JP2003101956A - Reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform reverse reproduction that can display the entire compressed frames by using frame correlation. SOLUTION: In the case that image data to be decoded area a B picture arranged ahead of an I picture in a first GOP to which the image data to be decoded belong at reverse reproduction that is reverse single frame step reproduction or reverse slow reproduction, a reproducing means is controlled so as to access an entry point of a second GOP arranged just before the first GOP. A controlling part controls so as not to decode the B picture in the second GOP in obtaining a predictive image needed to decode the first image data to be decoded. At reverse reproduction, the reproducing means is controlled so as to access the entry point of the second GOP in the case that a reference image needed to decode the image data to be decoded is encoded by referring to an image belonging to the second GOP arranged just before the first GOP to which the image data to be decoded belong.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention reads data from a disc in which moving image data and audio data are compressed and time-division-multiplexed and time-division multiplexed signals are recorded to reversely reproduce images. The present invention relates to a reproducing apparatus suitable for reverse low speed (slow) reproduction and reverse frame advance reproduction.

[0002]

2. Description of the Related Art As a method for compressing moving image data and audio data, a conventional MPEG (Motion Picture Codin)
g Experts Group) method has been proposed and this MPE
The moving image data and the audio data are compressed by the G method and are time-division multiplexed and recorded on the disc. FIG. 11 shows an example of the data format of this multiplexed data. As shown in FIG. 11A, one unit of the multiplexed bitstream in the multiplexed data time-division multiplexed recorded on the disc is at least one pack (PACK),
It is composed of ISO_11172_end_code indicating the end. This ISO_11172_end_code is a 32-bit code and is defined as "0x000001b9" in hexadecimal. In addition,
The leading "0x" indicates that it is a hexadecimal number (C
language).

The pack is, as shown in FIG.
Pack_Start_Code, SCR (SystemClock Referece), M
It is composed of a header composed of UX_Rate and at least one packet. Pack_Start of this header
_Code is a 32-bit code and is hexadecimal "0x00".
0001b4 ”. Although the length of the pack is defined as a variable length, it may be fixed at 2048 bytes, for example, and the following description will be made assuming that it is fixed at 2048 bytes.

Further, the packets forming the pack are Packet_Start_Code_Prefi as shown in FIG.
x, Stream_ID, Packet_length, PTS (Presentation Ti
It is composed of a header composed of me Stamp) and DTS (Decoding Time Stamp) and packet data (Code Data). This Packet_Start_Code_Prefix is a 24-bit code and is defined as "0x000001" in hexadecimal. Stream_ID is an 8-bit code and indicates the packet type (Stream Type) as shown in FIG. Furthermore, Packet_length (16 bits) indicates the length of subsequent packets.

Packet data (Code Data) of each packet
In (), audio data is recorded in the case of an audio stream, and video data is recorded in the case of a video stream. There are 32 types of audio stream and 1 for video stream.
Since you can have 6 different stream ids,
Up to this number, a plurality of audio signals and video signals can be multiplexed.

FIG. 13 is a block diagram showing a schematic configuration of a conventional coded data decoding device for reproducing such multiplexed data. In the encoded data reproducing apparatus shown in this figure, the time division multiplexed data read from the DSM (Digital Storage Media) 10 is supplied to a demultiplexing apparatus 21, and a header demultiplexing circuit 22 demultiplexes the time division multiplexed data from the pack header. The packet header is separated and supplied to the control device 24, and the time-division multiplexed data is supplied to the input terminal G of the switching circuit 23. The output terminals H1 and H2 of the switching circuit 23
Are connected to the input terminals of the video decoder 25 and the audio decoder 26, respectively.

Further, the control device 24 of the separating device 21 controls the stre of the packet header supplied from the header separating circuit 22.
Switching circuit 2 for video signals according to am id
3 is connected to the output terminal H1 to send a signal to the video decoder 25, and in the case of an audio signal, the switching circuit 23 is connected to the output terminal H2 to send a signal to the audio decoder 26. Then, the time-division multiplexed data is correctly separated and supplied to the corresponding decoder.

[0008] Next, the MPEG will be briefly described. In MPEG, an intra-frame coded picture (I picture), an inter-frame forward predictive coded picture (P picture), and a bidirectional predictive coded picture (B picture). )have. The I-picture is encoded by using only the data in the image (frame or field). Therefore, the compression efficiency of the I picture data is low, but since it is encoded in the frame, it can be decoded by itself.

On the other hand, the P picture is a picture which is predictively coded by calculating the difference from the previous I picture or P picture, and the compression efficiency is improved.
However, when decoding, it is necessary that the previous I picture or P picture referred to at the time of predictive coding has been decoded. The B picture is a picture that is predictively coded by calculating the difference from the preceding or following I picture or P picture, and has the highest compression efficiency. However, in the case of decoding, the I picture or P before and after the reference referred to at the time of predictive coding is used.
The picture is required to be decoded.

Here, FIG. 14 shows a conceptual diagram of a bit stream including I-pictures, P-pictures and B-pictures recorded in the DSM 10 which is a disc. As shown in this figure, a series of video bitstreams is divided into one or more GOPs (Group of Pictures). This GOP is a Group-Start-Code (GSC), Time-Cod
It is configured to have e (TC), Closed GOP (CG), and Broken-Link (BC) data, followed by multiple picture data. The beginning of the picture data in the GOP is specified by the I picture.

By the way, when the time-division multiplexed video data is compressed by the MPEG encoding system,
Random access and search operations will be restricted. That is, when random access is performed to the disc recorded by MPEG, the pictures other than the I picture cannot be decoded because the pictures prior to the accessed picture are not read at the time of access. Need access to the I picture. For this reason, in the normal MPEG system, about two I pictures appear in one second so that random accessibility and compression efficiency are balanced.

Further, when the video data is compressed at a fixed rate, the I picture periodically appears at a predetermined position, so that the position can be calculated and accessed. However, when the variable rate compression is performed, the position of the I picture becomes indefinite, making it difficult to access. In addition, MPE
In the case of G, compression is usually performed at a variable rate.

Therefore, in the coded data decoding apparatus shown in FIG. 13, when a search command for random access is received, the following operation is performed. When receiving the search command, the main controller (not shown)
The controller 24, the video decoder 25, and the audio decoder 26 are instructed to transition to the search mode. In the transitioned search mode, the video decoder 25
Only I-picture data is selected from the input video data and decoded. Alternatively, only the I picture data is selected in the separation device 21 and input to the video decoder 25. Then, in the video decoder 25, the input I-picture data is decoded and output. As a result, the searched image is displayed.

In the search mode, the control device 2
4 issues a command to the DSM 10 to move the data read position on the disk forward (or backward).
The moving amount of the read position at this time depends on the search speed, the coding rate, and the like, but generally, the moving amount increases as the search speed increases and the coding rate increases. Then, when the read position moves to a predetermined position, the data output from the DSM 10 is transferred to the separation device 21.
The header separation circuit 22 separates the video data and supplies it to the video decoder 25. The video decoder 25 decodes and outputs the first I picture that appears. As a result, the searched image is displayed. In the search mode, the audio decoder 26 is in the mute state.

As described above, the search operation for continuously reproducing I pictures is realized by repeating random access. That is, when the user instructs, for example, a high-speed forward search, the video decoder 25 skips data of a predetermined number of frames among the input data, searches for an I picture, decodes and outputs the I picture. . Alternatively, the separation device 21 searches for an I picture, and only the I picture data is supplied to the video decoder 25 and decoded. By repeating such an operation, a search operation in which I pictures are continuously reproduced is realized.

However, conventionally, it is not known where an I picture for searching and displaying a desired image is located, so that there is a drawback that it takes a lot of time to access the I picture on which the desired image is displayed. It was
Therefore, an encoded data decoding device that solves this problem has been proposed. FIG. 15 shows an example of the configuration of the encoded data decoding device.

In the coded data decoding apparatus shown in this figure, the data recorded on the optical disk 60 is reproduced by the pickup 61. The output signal of the pickup 61 is input to the demodulation circuit 62. The demodulation circuit 62 demodulates the reproduction signal output by the pickup 61 and outputs the demodulated signal to the ECC circuit 63. ECC circuit 6
3 performs error detection / correction on the input data and outputs the data to the demultiplexer circuit 64. The video data separated in the demultiplexer circuit 64 is input to the video decoder 65, and the audio data is
The signals are input to the audio decoder 66, and the decompressed signals are output from each.

The controller (main controller) 67 sends commands to the video decoder 65 and the audio decoder 66 for control in response to a command from the user, and gives an access command to the drive controller 69. As a result, the drive control device 69 drives the pickup 61 using the tracking servo circuit 70 in accordance with a command from the controller 67 to access the disc 60. Further, TOC data located at the head of the data is separated by the demultiplexer circuit 64 and input to the controller 67,
It is stored in the storage device 68. TOC data is T when needed
It is read from the OC storage device 68 and used by the controller 67.

Next, the operation of the data decoding apparatus shown in FIG. 15 will be described. When the optical disk 60 is inserted, the controller 67 gives the drive controller 69 a command for reading the first sector. The drive control device 69 drives the pickup 61 by the tracking servo circuit 70 to start reproduction from the position of the head sector on the disk 60. The pickup 61 irradiates the optical disc 60 with a laser beam and reproduces the data recorded on the optical disc 60 from the reflected light. The reproduction signal output from the pickup 61 is input to the demodulation circuit 62 and demodulated. The demodulated data string is input to the ECC circuit 63, and data error detection / correction is performed. The error-corrected data is input to the demultiplexer circuit 64.

Next, in the coded data decoding apparatus shown in FIG. 15, when the search instruction for random access is received, the controller 67 causes the drive control apparatus 6 to operate.
9. The video decoder 65 and the audio decoder 66 are instructed to transit to the search mode. In the search mode, the video decoder 66 decodes only I picture data from the input video data. Alternatively, only the I picture data is selected by the demultiplexer circuit 64 and input to the video decoder 65.
As a result, in the video decoder 65, the I picture data is decoded and displayed on the display (not shown).

Further, in the search mode, the drive controller 69 commands the tracking servo circuit 70 to move the data read position on the disk forward (or backward). The movement amount of the read position at this time depends on the search speed, the coding rate, and the like, but generally, the higher the search speed and the higher the coding rate, the larger the movement amount. When the read position moves to a predetermined position, the data output from the pickup 61 is input to the demultiplexer circuit 64 via the demodulation circuit 62 and the ECC circuit 63. The demultiplexer circuit 64 separates the video data into a video decoder 65.
Supply to. The video decoder 65 decodes and outputs the first I picture that appears. The audio decoder 66 is muted in the search mode.

That is, the search operation is realized by repeating random access to the I picture. That is, when the user instructs, for example, a fast forward search, the video decoder 65 skips a predetermined number of frames of the input data, searches for an I picture, decodes it, and outputs it. Alternatively, the controller 6 may be used according to the information recorded in the TOC storage device 68.
7 gives a command to the drive control circuit 69, the drive control circuit 69 drives the tracking servo circuit 70 to search for an I picture, and only the I picture data is supplied to the video decoder 65 and decoded. The search operation is performed by repeating such operations.

By the way, the positions of I pictures can be stored in the disk 60 as TOC data and can be used by detecting the information. However, all the positions of I pictures were stored in the disk 60. If
Assuming that two I pictures are generated per second,
The number of I pictures recorded on a recording medium having a capacity of 1 hour is 7
It becomes 200. It is difficult to store all the positions in the disk 60 as TOC data from the viewpoint of the capacity, and in reality, the disk 60 can record only the position of the I picture at the head of each track. Therefore,
There is a problem that it is practically difficult to perform a precise search operation using the TOC data stored in the disk 60. Further, this causes a similar problem not only in the disk but also in other storage media.

Therefore, a data decoding apparatus described in Japanese Patent Application No. 5-235121 has been proposed in which an entry packet is provided in the bitstream to facilitate the search. A block diagram is shown in FIG. An example of the data format in this case is shown in FIG. 17. In this data format, an entry packet (Entry Packet) is multiplexed before the video packet including the entry point in the multiplexed data. Has been done. Further, another example of the data format is shown in FIG. 18. In this data format, position information of three entry points in the forward direction and three entry points in the backward direction is recorded in each packet.

Next, the operation of the encoded data decoding apparatus shown in FIG. 16 will be described for the case of decoding the formatted data shown in FIG. 17 or FIG.
The header separation circuit 22 of the separation device 21 separates the pack header, the packet header and the entry packet from the data read from the DSM 10 and supplies the separated data to the control device 24, and the time division multiplexed data of the switching circuit 23. Supply to the input terminal G. The output terminals H1 and H2 of the switching circuit 23 are connected to the input terminals of the video decoder 25 and the audio decoder 26, respectively.

The control unit 24 also includes a header separation circuit 22.
Information about the entry point (entry packet information) is read from the input data and supplied to the entry point storage device 41 for writing.
Since information about the current read position is supplied from the DSM 10 to the control device 24, the control device 24 stores the position of the entry point and its content in the entry point storage device 41 in association with the current read position. You can

The control device 24 of the separating device 21 controls the stream i in the packet header supplied from the header separating circuit 22.
According to d, the input terminal G and the output terminals H1 and H2 of the switching circuit 23 are sequentially connected to correctly separate the time-division multiplexed data, and the video data is sent to the video decoder 25 and the audio data is sent to the audio decoder 26.
, Respectively.

Next, when a search operation is instructed, the main control device (not shown) causes the control device 24 and the video decoder 25 to operate.
And instructing the audio decoder 26 to transition to the search mode. Further, the control device 24 reads the current read position from the output of the DSM 10 and extracts the entry point near that position from the entry point storage device 41. The entry point storage device 41 stores the information of the entry packet reproduced in the reproduction mode as needed. Alternatively, when the power of the apparatus is turned on, when a disc is loaded, when playback is instructed, etc., all or a predetermined range of entry packets recorded on the disc loaded in the DSM 10 is set at a predetermined timing. Information may be read and stored in advance.

When the controller 24 finds the entry point, it sends a search command to the DSM 10 to move the read position to the entry point at high speed. When the move is complete, the DSM 10 replays the data from that entry point and supplies it to the separator 21. When the data format is as shown in FIG. 17, the entry packet is placed immediately before the video data in which the I picture is recorded. Therefore, the header separation circuit 2
When the video data following the entry packet is separated by 2 and supplied to the video decoder 25, the first picture of this video data becomes an I picture. Therefore, the video decoder 25 immediately decodes and outputs the first I picture that appears. As a result, the searched image is displayed.

If the data format is as shown in FIG. 18, the position information of three entry points before and after the entry packet is recorded in the entry packet. The next entry packet is searched for and the reproduction operation is repeated. As a result, the I-pictures are rapidly and successively decoded and the searched image is displayed.

When the search speed is high, the control unit 24 makes access to a farther entry point, and when the search speed is slow, makes the access point closer. The entry point is shown in FIG.
As shown in (3), when three records are recorded in each of the forward direction and the reverse direction, it is possible to provide three or more stages of variations in search speed depending on the combination of entry points to be selected. As described above, in the encoded data decoding device shown in FIG. 17, since the position of the I picture as the access point can be known from the entry packet, the search operation can be completed quickly.

A technique has been proposed in which the position (access point) of a nearby I picture is recorded as subcode data in the sector header, and a similar search operation can be realized by using this technique. it can.

Further, in the MPEG2 System, PSD (Pr
ogram Stream Directory) is defined. The structure of this PSD is shown in FIG. 19, which is shown in FIG.
prev_directory_offset, next_directory_offset, PES_
Even if the position of the entry point is recorded by header_position_offset, the same search operation can be realized.

[0034]

By the way, in addition to normal playback and search operations, video equipment such as VTRs are required to have functions such as backward playback, backward slow playback, and backward frame advance playback. . However, ISO11172-2 (MPEG1 Vide
o), the moving image data recorded in accordance with ISO13818-2 (MPEG2 Video) is referred to as shown in FIG. 20 (a), and therefore, the picture data is stored in the order shown in FIG. 20 (b). The order is changed and recorded. That is, the structure of inter-frame prediction in the MPEG system is shown in FIG.
As shown in 0 (a), 1 GOP (Group Of Pictures)
) Is composed of, for example, 15 frames, and 1 GOP
In the above, the I picture is 1 frame, the P picture is 4 frames, and the remaining 10 frames are B pictures.

This I picture is predicted within one frame.
It is an intra-frame predictive-coded image that has been measured and encoded, and
Kucha is the previously encoded temporally previous frame (I
Picture or P picture)
Inter-frame forward prediction coded image, B picture is temporal
Bidirectional prediction code that predicts by referring to the two frames before and after
Since it is an encoded image, as shown by the arrow, I picture
I₂ Is predictively coded only within that frame,
P picture P_Five Is I picture I₂ Refer to between frames
It is predictively coded and P picture P₈ Is a P picture P
_Five The inter-frame predictive coding is performed with reference to. Furthermore
, B picture B₃ , B_Four Is I picture I ₂ And P Picchi
P_Five Interframe predictive coding with reference to two
Cage, B picture B₆ , B₇ Is a P picture P_Five And P
Cha P₈ Interframe predictive coding with reference to
ing. Similarly, the subsequent pictures are predictively coded
Has been created.

By the way, in order to decode a picture that is predictively coded in this way, since the I picture is predictively coded within a frame, it is possible to decode only an I picture, but a P picture. Is predictively coded with reference to a temporally previous I picture or P picture, a temporally previous I picture or P picture is required at the time of decoding, and a B picture is temporally preceding or following I picture. Alternatively, since it is predictively coded with reference to a P picture, temporally preceding and following I pictures or P pictures are required at the time of decoding. Therefore, the pictures are exchanged as shown in FIG. 20B so that the pictures required for decoding can be decoded first.

As shown in the figure, this interchange requires the B pictures B ₀ and B ₁ to require the I picture I ₂ at the time of decoding, so that the I picture I ₂ precedes the B pictures B ₀ and B _1. Since the pictures B ₃ and B ₄ require the I picture I ₂ and the P picture P ₅ at the time of decoding, the B pictures B ₆ and B ₇ are similarly arranged so that the P picture P ₅ precedes the B pictures B ₃ and B _4. Is a P picture P when decoding
B picture B because ₅ and P picture P ₈ are required
_Since B picture B ₉ and B ₁₀ require P picture P ₈ and P picture P ₁₁ at the time of decoding so that P picture P ₈ precedes ₆ and B ₇ , P picture from B picture B ₉ and B ₁₀
The picture P ₁₁ has been replaced so that it precedes. Similarly, the B pictures B ₁₂ and B ₁₃ are interchanged so that the P picture P ₁₄ precedes them.

Such a bit stream is shown in FIG.
If the video decoder is transferred to the video decoder in the order shown in (b), normal reproduction is possible, but there is a problem that other special reproduction cannot be performed as it is. Further, in the above-described conventional encoded data decoding device,
Regarding the search operation for continuously displaying only I pictures, a method of implementation has been disclosed. However, by displaying all frames including P pictures and B pictures as well as I pictures continuously in reverse order, There is a problem that it is difficult to realize the directional reproduction, the reverse slow reproduction, and the reverse frame advance reproduction.

Therefore, the present invention has been made in view of such a situation, and not only I pictures but all frames including P pictures and B pictures can be continuously displayed in reverse order. It is an object of the present invention to provide a reproducing device capable of directional reproduction, reverse slow reproduction, and reverse frame-by-frame reproduction.

[0040]

In order to achieve the above object, the present invention provides a recording medium for reproducing a recording medium in which image data of at least I picture, P picture and B picture is recorded. A reproduction unit that reproduces the image data from, a decoding unit that decodes the reproduced image data, and the image data to be decoded is the first GOP to which the image data to be decoded belongs during reverse reproduction.
In the case of a B picture arranged before the I picture in the second picture, a second picture arranged immediately before the first GOP is arranged.
The reproducing means is controlled so as to access the entry point of the GOP. Further, the control unit controls so as not to decode the B picture in the second GOP when obtaining the predicted image required to decode the image data to be first decoded. The reverse reproduction is reverse rotation frame advance reproduction or reverse rotation slow reproduction. Further, in a reproducing device for reproducing a recording medium on which image data encoded by using predictive encoding is recorded, a reproducing means for reproducing the image data from the recording medium, and a decoding means for decoding the reproduced image data. , The reference image necessary for decoding the image data to be decoded in the backward reproduction refers to the image belonging to the second GOP arranged immediately before the first GOP to which the image data to be decoded belongs. If it is encoded, the control means controls the reproducing means so as to access the entry point of the second GOP.

[0041]

According to the present invention, the image data to be decoded belongs to the first frame to which the image data to be decoded belongs in the reverse reproduction which is the reverse frame advance reproduction or the reverse slow reproduction.
In the case of a B picture arranged before the I picture in the GOP, the reproducing means is controlled to access the entry point of the second GOP arranged immediately before the first GOP. Further, the control unit controls so as not to decode the B picture in the second GOP when obtaining the predicted image required to decode the image data to be first decoded. In the backward reproduction, the reference image necessary for decoding the image data to be decoded is the first GOP to which the image data to be decoded belongs.
When the image is encoded by referring to the image belonging to the second GOP arranged immediately before, the reproducing means is controlled so as to access the entry point of the second GOP.

[0042]

FIG. 1 shows the configuration of a first embodiment of an encoded data decoding device embodying a reproducing device of the present invention. In the encoded data decoding device shown in this figure, the data recorded on the optical disc 60 is reproduced by the pickup 61. The pickup 61 reads out the data recorded on the optical disc 60 and supplies the output signal thereof to the demodulation circuit 62. The demodulation circuit 62 demodulates the reproduction signal output by the pickup 61, and the ECC circuit 6
Supply to 3. The ECC circuit 63 performs error detection / correction on the input data and sends it to the demultiplexer 21.

Then, the header separation circuit 22 provided in the demultiplexer 21 separates the pack header and the packet header from the data sent from the ECC circuit 63 and supplies them to the control device 24, as well as time division. The multiplexed data is supplied to the input terminal G of the switching circuit 23. The output terminals H1 and H2 of the switching circuit 23 are connected to the input terminals of the video decoder 25 and the audio decoder 26, respectively.

Further, the control device 24 provided in the demultiplexer 21 reads out information (entry packet information) about the entry point from the data input from the header separation circuit 22 and sends it to the controller 67. Information about this entry point is stored in the entry point storage unit 41 in the controller 67. Since the controller 67 is supplied with the information on the current read position from the drive control circuit 69, the controller 6
7, the entry point position and the current read position information content can be associated with each other and stored in the entry point storage unit 41.

Further, the control device 24 sequentially connects the input terminal G and the output terminals H1 and H2 of the switching circuit 23 in accordance with the stream id information in the packet header supplied from the header separation circuit 22 to perform time division multiplexing. The separated data is correctly separated, and the video data is sent to the video decoder 25 and the audio data is sent to the audio decoder 26.
, Respectively. In this way, the video data separated by the demultiplexer circuit 64 is input to the video decoder 25 for decoding, and the audio data is input to the audio decoder 26 for decoding, and a signal decompressed from each of them. Is output.

By the way, the video data supplied from the demultiplexer 21 to the video decoder 25 is input to the video code buffer 121 in the video decoder 25 and the data is temporarily stored. Then, the data read from the video code buffer 121 is input to the picture header detection circuit 122, and the picture header detection circuit 1
At 22, the data sent from the video code buffer 121 to the video decoding circuit and the frame memory 125 is analyzed, and the picture header and GOP header are detected. The content of this picture header is sent to the controller 67 as picture header information. The picture header information indicates temporal reference (TR) information indicating the screen order in the GOP and one of the pictures I, P, and B. It is composed of picture type information. The picture header described in this embodiment is ISO1
1172-2 (MPEG1 Video) and ISO13818-2 (MPEG2 Video)
Is a picture_header defined in.

Next, the data from the picture header detection circuit 122 is supplied to the picture selection circuit 123,
By switching the movable contact P of the switch to the fixed contact Q1 or the fixed contact Q2, the data of a specific picture is selectively sent to the video decoding circuit and the frame memory 125.

That is, when the movable contact P is connected to the fixed contact Q1, data is sent to the video decoding circuit and the frame memory 125, and when the movable contact P is connected to the fixed contact Q2, The data is discarded without being sent to the video decoding circuit and frame memory 125. The condition for switching the movable contact P is
Although instructed by the controller 67, the movable contact P is switched in synchronization with the detection signal from the picture header detection circuit 122. Therefore, the video decoding circuit and the frame memory 1 can selectively select the data of a specific picture.
25 to be able to send.

The video decoding circuit and frame memory 125 decompresses the compressed video data sent from the video code buffer 121 through the picture header detection circuit 122 and the picture selection circuit 123, and outputs it to the external frame memory 131. . In this embodiment, the video decoding circuit and the frame memory 125 are ISO13818.
-It compresses the compressed data based on -2 (MPEG2 Video), so it has at least three frame memories inside.

By the way, the external frame memory 131
Holds the video signal sent from the video decoder 25 by a picture hold command signal from the controller 67. When the picture hold command signal is at low level, the video signals sent from the video decoder 25 are written one screen at a time and output as they are. Then, when the picture hold command signal from the controller 67 becomes high level, the data of the picture that was output last is held, and thereafter, the held picture data is held until the picture hold command signal returns to low level again. Will be output. That is,
The same image will be displayed continuously.

When the user gives an instruction for reproduction or the like to the encoded data decoding device, the controller 67
A control command is sent to the video decoder 25 and the audio decoder 26, and an access command is given to the drive control device 69. The drive control device 69 follows the command from the controller 67 and the tracking servo circuit 7
0 is used to drive the pickup 61, and the disc 60
To access. Further, the drive control device 69 supplies the accessed current read position information to the controller 67.

Next, the operation of the encoded data decoding apparatus shown in FIG. 1 when normal reproduction and subsequent reverse frame advance are performed will be described. It should be noted that, among the reverse reproduction, the reverse slow reproduction, the reverse constant speed reproduction, and the reverse high speed reproduction can be realized by the operation of repeatedly performing the reverse frame feed at each appropriate cycle. Therefore, in the present embodiment, the reverse frame feed is performed. Only the embodiment will be described, but other reverse reproduction can be similarly realized within the range of the speed permitted by the processing speed of the constituent elements.

When a normal reproduction is instructed by the user, the controller 67 gives a command to the drive control circuit 69 to start the supply of data. In this case, disk 6
If the start position of data supply on 0 is designated by the user, the designated position is set as the start position, and otherwise, the start is made from the first data recorded on the disc 60. The drive control circuit 69 to which the command is given controls the tracking servo circuit 70 to drive the pickup 61, and the disc 60 is accessed. Then, the pickup 61 sends the signal read from the disc 60 to the demodulation circuit 62, and the demodulation circuit 6
2 demodulates the reproduction signal output by the pickup 61,
Output to the ECC circuit 63. Further, the ECC circuit 63 performs data error detection / correction, and the demultiplexer 21
To supply data to.

When the data supply is started, the drive control device 69 comes to supply the current read position information to the controller 67. And EC
The header separation circuit 22 of the demultiplexer 21 to which the data is supplied from the C circuit 63 separates the pack header and the packet header from the data sent from the ECC circuit 63, supplies the separated pack header and packet header to the control device 24, and performs time division. Input the multiplexed data to the input terminal G of the switching circuit 23.
Supply to.

Further, the controller 2 of the demultiplexer 21
4, the input terminal G and the output terminals H1 and H2 of the switching circuit 23 are sequentially connected according to the stream_id information of the packet header supplied from the header separation circuit 22,
The time-division multiplexed data is correctly separated, and the video data is supplied to the video decoder 25 and the audio data is supplied to the audio decoder 26. Further, the control device 24 of the demultiplexer 21 uses the header separation circuit 2
Information regarding the entry point (entry packet information) is read out from the data input from No. 2 and sent to the controller 67. Information about this entry point is stored in the entry point storage unit 41.

The video data separated by the demultiplexer circuit 64 is input to the video decoder 25, and the audio data is input to the audio decoder 2
6 is input, and the signals that are decompressed from each are output. That is, the data input to the video decoder 25 is temporarily stored in the video code buffer 121,
The data read from the video code buffer 121 is
Picture header detection circuit 122 and picture selection circuit 12
3 through video decoding circuit and frame memory 12
5, the compressed video data is decompressed to be original video data.

During normal reproduction, the picture hold command signal supplied from the controller 67 to the external frame memory 131 is at low level, and the video signal sent from the video decoder 25 is output as it is. In this way, the moving image is reproduced in the normal reproduction. At this time, the entry point storage unit 41 of the controller 67 stores the information (entry packet information) about the entry point supplied from the control device 24 of the demultiplexer 21 at the current read position supplied from the drive control circuit 69. It is stored in association with the content of information. Further, the controller 67 always receives the picture header information sent from the picture header detection circuit 122, and holds the latest information.

When the user gives an instruction for reverse frame advance, the controller 67 starts preparation for reverse frame advance. That is, first, the picture hold signal to the external frame memory 131 is set to a high level to hold the last output picture. Then, from the information about the entry point stored during normal playback and the latest picture header information, the temporal reference information of the picture to be displayed first, and
The position on the disk 60 of the entry point of the GOP to which the picture belongs is determined. Then, when the picture to be displayed first is determined, the controller 67 starts the reverse frame advancing operation.

A flow chart of the reverse frame feed operation of the controller 67 is shown in FIG. 2, and the operation of the reverse frame feed controller 67 will be described with reference to the reverse frame feed operation flowchart. In this operation flowchart, when reverse frame advance is started, a work of loading a picture to be displayed is performed in step S10, and then,
In step S20, the work of displaying the picture to be displayed is performed. Further, in step S30, it is determined whether or not a command for performing the reverse frame advancing by one frame is input from the user. When it is determined that an instruction for advancing the reverse frame is input, the process proceeds to step S40 to display the next item. The work of determining the (new) picture is performed. Then, when the work of determining the next (new) picture is completed, the process returns to step S10, the processes of steps S10 and S20 are performed, and the image of the previous frame is displayed. The work processes of steps S10 to S40 are cyclically performed.

As a result, reverse frame feeding is sequentially performed. Here, the picture to be displayed is the target picture, and the GOP to which the target picture belongs is targeted.
The position of the entry point of the GOP that is accessed to decode the GOP and the target GOP is called the access position. Then, the work in the operation flowchart shown in FIG. 2 can be restated as a target picture loading work, a target picture displaying work, a user command waiting, and a next (new) picture determining work.

Next, step S1 of the operation flow chart.
FIG. 3 shows a target picture loading flowchart in the target picture loading operation of the controller 67 performed at 0. When the target picture load flowchart is started, the controller 67 sends a seek command to the drive control circuit 69 in step S100 so that the seek operation to the access position is performed. In this case, if the access position is not defined at the beginning of the reverse frame advancing operation, the position of the entry point of the GOP to which the target picture (the picture to be displayed first) belongs is used as the access position.

Then, the process proceeds to step S110 and waits until a picture header is detected. That is, as in the case of normal reproduction, the data read from the specified position on the disc 60 is subjected to necessary processing in each block and sent to the picture header detection circuit 122 of the video decoder.
By extracting the picture type information and the temporal reference information and sending them to the controller 67, the control waits until the picture header information is detected. When the picture header is detected, it is determined in step S120 whether the GOP to which the detected picture header belongs is the target GOP.

Here, the GOP in which the picture header is detected
Is different from the target GOP, when the B picture used by referencing the picture of the immediately preceding GOP at the time of decoding is the target picture, and when the GOP immediately before the target GOP is specified as the access position. is there. Therefore, if the GOP detected in this way is different from the target GOP, the process branches to step S160,
It is determined whether the detected picture is a B picture, and B
If it is determined to be a picture, the picture selection circuit 123
By operating the movable contact P of the above so as to be connected to the fixed contact Q2, the B picture data is discarded. Then, the process returns to step S110 via step S180 to enter the picture header detection waiting state again.

If it is determined that the picture detected in step S160 is an I picture or a P picture,
After proceeding to step S170, the movable contact P of the picture selection circuit 123 is operated so as to be connected to the fixed contact Q1, so that the decoding process is performed and then step S18.
The process returns to step S110 via 0 to return to the picture header detection waiting state. As described above, in the case of the I picture or the P picture, the decoding process is performed even for the picture that does not need to be displayed in order to create a predicted image for decoding the subsequent picture. However, since it is not the target picture, the picture hold signal to the external frame memory 131 is kept at the high level so that it is not displayed.

Next, GO detected in step S120
If it is determined that P matches the target GOP, the temporal
It is determined whether or not the reference value and the temporal reference value of the target picture match. If it is determined that they match, the movable contact P of the picture selection circuit 123 is connected to the fixed contact Q1 in step S140, the picture data is sent to the video decoding circuit and the frame memory 125, and the decoding process is performed. After being stored in the frame memory, the picture loading operation is completed.

If the temporal reference value of the picture detected in step S130 and the temporal reference value of the target picture do not match, the process proceeds to step S160 and it is determined whether the detected picture is a B picture or not. When the same process as above is performed and a B picture is determined, the B picture is discarded and the state returns to the picture header detection waiting state. Furthermore, if the detected picture is not the B picture in step S160 and the detected picture is an I picture or a P picture, the decoding process is performed in step S170 as described above, and then the picture header detection waiting state is set. Return. this is,
This is because, in the case of an I picture or a P picture, it is necessary to perform decoding processing on a picture that does not need to be displayed in order to create a predicted image for a subsequent picture. Since it is not necessary to display, the picture hold signal to the external frame memory 131 is kept at the high level.

If an I picture is detected by the picture selection circuit 123, that picture, that is, GO
Store the temporal reference value of the I picture at the beginning of P. This is to obtain information used in the work of determining the next picture to be displayed, which will be described later. Further, as will be described later in the work of determining a picture to be displayed, when a picture in the last display order of the GOP before the GOP displayed immediately before is specified as the target picture, and when a specific temporal reference value is unknown For the time being, "$ FFFF" indicating the end is set. As described above, when the temporal reference value of the target picture is “$ FFFF” indicating the last picture of the previous GOP, the GOP is set in step S180.
When the maximum temporal reference (TR) value in the above is detected, the process proceeds to step S190. This step S1
At 90, the maximum temporal reference value is retained, but at the beginning of the next GOP after the target GOP, that is,
When the second I picture or GOP header is detected, the maximum temporal reference value detected so far is updated to the original temporal reference value of the target picture, and the target picture loading process starts from the beginning. Repeated.

Next, FIG. 4 shows a flowchart of the target picture displaying operation of the controller 67 performed in step S20 of the reverse frame advancing operation flowchart. In this case, in this flowchart, since the display image holding operation is performed by the external frame memory 131, the display of the target picture in this flowchart is performed by the controller 67 in step S200.
The picture hold signal from the device to the external frame memory 131 is once set to the low level, and then step S210.
Then, the operation of returning the picture hold signal to the external frame memory 131 to the high level is performed again, and the operation is terminated.

Next, FIG. 5 shows a flowchart of the operation of determining the next (new) target picture by the controller 67, which is performed in step S40 of the reverse frame advance operation flowchart. Here, as information for identifying the next (new) target picture, the temporal reference value and access position of the next (new) target picture are determined. When the work of determining the next (new) target picture is started, it is determined in step S300 whether the temporal reference (TR) value of the target picture so far is "0" (zero),
If it is determined to be "0", in step S310, "$ FFFF" indicating the last picture of the GOP immediately before the target picture so far is set as the temporal reference value of the new target picture, and the new reference picture is set to the new reference picture. As the access position, the position of the entry point immediately before the entry point of the target GOP so far is read from the entry point storage device 41 and determined.

Then, in step S320, a new target picture is determined. That is, the temporal reference value of the new target picture is adopted as the temporal reference value of the target picture, and the new access position is adopted as the access position, and the step S10 shown in FIG. And used in step S20.
If it is determined in step S300 that the temporal reference value of the target picture is larger than "0" (zero), the temporal reference value of the first I picture of the GOP to which the target picture belongs is known in step S330. Whether or not it is determined, and if it is determined that the information is known, the process proceeds to step S340.

Then, in step S340, the magnitude relation between the temporal reference value of the target picture up to now and the temporal reference value of the I picture at the head of the GOP to which the target picture belongs so far is determined. In this case, if it is determined that the temporal reference value of the target picture up to now is the same as or smaller than the temporal reference value of the first I picture to which the target picture of the GOP belongs, step S
Similar to the case where it is determined in 330 that the I picture is not known, the process proceeds to step S360, and the temporal reference value of the new target picture is decremented by "1" from the temporal reference value of the target picture so far. The number is set. Also, as a new access position,
The position of the entry point of the GOP immediately before the GOP is read and determined from the entry point storage device 41, the process proceeds to step S320, the operation in step S320 is performed, and the work is completed.

If it is determined that the temporal reference value of the target picture so far is larger than the temporal reference value of the leading I picture to which the target picture of the GOP so far belongs, step S
Go to 350 and enter the new target picture temporal
As the reference value, a number decremented by "1" from the temporal reference value of the target picture up to now is set. As the new access position, the existing access position, that is, the existing entry point position of the target GOP is used as it is, the process proceeds to step S320, the operation in step S320 is performed, and the operation is completed.

Then, by repeating the above-described various operations, that is, the target picture loading operation, the target picture displaying operation, the waiting for a user command, and the operation of determining the next (new) target picture, the backward frame advance is realized. It

FIG. 6 is a block diagram showing the configuration of the second embodiment of the present invention. In the second embodiment shown in this figure, the external frame memory 131 is omitted by manipulating the inside of the video decoding circuit and the frame memory 125 in the first embodiment, which is the same as the first embodiment. The reverse frame feed can be realized. Video decoding circuit and frame memory 1 in the first embodiment
As shown in FIG. 6, 25 is internally a video decoding circuit 132 and a frame memory write control switch 12.
6, internal frame memory (FM) a127, internal frame memory (FM) b128, internal frame memory (F
M) c129, frame memory read control switch 1
Equipped with 30.

That is, in the normal reproduction, the frame memory write control switch 126 and the frame memory read control switch 130 are controlled according to the picture type detected by the detection circuit of the picture header detection circuit 122 at the timing shown in FIG. As a result, the I picture, P picture, and B picture can be decoded and displayed in the correct display order.

The switching timing of the write control switch 126 and the read control switch 130 will be described. First, the I picture I ₂ is decoded and output from the video decoding circuit 132. At this time, the write control switch 126 is applied to the contact a. Since it has been switched, the I picture I ₂ is stored in the frame memory a 127. Next, the B picture B ₀ is the frame memory a
Decoded by referring to the I picture I ₂ stored in 127 and the P picture of the immediately preceding GOP (not shown),
Write control switch 1 whose contact is switched to c
It is stored in the frame memory c129 via 26. At this time, the read control switch 130 is switched to the contact point c, so the B picture B ₀ is read and sent to the display.

Subsequently, the B picture B ₁ is decoded with reference to the I picture I ₂ stored in the frame memory a 127 and the P picture of the immediately preceding GOP (not shown), and the contact is switched to c. It is stored in the frame memory c129 via the switch 126. At this time, since the read control switch 130 is switched to the contact c, the B picture B ₁ is read and sent to the display. Further, the P picture P ₅ is the I picture I ₂ stored in the frame memory 16a.
The frame memory b is decoded through the write control switch 126, which is switched to the contact b.
Written to 128. At this time, read control switch 1
The switch 30 is switched to the contact a, and the I picture I ₂ is output and sent to the display.

Next, the B picture B ₃ is decoded with reference to the I picture I ₂ stored in the frame memory a 127 and the P picture P ₅ stored in the frame memory b 128, and the contact is switched to c. The frame memory c via the write control switch 126
129 is stored. At this time, the read control switch 13
Since 0 is switched to contact c, B picture B ₃
Is output and sent to the display.

Then, the write control switch 126 and the read control switch 1 are set at the timings shown in FIG.
30 are sequentially switched, and B ₀ →
The pictures are read out in the order of B ₁ → I ₂ → B ₃ → B ₄ → P ₅ → ... and sent to the display. In this way, the read control switch 130 is controlled so that the order of pictures is rearranged and sent to the display in the original picture order shown in 20 (a).

In the second embodiment shown in FIG. 6, the controller 67 controls the frame memory write control switch 126 and the frame memory read control switch 130, so that the frame memory can be effectively used during backward frame feeding. By doing so, the external frame memory 131 becomes unnecessary. Therefore, only the video decoding circuit and the frame memory different from those of the first embodiment will be described, and the description of the other parts will be omitted.

The video decoding circuit 132 in the second embodiment performs a calculation for expanding the compressed video data in picture units. At this time, when performing decompression using inter-frame correlation like P picture and B picture, the data stored in the internal frame memories a127, b128, and c129 can be referred to and decoded. ing. Then, the video data decoded by the arithmetic processing in the video decoding circuit 132 is sent to the frame memory write control switch 126.

Frame memory write control switch 12
The controller 6 switches the contact point w to the contact points a, b, and c according to an instruction from the controller 61, thereby sending the video data sent from the video decoding circuit 132 to the internal frame memories a127, b128, and c129, respectively. Internal frame memory a127, same b128, same c
129 is a frame memory write control switch 126
The video data sent from is stored (stored), and the video data read via the frame memory read control switch 130 is output. That is, the frame memory read control switch 130 is the controller 6
By switching the contact point z to a, b, c according to the instruction of 1, the internal frame memories a127, b128,
Among the c129, the frame stored in an arbitrary frame memory is output.

In the normal reproduction, the control is performed as described above with reference to the timing of FIG.
Obtains the picture type of the detected picture from the picture header detection circuit 122 each time the picture header is detected, and thereby the frame memory write control switch 126 and the frame memory read control switch 130 are obtained.
Are controlling.

Here, when the user instructs the backward frame advance, instead of performing the picture hold operation using the external frame memory 131, the frame memory read control switch 130 is fixed to the currently selected terminal. Hold the picture. As described above, in the second embodiment, one of the three surfaces of the internal frame memory is used for holding a picture, so in the reverse frame advancing operation, the remaining two surfaces are used for the target picture loading operation. There is a need to do. In this case, B
If the picture is not decoded, the decoding operation can be continued on the two surfaces by alternately storing the I and P pictures on the remaining two surfaces.

When the target picture is a B picture, an internal frame memory for the third plane is required when the target picture is decoded. In this case, it is used for holding the picture until the display timing of the target picture. The new target picture may be overwritten in the internal frame memory that has been selected as the display picture. Note that when writing and reading are performed at the same time, the screen is disturbed; therefore, writing of data to the internal frame memory is delayed by one field from the phase of reading data from the internal frame memory so that the screen is not disturbed.

The flow chart of the reverse frame advancing operation of the controller 67 in the second embodiment is as shown in FIG. 2 as in the first embodiment. That is, the target picture loading operation, the target picture displaying operation, waiting for a user command, and the next picture determining operation are repeated.

Next, FIG. 8 shows a flow chart of the target picture loading operation of the controller 67 in the second embodiment. Since this flowchart is almost the same as that in FIG. 5, only different parts will be described. The different part is that the decoding process in step S140 is not performed. In other words, if it is determined that the detected GOP is the same as the target GOP and the temporal reference value of the detected picture and the temporal reference value of the target picture match, no processing is performed. The loading operation of the picture is finished as it is.

Although not shown in the flow chart,
When it is determined that the temporal reference value of the detected picture and the temporal reference value of the target picture do not match and the detected picture is determined to be an I picture or a P picture, in step S170. Although the decoding process is performed, the remaining two-sided frame memories other than the internal frame memory used for the picture hold are alternately used as the internal frame memory for writing the decoding result. This switching is performed by the frame memory write control switch 1
This is done by operating 26. Further, the frame memory read control switch 130 keeps the internal frame memory used for the picture hold selected and does not display the image of the I picture or P picture in the process of decoding.

Next, FIG. 9 shows a flowchart of the target picture display operation of the controller 67. In the second embodiment, the target picture display operation is performed by decoding the detected picture in step S400 and overwriting the internal frame memory used for picture hold. Also, the flowchart of the target picture determination operation is shown in FIG.
Since it is the same as the above, its explanation is omitted.

As described above, the first embodiment is also implemented in the second embodiment in which the external frame memory 131 is omitted by controlling the timing of performing the write control and read control of the frame memory and the decoding process of the target picture. Reverse frame feed similar to the example can be performed.

In the first and second embodiments of the present invention described above, as the procedure of the reverse frame advance processing, as shown in FIG. 2, the target picture determination operation is performed after receiving the instruction from the user. The target picture is loaded and displayed. However, according to this method, each process is performed after receiving a command from the user, so that it takes time until the screen is updated from the command of the user, and the image is displayed with a delay. . In particular, it takes time to load the target picture.

Therefore, a third embodiment capable of improving this problem will be described with reference to the flow chart shown in FIG. In the third embodiment, the process of determining the target picture and the process of loading the target picture are performed in advance in step S500 without waiting for a command from the user, and the process waits for a command from the user in step S510. When it is determined in step S510 that there is a command from the user, the target picture display operation is performed in step S520. Upon completion of this display work, the next target picture is immediately determined in step S530, and the target picture is loaded in step S500.
In step S510, the process waits for a command from the user via 500.

As a result, the reaction time from the user's command input to the display can be almost eliminated. Generally, when a user issues a reverse frame advance operation command, the interval of reverse frame advance commands issued by the user does not occur in an extremely short period of time, for example, about once a second. In the third embodiment, it is possible to effectively use the time interval of the command from, and it is possible to realize the backward frame advancing function with almost no waiting time.
In the above description, reverse frame feed was explained,
Among the reverse reproduction, the reverse slow reproduction, the reverse constant speed reproduction, and the reverse high speed reproduction can be realized by repeatedly performing the reverse frame feed at appropriate intervals. It can also be applied to reverse reproduction.

[0094]

As described above, according to the present invention, the entry point currently being decoded is stored and the entry point immediately before that is extracted and the frame to be sent for decoding is specified. With, it is possible to reproduce moving image data composed of frames compressed in the reverse direction using inter-frame correlation, which is difficult to reproduce in the reverse direction. Further, since the decoding is started from the I picture located at the head of the GOP, even if the frame memory does not have data necessary for decoding,
Reverse playback capable of displaying all frames (frames), reverse slow playback, and reverse frame advance playback can be enabled.

Further, in the process of detecting the data to be displayed next, unnecessary frames are discarded, and the frames necessary for the next decoding frame are decoded, so that all pictures including B pictures can be quickly extracted. You will be able to play backwards. Furthermore, if the operation just before the display is performed before the user's command is input, it is possible to display the backward frame advance without any delay in the user's command.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of an encoded data decoding device which embodies a reproducing device of the present invention.

FIG. 2 is a flowchart executed by the controller during the reverse frame advancing operation in the first embodiment of the encoded data decoding device which embodies the reproducing device of the present invention.

FIG. 3 is a flowchart of a target picture loading operation performed by the controller during a reverse frame advancing operation in the first embodiment of the encoded data decoding device that embodies the reproducing device of the present invention.

FIG. 4 is a flowchart of a target picture display operation performed by the controller during a reverse frame advance operation in the first embodiment of the encoded data decoding device that embodies the reproduction device of the present invention.

FIG. 5 is a flowchart of a next picture determination operation performed by the controller during a reverse frame advance operation in the first embodiment of the encoded data decoding device that embodies the reproduction device of the present invention.

FIG. 6 is a block diagram showing a configuration of a second embodiment of an encoded data decoding device which embodies a reproducing device of the present invention.

FIG. 7 is a diagram showing write and read timings of a frame memory of an encoded data decoding device that embodies a reproducing device of the present invention.

FIG. 8 is a flowchart performed by the controller during the reverse frame advancing operation in the second embodiment of the encoded data decoding device which embodies the reproducing device of the present invention.

FIG. 9 is a flowchart of a target picture display operation performed by the controller during a reverse frame advance operation in the second embodiment of the encoded data decoding device that embodies the reproduction device of the present invention.

FIG. 10 is a flowchart executed by the controller during the reverse frame advancing operation in the third embodiment of the encoded data decoding device which embodies the reproducing device of the present invention.

FIG. 11 is a diagram showing a data format in a unit of a conventional multiplexed bit stream.

FIG. 12 is a diagram showing a table of stream IDs forming a part of a packet header.

FIG. 13 is a block diagram showing a configuration of a conventional encoded data decoding device.

[Fig. 14] Fig. 14 is a conceptual diagram showing a configuration of a bitstream including an I picture, a P picture, and a B picture.

FIG. 15 is a block diagram showing the configuration of another conventional encoded data decoding device.

FIG. 16 is a block diagram showing the configuration of still another conventional encoded data decoding device.

FIG. 17 is a diagram showing an example of a data format in still another conventional coded data decoding device.

FIG. 18 is a diagram showing another example of a data format in still another conventional encoded data decoding device.

FIG. 19 is a diagram showing the structure of a program stream directory defined in the MPEG2 system.

FIG. 20 is a diagram showing a structure of inter-frame prediction and a structure of a recording frame in MPEG.

[Explanation of symbols]

21 Demultiplexer 22 Header separation circuit 23 Switching circuit 24 Control circuit 25 video decoder 26 Audio Decoder 41 Entry point storage 60 discs 61 pickup 62 Demodulation circuit 63 ECC circuit 69 Drive control circuit 67 Controller 70 Tracking servo circuit 121 video code buffer 122 Picture Header Detection Circuit 123 picture selection circuit 125 Video code circuit and frame memory 126 frame memory write control switch 127, 128, 129 frame memory 130 frame memory read control switch 131 External frame memory 132 Video decoding circuit

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 7/32 H04N 7/137 ZF term (reference) 5C053 FA23 GA11 GB06 GB08 GB15 GB37 HA25 JA21 5C059 MA00 MA04 MA05 MA14 PP05 PP06 PP07 SS13 SS18 SS30 UA02 UA05 UA33 5D044 AB05 AB07 BC02 CC06 DE03 DE12 FG10 FG23 GK08 5D090 AA01 BB02 BB03 BB04 CC04 CC16 DD03 HH03

Claims (4)

[Claims] 1. A reproducing apparatus for reproducing a recording medium on which image data consisting of at least I picture, P picture and B picture is recorded, and a reproducing section for reproducing image data from the recording medium, and the reproduced image data. In the case where the image data to be decoded is a B picture arranged before the I picture in the first GOP to which the image data to be decoded belongs, in the backward reproduction, And a control unit for controlling the reproducing means so as to access the entry point of the second GOP arranged immediately before the first GOP. 2. The control unit controls not to decode a B picture in the second GOP when obtaining a predicted image required to decode the image data to be first decoded. The reproducing apparatus according to claim 1, wherein the reproducing apparatus is characterized in that: 3. The reverse reproduction is reverse frame advance reproduction,
Alternatively, the slow reverse reproduction is performed.
The playback device described. 4. A reproducing apparatus for reproducing a recording medium on which image data encoded by predictive encoding is recorded, and reproducing means for reproducing the image data from the recording medium, and decoding the reproduced image data. And the reference image necessary for decoding the image data to be decoded in the backward reproduction is stored in the second GOP arranged immediately before the first GOP to which the image data to be decoded belongs. And a control unit for controlling the reproduction unit so as to access the entry point of the second GOP when the image is encoded with reference to the belonging image.