moving image signal is output from the moving image data output terminal 29 to an external device, and is also supplied to the MPEG encoding unit 23 via the moving image input signal selector 22.
The MPEG encoding unit 23 detects the screen switching by referring to the screen switching timing supplied from the system controller 28, and uses the method described in the above-described embodiment to input the moving image signal. Is encoded with MPEG to generate encoded data as shown in FIG.
[0093]
Since the MPEG stream format that can be input / output at the IEEE 1394 terminal 26 is defined in the MPEG_TS format, the MPEG stream is converted into the MPEG_TS format and output from the IEEE 1394 terminal 26 here. In this way, the MPEG_TS format signal output from the IEEE 1394 terminal 26 is received and decoded by a receiving apparatus having the IEEE 1394 terminal 26 (not shown), so that the moving picture recording / reproducing apparatus according to the present embodiment performs slow playback. The moving image signal obtained in this way can also be reproduced on the receiving device side.
In this embodiment, the example in which the data encoded by the MPEG encoding unit 23 is output via the IEEE 1394 terminal 26 is used. However, the data is recorded on another recording medium 27 or another recording medium (not shown). May be.
[0094]
<Recording method to recording media>
Next, a method of decoding a DV compressed signal input from the IEEE 1394 terminal 26, re-encoding it to MPEG, and recording it on the recording medium 27 will be described.
[0095]
A DV compressed signal input from an unshown DV recording / reproducing apparatus via the IEEE 1394 terminal 26 is supplied to the DV decoding unit 30.
The DV decoding unit 30 performs a decoding process on the DV compressed signal to restore the moving image signal. Here, when the DV recording / reproducing apparatus performs 1 / 3-speed slow reproduction, the DV compressed signal is different from MPEG and uses only intra-frame coding. Therefore, the same DV compressed signal is used every three frames. Is entered. Furthermore, since the additional information called VAUX data sent at the same time includes an identifier indicating whether or not the current frame FC (Frame Change Flag) is the same as the previous frame, the DV decoding unit 30 Then, by decoding and outputting the input DV compressed signal as it is, a moving image signal as shown in FIG. 2A is output, and the switching timing of the screen obtained by referring to the identifier FC is set in the system. Tell the controller 28.
[0096]
The moving image signal is output from the moving image data output terminal 29 to the external device via the moving image output signal selector 31 and is also supplied to the MPEG encoding unit 23 via the moving image input signal selector 22. .
The MPEG encoding section 23 refers to the screen switching timing supplied from the system controller 28 to detect the screen switching, MPEG-encodes the moving image signal using the above-described method, and the result shown in FIG. Coded data as shown is generated.
The recording / reproducing signal processing unit 25 performs error correction coding for adding a parity for correcting an error that may occur during reproduction to the coded data, and performs digital modulation processing on the recording medium 27. Record.
[0097]
In this way, even when the DV recording / reproducing apparatus performs slow reproduction, the input DV compressed signal can be decoded via the IEEE 1394 terminal 26, MPEG-encoded again, and recorded on the recording medium 27. it can.
In this embodiment, the example in which the data encoded by the MPEG encoding unit 23 is recorded on the recording medium 27 is used. However, the data may be output via the IEEE1394 terminal 26.
[0098]
As described above, by applying the moving image encoding apparatus according to the present invention to a moving image recording / reproducing apparatus, MPEG encoding can be efficiently performed when the same image is input during slow reproduction or still reproduction. The screen switching detection circuit inside the moving picture coding apparatus can be omitted by performing the screen switching detection by an instruction from the system controller with reference to the internal playback mode and additional information supplied from the outside. can do.
[0099]
【The invention's effect】
The moving picture coding apparatus according to the present invention performs coding by appropriately combining coded data of a B picture with a difference of 0 and normal coded data in coding when the same image signal is input in slow reproduction or the like. Generate data.
Therefore, when the encoded data is decoded, images that were the same image before encoding can be decoded as exactly the same image.
[0100]
In addition, the moving picture coding apparatus according to the present invention makes the number of pictures constituting a GOP variable below a set upper limit value, and appropriately changes the coded data of P or B pictures with a difference of 0 and normal coded data as appropriate. Combined to generate encoded data.
In addition, the moving picture coding apparatus according to the present invention sets the number of pictures constituting the GOP and the period in which the I picture or P picture appears, and selects the I picture, the P picture, and the B picture to match the setting. In addition, the number of pictures constituting the GOP is made variable below the set upper limit value, and the encoded data is generated by appropriately combining the encoded data of the P or B picture with the difference 0 and the normal encoded data.
Therefore, when the encoded data is decoded, the same image before encoding can be decoded as the same image, and the GOP configuration is advantageous for random access and GOP unit edit. be able to.
[0101]
In addition, the moving image encoding apparatus according to the present invention compares the moving image signal stored in the buffer memory for encoding by the moving image encoding means with the input moving image signal, immediately before or after. It is detected whether it is the same as the moving image signal of the frame.
Therefore, the encoding method described above can be applied to any input signal.
[0102]
In addition, the moving picture coding apparatus according to the present invention refers to switching information of a moving picture signal supplied from the outside, and whether the inputted moving picture signal is the same as the moving picture signal of the immediately preceding or following frame. Detect whether or not.
Therefore, the processing in the image switching detection unit can be reduced.
[0103]
The moving picture recording / reproducing apparatus according to the present invention refers to the moving picture signal switching information generated by the moving picture decoding means, and the moving picture of the frame immediately before or after the moving picture signal input to the moving picture coding apparatus is Whether the signal is the same is detected.
The moving picture recording / reproducing apparatus according to the present invention refers to moving picture signal switching information supplied from an external device, and the moving picture signal input to the moving picture coding apparatus is a moving picture of a frame immediately before or after the moving picture signal. Whether the signal is the same is detected.
Therefore, the processing in the image switching detection unit can be reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a system configuration including a moving image encoding apparatus according to an embodiment of the present invention.
FIG. 2 is a schematic diagram illustrating an example of a moving image encoding operation timing for 1 / 3-speed slow playback.
FIG. 3 is a schematic diagram illustrating an example of a moving image encoding operation timing for 1 / 5-speed slow playback.
FIG. 4 is a schematic diagram illustrating an example of a video encoding operation timing with respect to a video signal including a portion that shifts from normal playback to 1 / 3-speed slow playback.
FIG. 5 is a schematic diagram illustrating an example of a moving image encoding operation timing for a moving image signal including a portion that shifts from 1 / 3-speed slow reproduction to normal reproduction.
FIG. 6 is a block diagram showing a configuration of a moving image recording / playback apparatus according to an embodiment of the present invention.
FIG. 7 is a block diagram showing a configuration of a moving image recording / reproducing apparatus that decodes a DV input signal, re-encodes the MPEG signal, and records / reproduces the encoded MPEG signal.
FIG. 8 is a block diagram illustrating a configuration example of a system including a conventional moving image re-encoding device.
FIG. 9 is a block diagram illustrating a configuration example of a video encoder in a conventional moving image re-encoding device.
FIG. 10 is a schematic diagram showing the operation timing of a conventional moving image re-encoding device.
[Explanation of symbols]
1 Video signal input terminal
2 Video signal storage buffer
3 Screen switching detection circuit
4 B and P picture encoded data generation unit with difference 0
5 Coding control unit
6 DCT processor
7 Quantization part
8 Prediction memory “1”
9 Prediction memory “2”
10 Averaging circuit
11 Inverse quantization section
12 Inverse DCT processing unit
13 Video data output terminal
21 Video signal input terminal
22 Video input signal selector
23 MPEG encoding section
24 MPEG decoding unit
25 Recording / playback signal processing section
26 IEEE1394 terminal
27 Recording media
28 System controller
29 Video signal output terminal
30 DV decoder
31 Video output signal selector
110 Data separator
111 Audio data buffer memory
112 Video data buffer memory
113 Sub-picture data buffer memory
114 Navigation data buffer memory
115 Audio decoder
116 Video decoder
117 Sub-picture decoder
118 Video overlay processor
119 Adder
120 GUI image signal generator
121 audio encoder
122 video encoder
123 frame memory
124 Re-encoding unit audio data buffer memory
125 Re-encoding unit moving image data buffer memory
126 Multiplexer
127 IEEE 1394 interface
128 Re-encoding unit
129 D / A converter for audio signal
130 D / A converter for video signal
131 Data memory
150 Optical disk device
151 voice data
152 video data
153 Sub-picture data
154 Navigation data
155 Audio signal
156 Video signal
157 Sub-video signal
158 video signal
159 GUI image signal
160 User operations
161, 162 Analog signal
164 encoding parameters
165 Motion vector information
170 Motion compensation unit
172 Motion compensation mode and DCT type determination unit
173 Rate control unit
174 DCT processor
175 Quantizer
176 Inverse quantization unit
177 Inverse DCT processing unit
178 Variable length coding unit
180 Generated code amount

Claims (7)

In a moving image encoding device for encoding a moving image signal,
Image switching detection means for detecting whether or not the moving image signal to be encoded is the same as the moving image signal of the immediately preceding or subsequent frame;
A moving image encoding means for encoding a moving image signal to generate moving image encoded data;
Difference 0 encoded data generating means for generating encoded data of a B picture of difference 0;
Encoded data selection means for selecting encoded data output from the moving image encoding means and encoded data output from the difference 0 encoded data generation means;
With
The encoded data selection means, when the moving image signal to be encoded is the same as the image of the immediately following frame, is a B picture code for backward prediction of difference 0 output from the difference 0 encoded data generation means. A moving picture coding apparatus that selects coding data and selects coding data output from the moving picture coding means when a moving picture signal to be coded is different from an image of the immediately following frame. .   2. The upper limit of the number of pictures constituting the GOP is set, and the number of pictures constituting the GOP is made variable below the set upper limit according to the slow playback speed of the input moving image signal. The moving image encoding apparatus described. In a moving image encoding device for encoding a moving image signal,
Image switching detection means for detecting whether or not the moving image signal to be encoded is the same as the moving image signal of the immediately preceding or subsequent frame;
A moving image encoding means for encoding a moving image signal to generate moving image encoded data;
Difference 0 encoded data generation means for generating encoded data of B and P pictures of difference 0;
Encoded data selection means for selecting encoded data output from the moving image encoding means and encoded data output from the difference 0 encoded data generation means;
With
The encoded data selection means outputs the difference 0 encoded data generation means when the moving image signal to be encoded is the same as the image of the immediately preceding or immediately following frame and is encoded as a B picture or P picture. Selected from the B picture encoded data of the backward prediction of the difference 0, the B picture encoded data of the forward prediction of the difference 0, and the P picture encoded data of the forward prediction of the difference 0. when the moving image signal that is different from the image immediately following frame, the video encoding apparatus and selects the coded data output from the video encoding means.   The image switching detection unit compares the moving image signal stored in the buffer memory for encoding by the moving image encoding unit with the input moving image signal, thereby moving the moving image of the immediately previous or next frame. 4. The moving picture coding apparatus according to claim 1, wherein it is detected whether or not the signal is the same.   The image switching detection means detects whether or not the input moving image signal is the same as the moving image signal of the immediately preceding or immediately following frame with reference to switching information of the moving image signal supplied from the outside. The moving picture coding apparatus according to claim 1 or 3, characterized in that: In a moving image recording / reproducing apparatus for recording / reproducing encoded data obtained by encoding a moving image signal,
A video encoding device according to claim 1 or 3,
A recording medium for recording and reproducing moving image encoded data, and a moving image decoding means for decoding the encoded data,
When performing slow playback or still playback of the signal recorded on the recording medium, whether or not the video signal decoded by the video decoding means is the same as the video signal of the immediately preceding or subsequent frame Switching information is generated, and it is detected by the image switching detection means in the moving image encoding device whether or not it is the same as the moving image signal of the immediately preceding or immediately following frame with reference to the switching information. A moving image recording / reproducing apparatus. In a moving image recording / reproducing apparatus for recording / reproducing encoded data obtained by encoding a moving image signal,
A video encoding device according to claim 1 or 3,
A recording medium for recording and reproducing moving image encoded data;
Moving image decoding means for decoding moving image encoded data supplied from an external device,
When an external device performs slow playback or still playback, the moving image signal decoded based on the screen switching information added to the moving image encoded data by the moving image decoding means is the moving image of the frame immediately before or after Switching information as to whether or not it is the same as the image signal is generated, and the same as the moving image signal of the frame immediately before or after referring to the switching information by the image switching detecting means in the moving image encoding device A moving image recording / reproducing apparatus for detecting whether or not the moving image is recorded.

Images