JP4148200B2 - MPEG image data recording apparatus and MPEG image data recording method - Google Patents

Description

  The present invention relates to an MPEG image data recording apparatus and an MPEG image data recording method, and in particular, two MPEG image data obtained by encoding with an MPEG (Moving Picture Experts Group) encoding method are designated in each MPEG image data. The present invention relates to an MPEG image data recording apparatus and an MPEG image data recording method for performing recording for realizing seamless reproduction when reproducing at a connection designated position.

  First, an MPEG encoding system which is a premise of the present invention will be briefly described. As for the MPEG encoding method, ISO / IEC 11172-2, ITU-T recommendation H.264, etc. 262, ISO / IEC 13818-2 has been described in detail, and only the outline will be described here. MPEG is the name of an organization that examines video coding standards established in 1988 by ISO / IEC JTC1 / SC2 (International Organization for Standardization / International Electrotechnical Standards Meeting Technical Committee 1 / Technical Committee 2, now SC29). Abbreviation for Moving Picture Experts Group.

  MPEG1 (MPEG Phase 1) is a standard for storage media of about 1.5 Mbps, for low-transfer rates of JPEG and ISDN (Service Integrated Digital Network) video conferences and videophones for the purpose of still image coding. Inheriting the basic technology of H.261 (CCITT SGXV, standardized by the current IU-TSG15) for the purpose of moving image compression, a new technology for storage media is introduced. These were established in August 1993 as ISO / TEC 11172. MPEG1 is created by combining several technologies.

  FIG. 9 shows a conventional MPEG encoder that performs encoding by the MPEG encoding method, which will be briefly described below. The input image signal is first subjected to the difference between the image signal decoded by the motion compensated predictor 1 and the differencer 2 to reduce the time redundant portion. There are three modes of prediction from the past and the future. These can be switched and used for each 16 pixels × 16 pixels MB (Macro block). The prediction direction is determined by the picture type given to the input image.

  There are two modes, a mode in which encoding is performed based on prediction from the past and a mode in which the MB is independently encoded without prediction, and a one-way inter-picture predictive encoded image (P picture). In addition, there are four modes of bi-directional inter-picture prediction: a mode that encodes by prediction from the future, a mode that encodes by prediction from the past, a mode that encodes by prediction from both, and a mode that encodes independently. It is an encoded image (B picture). In addition, it is an intra-image independent encoded image (I picture) that all MBs are independently encoded.

  The motion compensation in the motion compensation predictor 1 is performed by performing pattern matching on the motion region for each MB between the input image signal and the reference image signal from the image memory 11 to detect a motion vector with half-pel accuracy, and only for the motion. Predict after shifting. The motion vector has a horizontal direction and a vertical direction, and is transmitted as additional information of the MB together with an MC (Motion Compensation) mode indicating where the motion vector is predicted.

  In general, the above three pictures are transmitted in combination, but from the I picture to the picture before the next I picture is called GOP (Group Of Picture). Generally, about 15 pictures are used as one GOP section (however, two or more I pictures may be included in one GOP section. In short, one or more I pictures may be included in one GOP section). .

  The differential image signal output from the differentiator 2 is subjected to discrete cosine transform (DCT) in the DCT unit 3 to obtain DCT coefficients. The discrete cosine transform (DCT) is an orthogonal transform that discretely transforms an integral transform with a cosine function as an integral kernel into a finite space. In MPEG, MB is divided into four, and two-dimensional DCT is performed on a DCT block of 8 × 8 pixels. In general, since an image signal has many low-frequency components and few high-frequency components, when DCT is performed, the coefficients are concentrated in the low frequency.

  The image data (DCT coefficient) subjected to DCT is quantized by the quantizer 4. Quantization is a quantized value obtained by multiplying a value obtained by weighting the 8 × 8 two-dimensional frequency called the quantization matrix with a visual characteristic and a value called a quantization scale that is a scalar multiplication of the whole, and the DCT coefficient is quantized. This is done by dividing by value. When inverse quantization is performed by the MPEG decoder (decoder), a value approximate to the original DCT coefficient is obtained by multiplying by the quantized value.

  The data quantized by the quantizer 4 is converted into a variable length code (VLC) by the VLC unit 5. In the VLC unit 5, the direct current (DC) component of the quantized value uses differential pulse code modulation (DPCM) which is one of predictive coding. In addition, the alternating current (AC) component is zigzag scanned from low to high, with zero run length and effective coefficient value as one event, and Huffman code that assigns a code with a short code length from the highest appearance probability Is done.

  The data variable-length encoded by the VLC unit 5 is temporarily stored in the buffer 6 and then output as encoded data at a predetermined transfer rate. Further, the generated code amount for each MB of the output encoded data is supplied to the code amount controller 7, where it is set as an error code amount with respect to the generated code amount with respect to the target code amount, and then the quantizer 4. The amount of code is controlled by adjusting the quantization scale.

  The image data quantized by the quantizer 4 is inversely quantized by the inverse quantizer 8 and further inversely DCTed by the inverse DCT device 9 and then stored in the image memory 11 via the adder 10. It is stored temporarily. The motion compensation predictor 1 uses the signal stored in the image memory 11 as a reference decoded image (reference image) for performing motion compensation with the input image signal.

  FIG. 10 is a block diagram showing an example of a decoder that decodes encoded data that has been MPEG-encoded by the encoding apparatus of FIG. In FIG. 10, encoded data (stream) is buffered by the buffer 13 and then input to the VLD unit 14. The VLD unit 14 performs variable length decoding to obtain a direct current (DC) component and an alternating current (AC) component. The alternating current (AC) component data is arranged in an 8 × 8 matrix in the zigzag scan order from the low range to the high range.

  The data output from the VLD unit 14 is input to the inverse quantizer 15 and is inversely quantized by the quantization matrix. The inversely quantized data is input to the inverse DCT unit 16 and subjected to inverse DCT, and is added to the signal from the motion compensation predictor 18 by the adder 17 to be output as image data (decoded data). Further, after the decoded data is temporarily stored in the image memory 19, a reference decoded image (reference image) for calculating a difference image from the data from the VLD unit 14 in the motion compensation predictor 18. ) Is used.

  Here, the bit stream of encoded data output from the buffer 13 has a variable amount of code for each picture in the case of video. This is because MPEG uses information conversion such as DCT, quantization, and Huffman coding, and at the same time, the amount of code allocated to each picture needs to be adaptively changed to improve image quality. This is because the entropy of the encoded image itself changes greatly, such as encoding the input image as it is in some cases and encoding a difference image that is the difference between the predicted images in some cases.

  In this case, in many cases, the code amount is controlled while allocating the entropy ratio of the image and keeping the limit of the storage capacity (buffer capacity) of the buffer 6. The code amount controller 7 monitors the relationship between the generated code amount and the encoding rate, and sets the target code amount so as to be within a predetermined buffer capacity. This value is fed back to the VLC unit 5 and enters the code amount controller 7, where the quantization value set in the quantizer 4 is increased to suppress the generated code amount or the quantized value is decreased to generate the generated code amount. Or make it smaller.

  When encoding such variable length data at a fixed transfer rate (encoding rate), if the maximum buffer amount of the decoder is set as the upper limit value, the data is input at a constant speed and only a predetermined value is accumulated. Therefore, it is MPEG to use a model in which decoding is performed at a predetermined time (1 / 29.97 sec unit for NTSC video signals) and the buffer is encoded so that neither overflow nor underflow occurs. It is prescribed.

  If this rule (VBV buffer rule) is observed, the rate in the VBV buffer changes locally, but if the observation time is increased, the transfer rate becomes a fixed transfer rate. In MPEG, this is a fixed transfer rate. It is defined as being.

  In the case of a fixed transfer rate, when the amount of generated code is small, the buffer occupation amount is stuck to the upper limit value. In this case, it is necessary to increase the code amount so as not to overflow by adding an invalid bit. In the case of variable transfer rate, the definition of this fixed transfer rate has been expanded so that when the buffer occupancy reaches the upper limit value, in principle, the decoder reading is stopped so that overflow does not occur. Has been.

  Such a buffer transition is shown in FIG. In FIG. 11, the vertical axis represents the buffer occupation amount, and the horizontal axis represents the display time. As shown in the figure, in MPEG encoding, even if the amount of generated code is very small, reading of the decoder is stopped, so there is no need to insert invalid bits as in the case of a fixed transfer rate. Therefore, encoding is performed so that only underflow does not occur.

  The MPEG system stipulates a method of multiplexing a bit stream encoded with MPEG video and audio into one bit stream and reproducing it while ensuring synchronization. The contents specified in the system are roughly divided into the following five points.

1) Synchronous playback of multiple encoded bitstreams 2) Multiplexing of multiple encoded bitstreams into a single bitstream 3) Buffer initialization at the start of playback 4) Continuous buffer management 5) Determination of time for decoding, reproduction, etc. Information must be packetized for multiplexing in the MPEG system. For example, when multiplexing video data and audio data, each packet is divided into streams of appropriate lengths called packets, attached with additional information such as headers, and video and audio packets as appropriate. Is a time-division transmission method. The header includes information for identifying video, audio, etc., and time information for synchronization. The packet length depends on the transmission medium and application, and ranges from 53 bytes as in the asynchronous transfer mode (ATM) to as long as 4 kbytes as in the optical disk. In MPEG, the packet length is variable and can be arbitrarily specified.

  Data is packed and packetized, and one pack is composed of several packets. A pack start code and a system time reference reference value (SCR: System Clock Reference) are described at the top of each pack, and a stream ID and a time stamp are described at the top of the packet. ing. The time stamp describes time information for synchronizing audio, video, etc., and there are two types, DTS (Decoding Time Stamp) and PTS (Presentation Time Stamp).

  DTS is decoding time management information, and the decoding start time of the first access unit (1 picture for video, 1152 samples for audio) in the packet data, PTS is time management information for reproduction output, and is displayed (reproduction). The start time is shown. These time stamps are given a time reference by SCR or PCR (Program Clock Reference). The PCR is described with a time accuracy of 27 MHz, and is information (time reference reference value) for locking the reference clock of the decoder.

  FIG. 12 shows a block diagram of an example of a conventional MPEG multiplexing system. As shown in the figure, the system decoder 21 looks at the audio and video identifiers described in the header of the multiplexing unit pack in the input multiplexed stream, and the audio pack data is sent to the audio decoder 22. The video pack data is separated and output to the video decoder 23.

  At the same time, the system decoder 21 receives information of SCR (system clock reference), DTS (decoding time stamp), and PST (presentation time stamp) described in the header of the pack in the input multiplexed stream. , SCR is output to the clock controller 24, and DTS and PTS are output to the comparator 25. The clock controller 24 generates a 27 MHz clock by synchronizing the input SCR.

  On the other hand, the audio decoder 22 removes the pack header from the input audio pack data, decodes it to audio elementary data, and stores it in the temporary audio memory 26. Further, the video decoder 23 removes the pack header from the input video pack data, converts it to video elementary data, performs decoding, and stores it in the temporary video memory 27.

  The comparator 25 compares the clock information from the clock with the input time stamp such as PTS, and outputs a signal to be decoded or presented to the decoder or the memories 26 and 27 when they coincide with each other to match the DTS. The audio and video buffer control and AV synchronization control are performed by decoding and presenting in accordance with the PTS. As described above, the audio decoder 22, the video decoder 23, and the like are configured to always monitor a common reference clock locked by PCR and to perform decoding and display when the time matches with the time of DTS or PTS.

  Here, the multiplexed data is buffered by each decoder, and a virtual decoder (audio decoder 22 and video decoder 23 in FIG. 12) for performing synchronized display is called an STD (System Target Decoder). The STD must be multiplexed to prevent overflow or underflow.

  The MPEG system is roughly divided into TS (Transport Stream) and PS (Program Stream). These are composed of packets including PES (Packetized Elementary Stream) and other necessary information. PES is defined as an intermediate stream for enabling conversion between both streams, and is a packet of a private stream or the like in addition to video and audio data encoded by MPEG.

  As shown in FIG. 13E, the PS is a continuous stream in units of variable-length blocks called PS packs (or simply packs), and the video and audio of a program having a common reference time are multiplexed. Is possible. As shown in FIG. 13D, the PS pack is composed of a pack header and a plurality of PESs. As shown in FIG. 13C, the structure of the PES is composed of a PES header and a PES payload, which are used in common with a TS to be described later, and make these compatible. In the PS STD model, a stream is switched by a stream ID (stream id) in a PES packet.

  TS can multiplex video and audio of programs having a common reference time as well as PS, but TS can also multiplex multi-programs such as communication and broadcasting having different reference times. It is said. The TS is composed of a continuous stream of TS packets having a fixed length of 188 bytes as shown in FIG. 13A in consideration of the ATM cell length and error correction coding. Considered to be usable.

  As shown in FIG. 13B, the structure of the TS packet itself is composed of one TS header and one TS payload, and is not so complicated. However, since the TS is a multi-program stream, its operation is complicated. It is. Characteristic compared to PS is that although the TS packet with the structure shown in FIG. 13B is a higher structure, it is shorter (usually) than the PES packet shown in FIG. This is the point of transmission on the TS payload of the TS packet shown in FIG. In the STD model of TS, the stream is switched by the PID (packet ID) in the TS packet.

  The MPEG system TS has a mechanism for instructing which packet identification information (PID: Packet Identification) the packet relating to the information of the multiplexed program is. This will be described with reference to FIG. First, the TS packet group is searched for PID = 0. A TS packet 31 with PID = 0 is an information packet called a PAT (Program Association Table), and in the packet, information PID corresponding to the PROGRAM number PR is described in a linked form as indicated by 32. Yes.

  Next, when a PID packet 33 (here, TS packet with PID = X) corresponding to the target PR is read, it is called a PMT (Program Map Table) as indicated by 34. There is an information packet, and the PMT describes information on the PID of the video packet of the program corresponding to the PR and the PID of the audio packet. Thus, the TS packet 35 of PID = PV is received, the TS packet 36 of PID = PC is received, and the TS packet 37 of PID = PA is received, thereby obtaining the video, clock, and audio signals of the target program. be able to. The above PAT and PMT are called PSI (Program Specific Information), which is an information system that can access (entry) the channel of the target program.

  When two MPEG image data obtained by encoding with the MPEG encoding method described above are simply connected and recorded on a recording medium for editing, a contradiction arises in the connection of the VBV buffer described above and an overflow occurs. And underflow occurs.

  In order to maintain the continuity of the MPEG image data in order to perform recording for seamless reproduction when the two MPEG image data are reproduced at the designated connection designated positions, the two MPEG image data are maintained. Image data recording apparatus that performs encoding in consideration of continuity, such as controlling the amount of generated code so that the VBV buffer is always fixed at the edit point of the connection designated position of GOP or encoding GOP as closed GOP Is conventionally known (see, for example, Patent Document 1).

  In addition, the encoded data is not subject to any restrictions, and in the partial section of the data, information indicating the data extracted as editing material and information regarding the reproduction order are described, and the recorded data is changed. In addition, an image data recording apparatus that can realize video editing on a single recording medium is known (for example, see Patent Document 2).

JP-A-11-74799 JP-A-11-187354

  However, in the conventional image data recording apparatus described in Patent Document 1, the generated code amount is controlled so that the VBV buffer is always fixed for each GOP so that the GOP is a closed GOP so that the GOP can be edited anywhere. It is necessary to apply an encoding constraint considering continuity such as encoding, which is a disadvantageous factor in terms of encoding efficiency.

  On the other hand, the conventional image data recording apparatus described in Patent Document 2 reproduces and displays as if edited, but the continuity at the editing point is incomplete, initialization of the decoder buffer for MPEG data, etc. There may be a temporary quiescence phenomenon.

  The present invention has been made in view of the above points. The first and second MPEG image data (or the packet-multiplexed first image data including the first MPEG image data as element encoded data) is provided. MPEG multiplexed data and packet-multiplexed second MPEG multiplexed data including second MPEG image data as element encoded data) at each designated connection designation position, the first MPEG When the image data is connected to the second MPEG image data for reproduction (when the first MPEG multiplexed data is connected to the second MPEG multiplexed data for reproduction), the VBV buffer connection overflows MPEG image data recording apparatus and MP capable of realizing seamless and high-quality reproduction without causing inconsistency of underflow and underflow And to provide the G image data recording method.

To achieve the above object, an MPEG image data recording apparatus according to a first aspect of the present invention converts first and second MPEG image data, which are image data encoded by the MPEG encoding method, into respective MPEG images. The first joint section re-encoding encoded by the MPEG encoding method as data for reproduction from the first MPEG image data to the second MPEG image data at the designated joint designation position in the data An MPEG image data recording apparatus provided with recording means for generating and recording data,
The first MPEG image data has at least the first time management information indicating the decoding start time of the access unit, which is a unit of decoding and reproduction, and the decoding time reference is the same as that of the encoding side in accordance with the MPEG encoding standard. The first time reference information for determination and the first transfer rate information of the multiplexed data are multiplexed in time series and stored in the VBV buffer which is a virtual buffer on the decoding side. The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the time management information.
The second MPEG image data has at least the second time management information indicating the decoding start time of the access unit, which is a unit of decoding and reproduction, and the decoding side time reference is the same as the encoding side in accordance with the MPEG encoding standard. The second time reference information for determination and the second transfer rate information of the multiplexed data are multiplexed in time series and stored in the VBV buffer which is a virtual buffer on the decoding side. The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the time management information.
The recording means sets a section having a start position as a predetermined time before the connection designation position in the first MPEG image data as a start position and a section having the connection designation position in the first MPEG image data as an end position as a first connection section. Recoded image data, which is image data obtained by decoding the first MPEG image data corresponding to one connection section, is re-encoded by the MPEG encoding method to generate first connection section re-encoded data. An encoding means, and recording the first linked section re-encoded data on a recording medium;
The re-encoding means is
The value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the two recorded before and after the recording position of the first time management information The difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of the data lengths between the recording position of the first time reference information and the recording position of the first time management information, First predicted value calculating means for multiplying the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value;
The value of the second time management information of the first frame after the connection designated position in the second MPEG image data and two second time references recorded before and after the recording position of the second time management information The difference between the interpolation value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the information and the recording position of the second time management information is the second transfer rate. Second predicted value calculation means for multiplying the value of the information to obtain a second predicted value of the VBV buffer occupation value;
The decoded image data, which is the image data obtained by decoding the first MPEG image data corresponding to the first connection interval, is changed from the first predicted value to the second predicted value when the VBV buffer occupancy value is started from the first predicted value. And encoding means for performing encoding while performing code amount control so as to end after transition.

In order to achieve the above object, the MPEG image data recording apparatus of the second invention is replaced with the recording means and the re-encoding means of the MPEG image data recording apparatus of the first invention. Corresponds to the second connection section, with the connection specified position in the MPEG image data as a start position and a section having a position after a predetermined time from the connection specified position in the second MPEG image data as an end position as a second connection section. Re-encoding means for re-encoding the decoded image data, which is the image data obtained by decoding the second MPEG image data, using the MPEG encoding method to generate second linked section re-encoded data, Second re-encoded data is recorded on a recording medium,
The re-encoding means is recorded before and after the value of the first time management information of the first frame after the connection designation position in the second MPEG image data and the recording position of the first time management information. Difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the first time reference information and the recording position of the first time management information First predicted value calculation means for multiplying the value of the first transfer rate information to obtain the first predicted value of the VBV buffer occupation value, and after the end position of the second connection section in the second MPEG image data The second time management information value of the first frame, the recording position of the second time reference information recorded before and after the recording position of the second time management information, and the second time management information By interpolation based on the ratio of each data length to the recording position of Second predicted value calculation means for obtaining a second predicted value of the VBV buffer occupation value by multiplying the difference between the calculated second time management information and the interpolated value by the value of the second transfer rate information; The decoded image data, which is the image data obtained by decoding the second MPEG image data corresponding to the second connection section, is changed from the first predicted value to the second predicted value when the VBV buffer occupancy value is started from the first predicted value. And encoding means for performing encoding while performing code amount control so as to end after transition.

In order to achieve the above object, the MPEG image data recording apparatus of the third invention is replaced with the recording means and the re-encoding means of the first and second inventions. A section having a start position at a first predetermined time before the connection specified position in the MPEG image data as a start position, a section having the connection specified position in the first MPEG image data as an end position is set as a first connection section, and the second MPEG image The first and second joints are defined as a second joint section in which a joint designated position in the data is a start position and a section having a position after a second predetermined time from the joint designated position in the second MPEG image data is an end position. The first and second decoded image data, which are image data obtained by decoding the first and second MPEG image data corresponding to the section, are re-encoded by the MPEG encoding method, and the first and second connection images are connected. Comprising a re-encoding means for generating a segment re-encoded data, which is recorded in the first and second connecting sections re-encoded data recording medium,
The re-encoding means includes the value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data, and before and after the recording position of the first time management information. Interpolation of the first time management information obtained by interpolation based on the ratio of each data length between the recording positions of the two recorded first time reference information and the recording position of the first time management information A first predicted value calculating means for obtaining a first predicted value of the VBV buffer occupancy value by multiplying the first difference with the value by the value of the first transfer rate information; and a second predicted value in the second MPEG image data. The value of the second time management information of the first frame after the end position of the two connecting sections and the two second time reference information recorded before and after the recording position of the second time management information Ratio of each data length between the position and the recording position of the second time management information The second difference from the interpolation value of the second time management information obtained by interpolation based on the value is multiplied by the value of the second transfer rate information to obtain the second predicted value of the VBV buffer occupation value. 2 prediction value calculation means, and first decoded image data which is image data obtained by decoding the first MPEG image data corresponding to the first connection interval and the second corresponding to the second connection interval. The second decoded image data, which is image data obtained by decoding the MPEG image data, sequentially, so that the VBV buffer occupancy value starts from the first predicted value, transitions to the second predicted value, and ends. And an encoding means for performing encoding while performing code amount control.

In order to achieve the above object, an MPEG image data recording method according to a fourth aspect of the present invention relates to first and second MPEG image data, which are image data encoded by the MPEG encoding method, respectively. As the data to be played back by connecting from the first MPEG image data to the second MPEG image data at the specified connection specified position in the MPEG image data, the first connection section recoded by the MPEG encoding method is reproduced. An MPEG image data recording method provided with a recording step for generating and recording encoded data,
The first MPEG image data has at least the first time management information indicating the decoding start time of the access unit, which is a unit of decoding and reproduction, and the decoding time reference is the same as that of the encoding side in accordance with the MPEG encoding standard. The first time reference information for determination and the first transfer rate information of the multiplexed data are multiplexed in time series and stored in the VBV buffer which is a virtual buffer on the decoding side. The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the time management information.
The second MPEG image data has at least the second time management information indicating the decoding start time of the access unit, which is a unit of decoding and reproduction, and the decoding side time reference is the same as the encoding side in accordance with the MPEG encoding standard. The second time reference information for determination and the second transfer rate information of the multiplexed data are multiplexed in time series and stored in the VBV buffer which is a virtual buffer on the decoding side. The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the time management information.
In the recording step, a section having a predetermined time before the connection designation position in the first MPEG image data as a start position and a section having the connection designation position in the first MPEG image data as an end position is defined as a first connection section. Recoded image data, which is image data obtained by decoding the first MPEG image data corresponding to one connection section, is re-encoded by the MPEG encoding method to generate first connection section re-encoded data. An encoding step, and recording the first linked section re-encoded data on a recording medium;
The re-encoding step is
The value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the two recorded before and after the recording position of the first time management information The difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of the data lengths between the recording position of the first time reference information and the recording position of the first time management information, A first predicted value calculating step of multiplying the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value;
The value of the second time management information of the first frame after the connection designated position in the second MPEG image data and two second time references recorded before and after the recording position of the second time management information The difference between the interpolation value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the information and the recording position of the second time management information is the second transfer rate. A second predicted value calculating step of multiplying the value of the information to obtain a second predicted value of the VBV buffer occupation value;
The decoded image data, which is the image data obtained by decoding the first MPEG image data corresponding to the first connection interval, is changed from the first predicted value to the second predicted value when the VBV buffer occupancy value is started from the first predicted value. And an encoding step of performing encoding while performing code amount control so as to end after transition.

In order to achieve the above object, the MPEG image data recording method of the fifth invention is replaced with the recording step and the re-encoding step of the MPEG image data recording method of the fourth invention. Corresponds to the second connection section, with the connection specified position in the MPEG image data as a start position and a section having a position after a predetermined time from the connection specified position in the second MPEG image data as an end position as a second connection section. A re-encoding step for re-encoding the decoded image data, which is image data obtained by decoding the second MPEG image data, by the MPEG encoding method to generate second re-encoded data. Second re-encoded data is recorded on a recording medium,
The re-encoding step 2 is recorded before and after the value of the first time management information of the first frame after the connection designation position in the second MPEG image data and the recording position of the first time management information. Difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the first time reference information and the recording position of the first time management information A first predicted value calculating step of multiplying the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value, and after the end position of the second joint section in the second MPEG image data The second time management information value of the first frame, the recording position of the second time reference information recorded before and after the recording position of the second time management information, and the second time management information Based on the ratio of each data length to the recording position of A second predicted value calculation step of obtaining a second predicted value of the VBV buffer occupation value by multiplying the difference from the interpolated value of the second time management information obtained by the insertion by the value of the second transfer rate information Then, the decoded image data, which is the image data obtained by decoding the second MPEG image data corresponding to the second connection interval, is obtained from the first prediction value when the VBV buffer occupancy value is started from the first prediction value. And an encoding step of performing encoding while performing code amount control so as to end after transitioning to a value.

Furthermore, in order to achieve the above object, the MPEG image data recording method of the sixth invention is replaced with the recording step and the re-encoding step of the fourth and fifth inventions, and the recording step is the first MPEG image. In the second MPEG image data, a section having a start position at a first predetermined time before the connection specified position in the data as a start position, a section having the connection specified position in the first MPEG image data as an end position is set as the first connection section. A section having a connection designated position as a start position and an end position at a position after a second predetermined time from the connection designated position in the second MPEG image data is defined as a second connection section. The first and second decoded image data, which are image data obtained by decoding the corresponding first and second MPEG image data, are re-encoded by the MPEG encoding method, and the first and second MPEG image data are decoded. Comprising a re-encoding step of generating a tether segment re-encoded data, which is recorded in the first and second connecting sections re-encoded data recording medium,
The re-encoding step is
The value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the two recorded before and after the recording position of the first time management information The first of the interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the first time reference information and the recording position of the first time management information A first predicted value calculating step of multiplying the difference by the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value, and the end of the second connection section in the second MPEG image data The value of the second time management information of the first frame after the position, the recording position of the second time reference information recorded before and after the recording position of the second time management information, and the second time Based on the ratio of each data length to the management information recording position The second predicted value for obtaining the second predicted value of the VBV buffer occupancy value by multiplying the second difference from the interpolated value of the second time management information obtained by the above by the value of the second transfer rate information The calculating step, the first decoded image data which is the image data obtained by decoding the first MPEG image data corresponding to the first connection section, and the second MPEG image data corresponding to the second connection section are obtained. The code amount control is performed so that the VBV buffer occupancy value starts from the first predicted value, transitions to the second predicted value, and ends with the second decoded image data that is the image data obtained by decoding. And an encoding step of encoding while performing the above.

  In the MPEG image data recording apparatus and the recording method of the present invention, the encoded image data of the connecting section consisting of either one or both of the first connecting section and the second connecting section is decoded, and When the decoded image data is re-encoded with the MPEG encoding method, the value of the first time management information of the first frame after the start position of the first joint section or after the joint designation position, and the first time First obtained by interpolation based on the ratio of each data length between the recording position of the two first time reference information and the recording position of the first time management information recorded before and after the recording position of the management information The first predicted value of the VBV buffer occupancy value is obtained by multiplying the difference from the interpolated value of 1 time management information by the first transfer rate information, and after the start position of the second connection section or the connection designation position Second time management information of the first frame after The ratio of each data length between the value and the recording position of the second time management information and the recording position of the second time management information recorded before and after the recording position of the second time management information The second predicted value of the VBV buffer occupancy value is obtained by multiplying the difference from the interpolation value of the second time management information obtained by the interpolation based on the second transfer rate information to obtain the first and second values. One or both of the decoded image data are re-encoded while controlling the amount of code so that the VBV buffer occupancy value starts from the first predicted value and transitions to the second predicted value and ends. Since the section re-encoded data is generated and recorded, the first MPEG image data is reproduced up to the start position of the connection section, the recorded connection section re-encoded data is reproduced, and then the connection is performed. The second MPEG image data from the end position of the section By reproducing, there is no inconsistency of overflow or underflow in the connection of the VBV buffer, and the original first and second MPEG image data itself is not processed, and the first designated position first The MPEG image data can be reproduced by being connected to the second MPEG image data at the second designated position.

  The first MPEG image data and the second MPEG image data of the present invention are packet-multiplexed with other data such as audio encoded data encoded by the MPEG encoding method, and element encoding is performed. It is assumed that the encoded image data in the MPEG multiplexed data constituting the data is included.

  According to the present invention, the first MPEG image data is reproduced up to the start position of the joining section, the recorded joining section re-encoded data is reproduced, and then the second MPEG from the end position of the joining section. By reproducing the image data, there is no inconsistency of overflow or underflow in the connection of the VBV buffer, and the original first and second MPEG image data itself is not processed, and the first designated position is not changed. Since the first MPEG image data can be connected and reproduced from the first MPEG image data to the second MPEG image data at the second specified position, the first and second specified positions respectively specified as the connected positions Playback that connects MPEG image data to second MPEG image data can be realized seamlessly and with high quality.

  Further, in the present invention, in order to realize seamless and high-quality reproduction, the image data to be newly generated / recorded may be only the re-encoded data of the set connection section. Can be used efficiently.

  Next, the best mode for carrying out the present invention will be described with reference to the drawings.

  First, the concept of joint reproduction that can be realized by the present invention will be described. As shown in FIG. 1, when there are first MPEG image data and second MPEG image data which are image data encoded by the MPEG encoding method, the first MPEG image data is designated (specified). Let us consider connecting and reproducing the second MPEG image data from the designated connection position.

  Assuming that the connection designation position (connection point) in the first MPEG image data is the position b in FIG. 1, reproduction is performed up to the position b in the first MPEG image data, and then connection is made to the second MPEG image data. However, if the two MPEG image data are simply connected, there is a problem in that the connection of the VBV buffer is inconsistent and overflow or underflow occurs.

  Therefore, conventionally, the occupied value of the VBV buffer has been detected by two methods. In the MPEG standard, the first detection method is basically a CBR (constant bit rate), that is, in the case of a fixed transfer rate, the VBV delay indicating the buffer occupancy rate of the VBV buffer of the decoder in the MPEG video picture layer. A value is specified, and this is the method of using it. That is, in the first detection method, since the VBV delay value is described in units of pictures, the MPEG code of a picture corresponding to a predetermined position of the MPEG image data can be obtained by observing only the picture header of the MPEG image bit stream. It is possible to detect the VBV buffer occupancy value at the start (or end) of the conversion.

In this case, for example, the function of the VBV buffer information predictor 105 in FIG. 2, which will be described later, searches for a bit stream picture header (a 4-byte code of 0x00000100 in MPEG) and then a 10-bit temporal reference (temporal reference). ), And a 16-bit VBV delay following a 3-bit picture coding type indicating the picture type. What is VBV delay value? VBV delay = 90000xB / R
B: Buffer occupancy value R: Indicates the initial state of the buffer at the time of random access defined by the bit rate. In the case of a fixed transfer rate, this bit rate is used to occupy the VBV buffer from the VBV delay value. The value can also be obtained by calculation (for example, it is calculated by the VBV buffer information predictor 105 shown in FIG. 2). In MPEG, it is necessary to pay attention to the VBV delay value at the vertex position of the graph as shown in FIG. However, this method is only effective for CBR.

  The second detection method is not a fixed transfer rate but a variable transfer rate. In the MPEG standard, basically, in the case of VBR (variable bit rate), the VBV delay value indicating the buffer occupancy of the VBV buffer in the picture layer of MPEG video is all 1 (0xffff because all 16 bits are 1). It becomes. Therefore, the first detection method cannot be used.

  Therefore, in the second detection method, the following calculation is performed from information obtained by observing the MPEG image bit stream. First, it is assumed that the VBV buffer is occupied from the very beginning of the bitstream up to the maximum value defined by MPEG (for example, 1.75 Mbit at the main profile main level). Next, the code amount of the first picture is subtracted. Next, using the transmission rate information of the peak rate in variable transfer rate encoding, the transmission amount when the time between display pictures has elapsed is added, and the code amount of the next picture is subtracted.

  By repeating such a series of addition and subtraction processes up to a predetermined bitstream position, a picture corresponding to a predetermined position in the MPEG image data is obtained by simulating a graph as shown in FIG. The VBV buffer occupancy value at the MPEG encoding start time (or end time) can be obtained.

  The peak rate at the time of the variable transfer rate is defined in MPEG so as to be described in a syntax portion called a bit rate of the sequence header, and can be obtained by referring to it. For example, the VBV buffer information predictor 105 shown in FIG. 2 is realized by providing the above-mentioned MPEG image bit stream with an observation function and a calculation function (in the case of this second detection method, the head of the MPEG image bit stream is From this, the generated code amount of each picture must be observed and calculated.) However, this second detection method must analyze the bitstream from the beginning, and cannot perform an editing operation or the like at high speed.

  Therefore, in the present invention, the VBV buffer occupancy value information is predicted based on the information of the difference between the DTS and PCR at the connection designated position in the MPEG image data to be re-encoded and the transmission rate of the compressed data, and the VBV buffer occupancy information is Code amount control is performed so that the predicted value changes, and re-encoding is performed.

  This method is described with the process of re-encoding using FIG. The information value related to the VBV buffer occupancy value is, for example, a VBV buffer occupancy value or a VBV delay value defined in MPEG. As will be described later, one unit of a predetermined section for performing re-encoding necessary when two streams are connected may be, for example, about 3 frames or about 1 GOP. Temporarily, the predetermined interval unit for the re-encoding is set between ab in the first MPEG image data shown in FIG. 1 and between cd in FIG. 1 in the second MPEG image data. To do.

  The first MPEG image data between a and b (first connecting section: here, section A) is once decoded to obtain decoded image data, and the decoded image data (first connecting section). (Decoded image data) is re-encoded by the MPEG encoding method. The re-encoded data created thereby is referred to as interval A re-encoded MPEG image data (first connected interval re-encoded data).

  This re-encoding is started from the information value related to the VBV buffer occupancy value obtained from the transition of the information value related to the VBV buffer occupancy value based on the first VBV buffer occupancy value related information detected at the position a, The rate control is performed so that the processing ends up to the information value related to the VBV buffer occupation value obtained based on the detected second VBV buffer occupation value related information at the position d, and re-encoding is performed by the MPEG encoding method.

  In the present embodiment, instead of the conventional splice reproduction operation in which reproduction is performed up to the position of b of the first MPEG image data and then data is reproduced by connecting to the position of d of the second MPEG image data. The following operation is performed using the re-encoded MPEG image data in the section A. That is, in the present embodiment, the first MPEG image data is reproduced up to the position a (start position of section A) of the first MPEG image data, and then the re-encoded MPEG image data of section A is reproduced. Is reproduced from the start position to the end position of the section, and then connected to the position d (connection designated position) of the second MPEG image data to reproduce the data of the second MPEG image data after d. Make it work.

  This is realized by providing a connecting reproduction means for performing this operation on the reproduction apparatus side. By performing this joint playback operation, it is possible to realize seamless connection playback that has the same content as the original first and second MPEG image data and that does not cause a VBV buffer failure.

  Next, a connection reproduction example according to another embodiment of the present invention will be described with reference to FIGS. FIG. 5A is the same as the example shown in FIG. That is, in FIG. 5A, the reproduction of the first MPEG image data is performed up to the position e of the first MPEG image data (the start position of the section A), and then the ef of the first MPEG image data. The re-encoded MPEG image data of the section A is reproduced from the start position to the end position of the section A, and then connected to the position h (connection designated position) of the second MPEG image data, and the second after the h. MPEG image data is reproduced.

  In the example shown in FIG. 1 and FIG. 5 (A), the re-encoded MPEG image data of the connecting section is obtained based only on the first MPEG image data side, and the connected reproduction is realized. It is also possible to realize joint reproduction by obtaining re-encoded MPEG image data of a joint section based only on the side. An example thereof is shown in FIG. 5B, which uses the same encoded MPEG image data in the connection section B in the second MPEG image data.

  That is, in FIG. 5B, the connection point (connection designation position) in the first MPEG image data is set to the position i, and the connection point (connection designation position) in the second MPEG image data is set to the position k. One unit of the first and second predetermined sections, which are units having the information of the VBV buffer described above, is between ij in the first MPEG image data and between kl in the second MPEG image data. To do. The position l is the boundary position of the second predetermined section after the second predetermined time from the connection point (connection designated position) k in the second MPEG image data.

  A section having the connection point k as the start position and the position l as the end position is defined as a connection section B (second connection section). The second MPEG image data is connected to the data in the section B (k-1 section), and then decoded image data is obtained, and the decoded image data (second connection section decoded image data) is MPEG-encoded. Re-encoding is performed using this method. The re-encoded data created thereby will be referred to as re-encoded MPEG image data (second connected interval re-encoded data) in the connection interval B.

  In this re-encoding, the transition of the information value related to the VBV buffer occupancy value starts from the information value related to the VBV buffer occupancy value at the position i and ends up to the information value related to the VBV buffer occupancy value at the position l. Thus, rate control is performed and re-encoding is performed using the MPEG encoding method. Then, using the re-encoded MPEG image data in the connection section B, connection reproduction from the first MPEG image data to the second MPEG image data is performed. That is, the first MPEG image data is reproduced up to the connection point i (connection designated position) of the first MPEG image data, and then the re-encoded MPEG image data of the section B is ended from the start position of the section. The reproduction is performed up to the position, and then the second MPEG image data after the l position is reproduced by connecting to the l position of the second MPEG image data (end position of the connecting section B).

  This is realized by providing a connecting reproduction means for performing this operation on the reproduction apparatus side. By performing this joint playback operation, it is possible to realize seamless connection playback that has the same content as the original first and second MPEG image data and that does not cause a VBV buffer failure.

  Next, an embodiment shown in FIG. 5C will be described. In this example, re-encoded MPEG image data in a connection section is obtained based on the first and second MPEG image data before and after the connection point, and connection reproduction is realized. In FIG. 5C, a connection point (connection designation position) in the first MPEG image data is set as n, and a connection point (connection designation position) in the second MPEG image data is set as p.

  The position of m is the boundary position of the first predetermined section before the first predetermined time from the connection point (connection designated position) n in the first MPEG image data. A section having the position of m as the start position and the position of the connection point n as the end position is defined as a connection section A (first connection section). The position of q is the boundary position of the second predetermined section after a second predetermined time from the connection point (connection designated position) p in the second MPEG image data. A section having the connection point p as the start position and the position of q as the end position is defined as a connection section B (second connection section).

  The data of the connection section A (mn section) of the first MPEG image data is once decoded to obtain decoded image data (connection section A decoded image data: first connection section decoded image data). Also, the data of the connection section B (pq section) of the second MPEG image data is once decoded to obtain decoded image data (connection section B decoded image data: second connection section decoded image data). . The joint section A + B decoded image data (third joint section decoded image data) obtained by combining the joint section A decoded image data and the joint section B decoded image data is re-encoded by the MPEG encoding method. The re-encoded data created thereby is referred to as re-encoded MPEG image data (third re-encoded data of the third connection section) in the connection section A + B.

  In this re-encoding, the transition of the information value related to the VBV buffer occupancy value starts from the information value related to the first VBV buffer occupancy value at the position m, and the second VBV buffer occupancy value at the position q. Rate control is performed so that the process ends with the information value, and re-encoding is performed using the MPEG encoding method. Then, using the re-encoded MPEG image data in the connection section A + B, connection reproduction from the first MPEG image data to the second MPEG image data is performed.

  That is, the first MPEG image data is reproduced until the start position m of the first MPEG image data connection section A, and then the re-encoded MPEG image data of the connection section A + B is ended from the start position of the section. The reproduction is performed up to the position, and then the second MPEG image data after the position of q is reproduced by connecting to the end position q of the connection section B of the second MPEG image data. This is realized by providing a connecting reproduction means for performing this operation on the reproduction apparatus side.

  By performing this joint playback operation, it is possible to realize seamless connection playback that has the same content as the original first and second MPEG image data and that does not cause a VBV buffer failure.

  As described above, any of the methods shown in FIGS. 5A to 5C can realize seamless and high-quality joint reproduction. As examples of these developments, as shown in FIG. 6, the first MPEG image data and the second MPEG image data are connected and reproduced, the first MPEG image data, the second MPEG image data, and the fourth MPEG image data are connected. It is also possible to realize joint playback that branches in the middle, such as joint playback in which MPEG image data are sequentially connected, and joint playback in which first MPEG image data and third MPEG image data are connected. .

  As described above, by using the present invention, it is possible to freely connect to other MPEG data only by generating and using re-encoded MPEG image data in the connecting section without processing the original MPEG image data itself. Therefore, even when a program constituting various branch stories is encoded, it is possible to efficiently use the media without recording redundant MPEG image data over the entire story for each branch story ( If the present invention is used, a set of each of the first to fourth MPEG image data and the encoded MPEG image data of the connecting section for each branch story may be prepared.)

  Next, the configuration of an embodiment of an MPEG image data recording apparatus according to the present invention will be described. FIG. 2 shows a block diagram of an embodiment of an MPEG image data recording apparatus according to the present invention. This embodiment is applied to a digital signal recording / reproducing apparatus.

  First, an operation for creating VBV buffer information while encoding image data by the MPEG encoding method will be described. In the figure, when there is no encoded data in the recording medium 101, that is, when encoding is performed for the first time, the data reading unit 102 does not include the encoded data recorded on the recording medium 101. Information that there is no data is output to the VBV buffer information predictor 105.

  When the VBV buffer information predictor 105 receives the information that there is no encoded data, the initial value in which the parameters (VBV value, PTM time stamp information) output by the parameter setter 106 are set in advance, that is, the VBV value is MPEG, for example. The value is 80% of the maximum value of the specified VBV, and the PTM timestamp information is 0. Initial setting values of these parameters are supplied from the parameter setting unit 106 to the image re-encoder 107.

  In the image re-encoder 107, when the signal input from the encoded data search unit 103 is a bit stream, the data is decoded and re-encoded, and when it is image data, new encoding is performed from the initial setting value. Start. That is, the image re-encoder 107 is composed of a decoder having the same configuration as that shown in FIG. 10 and an MPEG encoder having the same configuration as that shown in FIG. 9, and at the time of re-encoding or encoding, VBV The initial value is set from the parameter setting unit 106 to the code amount controller (corresponding to 7 in FIG. 9) in the image re-encoder 107, and the VBV buffer is controlled by the code amount controller according to the definition of the MPEG VBV buffer. Sets a target code amount for each frame so that underflow does not occur, and performs a quantization step on the quantizer (corresponding to 4 in FIG. 9) in the image re-encoder 107 according to the target code amount. Is set, and variable length coding is performed by a VLC unit (corresponding to 5 in FIG. 9).

  In this way, the image re-encoder 107 supplies the obtained encoded data to the data writing unit 108 while performing re-encoding or encoding, thereby recording it on the recording medium 101.

  Next, the first and second MPEG image data recorded on the recording medium 101 are re-encoded in the section A in FIG. 1 in order to enable continuous reproduction from a predetermined position. 2 will be described with reference to FIGS. 1 and 7 together.

  First, from the user interface (not shown), the encoded data indicates from which point of the first and second MPEG image data (compressed encoded stream) already recorded on the recording medium 101 to be reproduced. The search unit 103 or the like is designated. The data reading unit 102 supplies the MPEG image data read from the recording medium 101 on which the first and second MPEG image data are already recorded to the encoded data search unit 103.

  The encoded data search unit 103 searches a frame several frames before the start position of the designated re-encoding section A based on the reproduced MPEG image data supplied from the data reading unit 102.

  The PCR / DTS detector 104 detects the DTS in the first frame of the re-encoding section A in the first MPEG image data output from the encoded data search unit 103. Further, the PCR values before and after the DTS data are detected. The positional relationship between DTS and PCR is shown in FIG. Since DTS and PCR are recorded in conformity with the MPEG standard, detection can be performed by reading a recording bit according to the syntax of the MPEG standard. The PCR of the DTS point (recording position) can be calculated by interpolation using the data length ratio from the previously detected PCR value before and after the DTS to the point where the DTS is recorded.

  That is, FIG. 7A shows that the DTS information in the first frame of the re-encoding section A is recorded on the recording medium 101 at the position 41 between two adjacent PCR positions PA and PB of the compressed stream. In the MPEG standard, the interval between the PCR at the PA position and the next PCR position at the PB position is defined as 100 msec or less.

  Here, the value of the top PCR of the frame in which the DTS is described is represented by the data length 42 from the top of the PCR information at the position PA to the top of the DTS, and from the top of the PCR information at the position PB to the top of the DTS. Using the ratio of the data length 43 of the above, it is calculated by interpolation as shown at 44.

  Subsequently, the VBV buffer information predictor 105 calculates the difference Δ between the DTS value at the position 41 and the PCR interpolation value at the position indicated by 44 and records it in the header of the system layer of this MPEG data. After the detected TS_mux_rate value is detected, the estimated value of the VBV buffer occupation value is obtained by multiplying them.

  Here, the above TS_mux_rate indicates the transfer rate of the multiplexed data of the MPEG transport stream (TS), and is defined by MPEG so as to be transmitted at a rate of once every 100 to 120 msec. . On the other hand, the PCR describes the clock reference value measured when transferring multiplexed data at the timing when the data is generated, and indicates the reference time that is not related to the contents of the data. Time reference information.

  Further, as described above, DTS is time management information indicating the decoding start time of an access unit, which is a unit of decoding and reproduction, and the VBV buffer is instantaneously stored in the decoder at the time indicated by the DTS (DTS value). The MPEG image data is encoded so that the image data and the like are read and transferred. The DTS is described for each access unit (each picture in the case of video). If a picture always has the same generated code amount (= fixed rate), the difference Δ between the DTS value and the PCR interpolation value is Since it is always fixed, it is sufficient to store the code amount for the fixed time in the buffer.

  However, since the generated code amount for each picture is actually variable, the above difference Δ is variable. Therefore, as shown in FIG. 7B, the product of the above difference Δ and the value of TS_mux_rate. Is substantially equivalent to the occupied value of the VBV buffer of the picture in which the DTS is described. That is, the code amount corresponding to the number of frames (number of pictures) reproduced in the section of the difference Δ from the time indicated by the actual DTS value to the time of the data position of the DTS indicated by the PCR interpolation value is the difference Δ and the difference Δ described above. It substantially corresponds to the product value α with the value of TS_mux_rate, and can be regarded as being stored in the VBV buffer.

  The product value α is exactly the occupancy value of the buffer that stores the entire TS. The occupancy value of the picture VBV buffer depends on the audio data multiplexed in the TS and the multiplexing packing. The occupation value is obtained by adding the amount of overhead data such as a header. However, since the latter data amount is small, the product value α can be regarded as the occupation value of the VBV buffer in practice.

  The VBV buffer information predictor 105 uses the same method as described above, and the DTS value of the first frame after the connection designated position (d in the example of FIG. 1) in the second MPEG image data corresponding to the end position of the connection section. And VBV corresponding to the end position of the connecting section by calculating the difference Δ between the PCR interpolation value obtained from the PCR in the second MPEG image data before and after that and multiplying the value of the TS_mux_rate. Get the predicted value of the buffer occupancy value.

  In this way, the VBV buffer prediction value (first prediction value) of the first frame after the start position of the connection section obtained by the VBV buffer information predictor 105 and the connection designation position of the second MPEG image data. The predicted value (second predicted value) of the VBV buffer of the first frame after is input to the parameter setting unit 106.

  On the other hand, the encoded data search unit 103 searches the bit stream already recorded on the recording medium 101 for the start position of the connection section A to be re-encoded. The search uses the relative address of the data and sets a pointer at the position from the beginning of the bitstream file. The image re-encoder 107 decodes the first MPEG image data corresponding to the connection section A, and the transition of the VBV buffer occupancy value is again transmitted from the parameter setter 106 using the obtained decoded image data. The rate control is performed so as to start from the predicted value of the VBV buffer occupation value at the connection start position a in FIG. 1 and end up to the predicted value of the VBV buffer occupation value at the end position d of the connection interval. Then, re-encoding is performed by the MPEG encoding method.

  The re-encoding may use an image that has been completely decoded as described above. For example, a technique for controlling the amount of code on a bit stream as disclosed in Japanese Patent Laid-Open No. 11-234777 is used. Also good.

  The image data of section A (joined section A re-encoded data) re-encoded by the image re-encoder 107 is separated from the first and second MPEG image data on the recording medium 101 by the data writing unit 108. It is recorded as a separate file, or connected to the head of the recorded second MPEG image data.

  In the above description, the first MPEG image data, the second MPEG image data, the re-encoded image data of the section A (or the re-encoded image data of the section B, or the re-encoded image of the section A + B). (Encoded image data) may be recorded on the same recording medium 101, may be recorded on a plurality of recording media in any combination, or may be recorded on different recording media. In the case of being recorded separately on a plurality of recording media, it indicates that they are the same information group so that each recording medium is linked and operated (each data and information is linked). Information such as an ID may be recorded on each recording medium.

  The first MPEG image data, the second MPEG image data, and the re-encoded image data of the section A (or the re-encoded image data of the section B or the re-encoded image data of the section A + B). When recorded on the same recording medium, it is possible to control splice playback on the playback apparatus side with a single recording medium.

  In the above embodiment, the recording medium is described as a recording medium in the recording / reproducing apparatus. However, the recording medium may be a recording medium detachably attached to the recording / reproducing apparatus or a recording medium (database) via a network.

  In the above embodiment, an example in which image data is connected by focusing on single MPEG image data has been described. Next, MPEG transformers, which are MPEG multiplexed data packet-multiplexed by MPEG encoding together with audio data, etc. Another embodiment for performing editing and recording connecting MPEG image data in a port stream will be described.

  In many cases, variable-length-encoded MPEG image data, fixed-length-encoded MPEG1 layer 2 audio, or AC3 is multiplexed in the transport stream. Therefore, when MPEG image data, which is one of element encoded data in the multiplexed data, is connected, the connection takes into account the consistency of the STD buffer (VBV buffer in video) defined by MPEG. As a method, the method of the above-described embodiment is applied.

  For example, the first and second MPEG image data to be connected are extracted from the first and second MPEG transport streams, respectively, and connected in the same manner as in the above embodiment. Linked segment re-encoded data (re-encoded image data of the segment A, re-encoded image data of the segment B, re-encoded image data of the segment A + B) used for joint reproduction is packetized by the MPEG encoding method. It may be generated and recorded as multiplexed MPEG multiplexed data.

  The state in FIG. 8A shows the state of the packet multiplexed data of the MPEG transport stream. In FIG. 8, a packet described as “V” is a video packet, a packet described as “A” is an audio packet, and a packet described as “S” is a PAT or PMT used in the system. Information packet. Each is recorded in a form compliant with the rules of the MPEG2 system. The video packet has a light gray color.

  FIG. 8B shows a state in which these are shown as a whole. FIG. 8C shows a state where only the video packets are collected. As shown in FIG. 8D, the contents of this video packet are typically an I picture, followed by two B pictures, followed by one P picture, and so on. It is. The code amount is adjusted by re-encoding in one or more of these pictures. For example, FIG. 8E shows a state where the code amount of the transport stream shown in FIG. 8D is reduced. As shown in FIG. 8E, the code amount of each picture is smaller than that of the same picture shown in FIG. FIG. 8F shows the packetized state.

  The reduced area is black. This packet reduces the overall amount of video. And TS is reconstructed. This state is expressed as a whole in FIG. FIG. 11H is an enlarged view of FIG. As a result, a part of the video packet is reduced, and the other element data packets are multiplexed as they are.

  In the MPEG standard, the PCR clock information must be recorded once every 100 msec. In addition, since the data length is changed, the PCR clock information described in each element packet is changed as necessary. Also, PTS and DTS are described in a packet in which a PES header at the head of an access unit (a frame or field picture unit) exists in a video packet.

  In the audio packet, one or a plurality of audio frames are packed with PES, and the PTS is described in the packet having the leading PES header. These time stamp information does not need to be changed if the number of pictures in the image is not increased or decreased, and the audio does not need to be changed unless the playback time length is increased or decreased. Correct PTS and DTS.

  Furthermore, when re-encoding is simply performed, the GOP unit from the I picture that is the reset timing of predictive encoding is easy to handle, but when the GOP is not independent, that is, the B picture at the boundary is the GOP of both sides. When the GOP is closed (when GOP closed gop = 0), the last reference picture of the previous GOP is decoded and stored in an image re-encoding memory (not shown) or the like. It may be necessary to keep it.

  First MPEG multiplexed data (first MPEG transport stream), second MPEG image data (second MPEG transport stream), information on the VBV buffer occupancy value, and its identification shown in FIG. The information address and the re-encoded image data of the section A (or the re-encoded image data of the section B or the re-encoded image data of the section A + B) are included as element encoded data, and the MPEG encoding method is used. The linked section MPEG multiplexed data generated by packet multiplexing may be recorded on the same recording medium, or may be recorded on a plurality of recording media in an arbitrary combination, but recorded on different recording media. It doesn't matter.

  In the case of being recorded separately on a plurality of recording media, it indicates that they are the same information group so that each recording medium is linked and operated (each data and information is linked). Information such as an ID may be recorded on each recording medium. First MPEG multiplexed image data, second MPEG multiplexed image data, re-encoded image data of section A (or re-encoded image data of section B, or re-encoded image of section A + B) (Data) connecting section MPEG multiplexed data and the specific information address are recorded on the same recording medium, the connecting playback on the playback device side can be controlled with one recording medium.

  In addition, this invention is not limited to the above embodiment, For example, SCR (System Clock Reference) can also be used instead of PCR as time information used as a reference | standard. The present invention also includes a computer program for executing the function of each block other than the recording medium 101, the data reading unit 102, and the data writing unit 108 of the image data recording apparatus shown in FIG. . This computer program may be taken into a computer via a recording medium, or may be downloaded to a computer via a network.

It is a figure for demonstrating the concept of the joint reproduction | regeneration implement | achieved by this invention. It is a block diagram of one embodiment of the device of the present invention. It is a figure for demonstrating the information value regarding the VBV buffer occupation value in one embodiment of this invention. It is explanatory drawing explaining the VBV value in one embodiment of this invention. It is a figure for demonstrating the concept of each example of the joint reproduction | regeneration implement | achieved by this invention. It is a figure for demonstrating the concept of the other example of the joint reproduction | regeneration implement | achieved by this invention. It is explanatory drawing which shows an example of the method of estimating a VBV buffer occupation value in this invention apparatus. It is explanatory drawing which shows MPEG TS packet arrangement | positioning. It is a block diagram of an example of a conventional MPEG encoder. It is a block diagram of an example of the conventional MPEG decoder. It is a figure for demonstrating the VBV buffer concept in MPEG. It is a block diagram of an example of the conventional MPEG multiplexing system. It is explanatory drawing which shows the relationship between MPEG TS, PS, and PES. It is explanatory drawing which shows the usage example of PSI of MPEG TS.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 18 Motion compensation predictor 3 DCT device 4 Quantizer 5 VLC device 7 Code amount controller 8, 15 Inverse quantizer 9, 16 Inverse DCT device 11, 19 Image memory 14 VLD device 101 Recording medium 102 Data reading part DESCRIPTION OF SYMBOLS 103 Coded data search part 104 PCR, DTS detector 105 VBV buffer information predictor 106 Parameter setter 107 Image re-encoder 108 Data writing part

Claims (6)

The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording apparatus provided with recording means for generating and recording the first linked section re-encoded data encoded by the MPEG encoding method as data to be connected to and reproduced by the MPEG image data of 2 Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
The recording means sets a section in which a position a predetermined time before the connection designated position in the first MPEG image data is a start position, and an end position is the end of the connection designation position in the first MPEG image data. As the connection section, decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, is re-encoded by the MPEG encoding method and the first connection section. Comprising re-encoding means for generating re-encoded data, and recording the first linked section re-encoded data on a recording medium;
The re-encoding means includes
Recorded before and after the value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the recording position of the first time management information. Interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the two first time reference information and the recording position of the first time management information And a first predicted value calculating means for multiplying the difference between the first transfer rate information value and the first predicted value of the VBV buffer occupation value,
The value of the second time management information of the first frame after the connection designation position in the second MPEG image data, and the two pieces of the second time management information recorded before and after the recording position of the second time management information. The difference between the second time management information interpolation value obtained by interpolation based on the ratio of each data length between the recording position of the time reference information 2 and the recording position of the second time management information, Second predicted value calculating means for multiplying the value of the second transfer rate information to obtain a second predicted value of the VBV buffer occupation value;
Decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, is obtained when the VBV buffer occupancy value starts from the first predicted value, An encoding unit that performs encoding while performing code amount control so that the process proceeds to a predicted value of 2 and ends.
An MPEG image data recording apparatus characterized by the above. The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording apparatus provided with recording means for generating and recording the second linked section re-encoded data encoded by the MPEG encoding method as data to be connected to and reproduced by the MPEG image data of 2 Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
The recording means has a second interval in which the connection designated position in the second MPEG image data is a start position, and a position after a predetermined time from the connection designated position in the second MPEG image data is an end position. As the connection section, decoded image data, which is image data obtained by decoding the second MPEG image data corresponding to the second connection section, is re-encoded by the MPEG encoding method and the second connection section. Re-encoding means for generating re-encoded data, and recording the second joint interval re-encoded data on a recording medium,
The re-encoding means includes
The value of the first time management information of the first frame after the connection designated position in the second MPEG image data and the two first recorded before and after the recording position of the first time management information. The difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the time reference information of 1 and the recording position of the first time management information, First predicted value calculation means for multiplying the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value;
Recorded before and after the value of the second time management information of the first frame after the end position of the second connection section in the second MPEG image data and the recording position of the second time management information. Interpolated value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the two second time reference information and the recording position of the second time management information And a second predicted value calculating means for multiplying the difference between the second transfer rate information value and the second predicted value of the VBV buffer occupation value,
Decoded image data, which is image data obtained by decoding the second MPEG image data corresponding to the second connection interval, is obtained when the VBV buffer occupancy value starts from the first predicted value, An encoding unit that performs encoding while performing code amount control so that the process proceeds to a predicted value of 2 and ends.
An MPEG image data recording apparatus characterized by the above. The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording apparatus provided with recording means for generating and recording the first linked section re-encoded data encoded by the MPEG encoding method as data to be connected to and reproduced by the MPEG image data of 2 Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
The recording means has a section in which a position that is a first predetermined time before the connection specified position in the first MPEG image data is a start position and an end position is the connection specified position in the first MPEG image data. The first connection section is set, the connection designation position in the second MPEG image data is set as a start position, and the position after the second predetermined time from the connection specification position in the second MPEG image data is set as an end position. First and second decoded image data, which are image data obtained by decoding the first and second MPEG image data corresponding to the first and second connection intervals, with the interval as a second connection interval. Re-encoding means for re-encoding with the MPEG encoding method to generate the first and second joint section re-encoded data, and comprising the first and second joint section re-encoding data. Is intended to record the data on a recording medium,
The re-encoding means includes
Recorded before and after the value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the recording position of the first time management information. Interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the two first time reference information and the recording position of the first time management information A first predicted value calculating means for multiplying the first difference with the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value;
Recorded before and after the value of the second time management information of the first frame after the end position of the second connection section in the second MPEG image data and the recording position of the second time management information. Interpolated value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the two second time reference information and the recording position of the second time management information Second predicted value calculation means for multiplying the second difference by the value of the second transfer rate information to obtain a second predicted value of the VBV buffer occupation value;
First decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, and the second MPEG image data corresponding to the second connection section In order so that the VBV buffer occupancy value starts from the first predicted value, transitions to the second predicted value, and finishes sequentially with the second decoded image data that is the image data obtained by decoding, Encoding means for encoding while performing code amount control,
An MPEG image data recording apparatus characterized by the above. The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording method provided with a recording step for generating and recording first linked section re-encoded data encoded by the MPEG encoding method as data to be connected to and reproduced by two MPEG image data Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
In the recording step, a section in which a position a predetermined time before the connection designation position in the first MPEG image data is set as a start position and an area in which the connection designation position in the first MPEG image data is set as an end position is a first position. As the connection section, decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, is re-encoded by the MPEG encoding method and the first connection section. Comprising a re-encoding step for generating re-encoded data, and recording the first linked section re-encoded data on a recording medium,
The re-encoding step includes:
Recorded before and after the value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the recording position of the first time management information. Interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the two first time reference information and the recording position of the first time management information A first predicted value calculation step of obtaining a first predicted value of the VBV buffer occupancy value by multiplying the difference by the value of the first transfer rate information,
The value of the second time management information of the first frame after the connection designation position in the second MPEG image data, and the two pieces of the second time management information recorded before and after the recording position of the second time management information. The difference between the second time management information interpolation value obtained by interpolation based on the ratio of each data length between the recording position of the time reference information 2 and the recording position of the second time management information, A second predicted value calculating step of obtaining a second predicted value of the VBV buffer occupation value by multiplying the value of the second transfer rate information;
Decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, is obtained when the VBV buffer occupancy value starts from the first predicted value, An encoding step of performing encoding while performing code amount control so as to transition to a prediction value of 2 and end.
An MPEG image data recording method characterized by the above. The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording method provided with a recording step for generating and recording second linked section re-encoded data encoded by the MPEG encoding method as data to be reproduced by being linked to two MPEG image data Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
In the recording step, a section in which the connection specified position in the second MPEG image data is a start position and a position after a predetermined time from the connection specified position in the second MPEG image data is an end position is a second position. As the connection section, decoded image data, which is image data obtained by decoding the second MPEG image data corresponding to the second connection section, is re-encoded by the MPEG encoding method and the second connection section. Comprising a re-encoding step for generating re-encoded data, and recording the second joint interval re-encoded data on a recording medium,
The re-encoding step includes:
The value of the first time management information of the first frame after the connection designated position in the second MPEG image data and the two first recorded before and after the recording position of the first time management information. The difference between the interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the time reference information of 1 and the recording position of the first time management information, A first predicted value calculating step of multiplying a value of the first transfer rate information to obtain a first predicted value of a VBV buffer occupation value;
Recorded before and after the value of the second time management information of the first frame after the end position of the second connection section in the second MPEG image data and the recording position of the second time management information. Interpolated value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the two second time reference information and the recording position of the second time management information And a second predicted value calculating step of multiplying the difference by the value of the second transfer rate information to obtain a second predicted value of the VBV buffer occupation value;
Decoded image data, which is image data obtained by decoding the second MPEG image data corresponding to the second connection interval, is obtained when the VBV buffer occupancy value starts from the first predicted value, An encoding step of performing encoding while performing code amount control so as to transition to a prediction value of 2 and end.
An MPEG image data recording method characterized by the above. The first and second MPEG image data, which are image data encoded by the MPEG encoding method, are converted from the first MPEG image data to the first MPEG image data at specified connection specification positions in the respective MPEG image data. MPEG image data recording method provided with a recording step for generating and recording first linked section re-encoded data encoded by the MPEG encoding method as data to be connected to and reproduced by two MPEG image data Because
The first MPEG image data includes first time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the MPEG coding standard, and a decoding time reference as a coding side. After being time-multiplexed with the first time reference information for determining the same and the first transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the first time management information,
The second MPEG image data includes second time management information indicating a decoding start time of an access unit, which is at least a unit of decoding / reproduction, according to the standard of the MPEG encoding method, and a time reference on the decoding side as an encoding side. After being time-multiplexed with the second time reference information for determining the same and the second transfer rate information of the multiplexed data, and stored in the VBV buffer which is a virtual buffer on the decoding side The code amount is controlled and encoded by a model that is instantaneously read and transferred to the decoder at the time indicated by the second time management information,
The recording step includes a section in which a position that is a first predetermined time before the connection designated position in the first MPEG image data is a start position and an end position is the connection designated position in the first MPEG image data. The first connection section is set, the connection designation position in the second MPEG image data is set as a start position, and the position after the second predetermined time from the connection specification position in the second MPEG image data is set as an end position. First and second decoded image data, which are image data obtained by decoding the first and second MPEG image data corresponding to the first and second connection intervals, with the interval as a second connection interval. A re-encoding step of re-encoding with the MPEG encoding method to generate the first and second joint section re-encoded data, and the first and second joint section re-encoding. The issue of data is intended to be recorded on a recording medium,
The re-encoding step includes:
Recorded before and after the value of the first time management information of the first frame after the start position of the first connection section in the first MPEG image data and the recording position of the first time management information. Interpolation value of the first time management information obtained by interpolation based on the ratio of each data length between the recording position of the two first time reference information and the recording position of the first time management information A first predicted value calculating step of multiplying the first difference with the value of the first transfer rate information to obtain a first predicted value of the VBV buffer occupation value;
Recorded before and after the value of the second time management information of the first frame after the end position of the second connection section in the second MPEG image data and the recording position of the second time management information. Interpolated value of the second time management information obtained by interpolation based on the ratio of each data length between the recording position of the two second time reference information and the recording position of the second time management information A second predicted value calculating step of multiplying the second difference with the value of the second transfer rate information to obtain a second predicted value of the VBV buffer occupation value;
First decoded image data, which is image data obtained by decoding the first MPEG image data corresponding to the first connection section, and the second MPEG image data corresponding to the second connection section In order so that the VBV buffer occupancy value starts from the first predicted value, transitions to the second predicted value, and finishes sequentially with the second decoded image data that is the image data obtained by decoding, An encoding step for encoding while performing code amount control,
An MPEG image data recording method characterized by the above.

Images