JPH10322704A - Bit stream data decoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To output normal images instead of abnormal images by comparing the information of a packet length detected from the packet header of an inputted bit stream with the value of measuring the packet length of the bit stream and replacing erroneous image data with stored image data, when judged that an error is present. SOLUTION: Inputted encoded data are decoded by a variable length decoding means 2 and an encoded prediction error, a prediction mode and a motion vector are outputted. The encoded prediction error is decoded by an inverse quantization means 3 and an inverse DCT transformer 4, added with a prediction image outputted from a motion compensation prediction means 8 in an adder 5 and stored in an image memory 6. Upon receiving error information, the image memory 6 inhibits the write of the image data inputted from the adder 5 and outputs the stored image data of one frame before to a format conversion means 7.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a bit stream data decoding device.

[0002]

2. Description of the Related Art FIG. 6 is a block diagram showing a conventional bit stream data decoding device. In the figure, reference numeral 1 denotes an input buffer for storing coded data of an input image, and an output of the input buffer 1 is provided to a variable length decoding means 2. The output of the variable length decoding means 2 is
And the output of the inverse quantization means 3 is supplied to the inverse DCT
Given to. The output of the inverse DCT converter 4 is provided to a first input of an adder 5. The output of the adder 5 is an image memory 6
Is given and stored. The output of the image memory 6 is provided to the format conversion means 7 and the motion compensation prediction means 8, and the output of the motion compensation prediction means 8 is provided to the second input of the adder 5 and the image memory 6. Also, the output of the format conversion means 7 is given to a D / A converter 9, and the D / A
The output of the converter 9 is the image output of the bitstream data decoding device.

[0003] First, an encoding method based on this decoding apparatus will be described. Motion-compensated predictive coding is widely used as a coding method for reducing the temporal redundancy of a moving image. FIG. 7 shows a conceptual diagram of the motion compensation predictive coding method. The input image is divided into, for example, macroblocks of 16 pixels × 16 lines, and a decoded image of an already coded image is used as a reference image, and a region of an image closest to each macroblock is searched from the reference image to obtain a predicted image. Is generated, and the position of the predicted image is encoded as a motion vector.

FIG. 7A shows a case of unidirectional prediction, and FIG. 7B shows a case of bidirectional prediction. In the case of unidirectional prediction, the reference image is an already encoded image, which is an image temporally earlier than the input image. In the case of bidirectional prediction, the reference image is an already coded image, and both images temporally before and after the input image are used. That is, it is necessary to change the order of the input images before encoding. In this case, the predicted image includes a predicted image when a temporally previous image is used as a reference image, a predicted image when a temporally later image is used as a reference image, and an average of these predicted images. Among the three images, an image having a small prediction error is selected. The difference between the input image and the predicted image is encoded using a discrete cosine transform (hereinafter, referred to as “DCT”) as a prediction error.

[0005] An image for which unidirectional prediction is performed is called a P picture, and an image for which bidirectional prediction is performed is called a B picture. In addition, due to an initial state of encoding or a case where a transmission error occurs, encoding may be performed within one screen without performing prediction. This image is called an I picture.

The image memory 6 in FIG. 6 decodes in the order of encoding, and outputs the images in the order of moving images input to an encoder (not shown) at the time of encoding. For example, four images are stored in the image memory 6 in chronological order (I picture, P picture, B picture, B picture).
When decoding the first picture I picture, decoding the second picture P picture, and decoding the third and fourth picture B pictures, the picture memory 6 stores Output the image first, then output the third and fourth images, and then output the second image. By outputting the images from the image memory 6 in such an order, it is possible to faithfully reproduce the moving image at the time of encoding.

Next, a description will be given of a method of transmitting a video coded by the above method. Video / audio information compressed and encoded by an MPEG encoder (not shown) is usually transmitted to another device having a decoder via a communication medium. At this time, the compression-coded video / audio information is divided and transmitted in a time-series manner. As a transmission method, a program stream (hereinafter, referred to as “program stream”) that handles video / audio compression-coded data in one program unit is used. PS) and a transport stream (hereinafter, “T”) that handles video / audio compression-encoded data in a plurality of program units.
S ”).

The PS is used when compressed and encoded data is transmitted from a single medium such as a DVD.
The TS is used when multi-channel information is multiplexed into one channel information in time series and transmitted as in CS digital broadcasting.

On the other hand, in MPEG, in order to facilitate the data transfer between the PS and the TS, a PES (Packetized Elementary) having a data stream structure intermediate between the two.
Stream). The PES is composed of a chunk of data having a predetermined length (hereinafter, referred to as a “packet”), and is treated as a PES packet. P
In the case of S, based on this PES packet,
The information unique to the PS is added to the ES packet to handle it as one stream.

A PS composed of a plurality of PES packets
, Various information such as a video signal and an audio signal are transmitted therein. These pieces of information are transmitted using an elementary stream (hereinafter, referred to as “ES”) composed of a plurality of PES packets as one information unit. A video signal is referred to as “video ES”, and an audio signal is referred to as “audio ES”.

Here, the structure of the video ES will be described. FIG. 8 is a diagram showing a hierarchical structure of the video ES. The video ES is stream data having a hierarchical structure, and the highest layer is called a sequence layer. 32-bit SH
C (Sequence Header Code) 50 is data indicating the start of the sequence layer and information unique to the sequence layer, and SEC (Sequence End Code) 51 is data indicating the end of the sequence layer. For MPEG-2, S
Three extension data exist after the HC 50, and a plurality of GOPs (Group of Picture) 52 continue behind the extension data.

[0012] 32-bit GSC (Group Start Code)
Numeral 53 denotes data indicating the start of the GOP layer. TC (Time Code), CG (Closed Gop), BL (Broke)
After the information such as (n Link), an I picture 57, a B picture 58, and a P picture 59 are generated alternately according to the video signal encoding method.

Each picture exists below a GOP layer called a picture layer. 32-bit PSC
(Picture Start Code) 60 is data indicating the start of a picture layer, and is followed by a plurality of slice data.

The relationship between the GOP 52 which is a part of the PS having the above data hierarchy and the PES is as shown in FIG. 9 for a DVD, for example. The PS can usually be divided into a plurality of video packs 70, which are basic data units, and a plurality of packets 72, which are data units constituting the video packs 70. In the case of DVD, one pack = 204
In one pack, there is a pack header 71 indicating the start position of the pack, and in one packet 72, a packet header 74 indicating the start position of the packet, a video start code 83, Video data 75 exists.

In the video data 75, the data in the GOP layer is divided and entered with the GHC 53 at the head.
With this division, it is possible to transmit 1 GOP data by a plurality of video packs.
Since the data length of the OP is variable, if the OP cannot be divided into an integer number of video packs, the final video pack of the video packs constituting one GOP will be shorter than 2048 bytes. In that case, padding 73 is applied to the shortage to make a 2048-byte video pack.

The packet header 74 is composed of 9 to 29 bytes. The packet start code 76 indicates the start of the packet header 74, and the stream ID 77 is information indicating what type (video, audio, etc.) of the packet 72 is. After this data, the PES packet length 78
2 represents the number of data bytes to the end. The video start code 83 in the video pack 70 is the packet 7
2 is a code indicating the start of the video data 75, and this code is followed by actual video data 75.

FIG. 10 is a configuration diagram showing the data length of the PES packet length 78. For example, when the value of the PES packet length 78 is 2025, the data length from the data immediately after the PES packet length 78 to the end of the video data 75 is 2025 bytes.

Next, the operation of the bit stream data decoding device will be described with reference to FIG. The bit stream data encoded as described above is input to the variable length decoding means 2 and subjected to variable length decoding. At the same time, the encoded prediction error, the prediction mode, and the motion vector are output from the variable length decoding means 2. The encoded prediction error is decoded by the inverse quantization means 3 and the inverse DCT transformer 4, and is added by the adder 5 to the prediction image output from the motion compensation prediction means 8. The image output from the adder 5 is a decoded image, which is provided to the image memory 6 and stored. Since the adder 5 outputs the decoded images in the order of encoding, the image memory 6 rearranges and outputs the decoded images in the order of image input during encoding, that is, in the order of image display. The output of the image memory 6 is converted into the format of the image to be displayed by the format conversion means 7, and is converted and output by the D / A converter 9.

[0019]

Since the conventional bit stream data decoding device is configured as described above, for example, video data in an input bit stream signal is correctly transmitted due to noise during transmission. Even if there is no such data, there is a problem that an abnormal image obtained by decoding this data is output.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and in the case where an input bit stream is not correctly transmitted, a bit stream data decoding in which a normal image is output instead of the above image The aim is to obtain a device.

[0021]

According to the present invention, there is provided a bit stream data decoding apparatus comprising: a packet header detector for detecting information on a packet length of input bit stream data;
A packet length counter for measuring the packet length of the input bit stream data, and a data comparator for comparing the information on the packet length and the output value of the packet length counter, and the bits input based on the comparison result. A processing unit configured to determine whether there is an error in the stream data, and when it is determined that there is an error, replacing the image data stored in the image memory or the format conversion unit with the error image data and outputting the error image data It is provided with.

Further, an error determination means for the input bit stream includes a packet header detector for detecting information of a packet header length of the input bit stream data, and a packet header for measuring the packet header length of the input bit stream data. A length comparator and a data comparator for comparing the information of the packet length with the output value of the packet header length counter, and determining whether or not there is an error in the input bit stream data based on the comparison result. And processing means for, when it is determined that there is an error, replacing the image data stored in the image memory or the format conversion means with the error image data and outputting.

[0023] Further, the input bit stream error determining means includes a packet header detector for detecting packet header information of the input bit stream data, and an initial value of a parameter of a data protocol that can be decoded by the bit stream data decoding apparatus. And a data comparator for comparing the packet header information with an initial value stored in the format information storage means, and the bit stream data input based on the comparison result. Processing means for determining whether there is an error, and when it is determined that there is an error, replacing the image data stored in the image memory or the format conversion means with the error image data and outputting the error data. It is provided.

[0024] Further, the error judging means for the input bit stream is determined based on the sequence header detector for detecting the sequence header information of the input bit stream data and the profile information and level information written in the sequence header. A bit rate calculator for calculating the maximum bit rate of the bit stream data; and a bit rate comparator for comparing the bit rate information detected by the sequence header detector with the output of the bit rate calculator. It is configured to determine the presence or absence of an error in the input bit stream data based on the result, and when it is determined that there is an error,
Processing means for replacing the image data stored in the image memory or the format conversion means with the error image data and outputting the error image data.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In a bit stream data decoding apparatus according to an embodiment of the present invention, the packet length of an input bit stream is detected, and the packet length is determined by the packet length in the header information of the input bit stream. If the input bit stream data is different from the information, an error determination unit is provided for the input bit stream to determine that there is an error in the input bit stream data. Is replaced with the error image data and output.

The error determination means for the input bit stream detects the packet header length of the input bit stream. If the packet header length is different from the information of the packet length in the header information of the input bit stream, the input is determined. It is determined that there is an error in the bit stream data.

Further, the input bit stream error judging means judges that there is an error in the input bit stream data when the information recorded in the packet header is different from the initial value stored in the format information storage means. to decide.

The input bit stream error determination means calculates the maximum bit rate from the profile information and the level information written in the sequence header in the input bit stream, and calculates the bit rate detected from the sequence header. If the value of the rate information exceeds this calculated value, it is determined that there is an error in the input bit stream data.

Hereinafter, the present invention will be described in detail with reference to the drawings showing the embodiments. Embodiment 1 FIG. FIG. 1 is a block diagram showing a bit stream data decoding apparatus according to Embodiment 1 of the present invention. In the figure, reference numeral 1 denotes an input buffer for storing coded data of an input image, and an output of the input buffer 1 is provided to a variable length decoding means 2. Variable length decoding means 2
Is supplied to the inverse quantization means 3, and the output of the inverse quantization means 3 is supplied to the inverse DCT transformer 4. The output of the inverse DCT converter 4 is provided to a first input of an adder 5,
Are given to the image memory 6 and stored. The output of the image memory 6 is provided to the format conversion means 7 and the motion compensation prediction means 8, and the output of the motion compensation prediction means 8 is provided to the second input of the adder 5 and the image memory 6.

On the other hand, the encoded data is given to a packet header detector 11 and a packet length counter 12, and the output of the packet header detector 11 is
And the data comparator 13, and the output of the packet length counter 12 is provided to the data comparator 13. The output of the data comparator 13 is provided to the image memory 6 and the format converter 7, the output of the format converter 7 is provided to the D / A converter 9, and the output of the D / A converter 9 is the bit stream data. The image is output from the decoding device. In the above configuration, the packet header detector 11, the packet length counter 12, and the data comparator 13 constitute the error determination means 20.

Next, the operation of the first embodiment will be described. The coded data input via the input buffer 1 is variable-length decoded by the variable-length decoding means 2, and the coded prediction error, prediction mode, and motion vector are output. The encoded prediction error is calculated by the inverse quantization means 3 and the inverse D
The decoded image is decoded by the CT converter 4, and is added to the predicted image output from the motion compensation prediction unit 8 in the adder 5. The image output from the adder 5 is a decoded image, which is provided to the image memory 6 and stored. Since the adder 5 outputs the decoded image in the encoding order, the image memory 6 rearranges and outputs the decoded image in the input order, that is, the display order.

On the other hand, the input encoded data is supplied to a packet header detector 11 and a packet length counter 12, and the packet header detector 11 detects the packet start code 76 shown in FIG. The PES packet length 78 is detected from the packet header, the value is output to the data comparator 13, and the packet start code 76 and the detection information of the PES packet length 78 are transmitted to the packet length counter 12. Output.

The packet length counter 12 starts counting the input data length in synchronization with the transmission timing of the input data at the same time as the detection information of the PES packet length 78 is input. The count value until input is calculated by the data comparator 13
Output to

The data comparator 13 compares the value of the PES packet length 78 input from the packet header detector 11 with the count value input from the packet length counter 12 which is the actual packet length. It is determined that the input coded data contains some data loss or error data due to noise, and outputs error information to the image memory 6.

The image memory 6 receiving the error information prohibits the writing of the image data input from the adder 5 and outputs the stored image data one frame before to the format conversion means 7.

The data comparator 13 compares the value of the PES packet length 78 detected by the packet header detector 11 and the count value input from the packet length counter 12 with respect to the next input coded data. If not, error information is output to the image memory 6. The image memory 6 prohibits writing of the image data input from the adder 5, and outputs the image data stored in the memory to the format conversion means 7. The output of error information is stopped when they match. The image memory 6 from which the input of the error information is stopped resumes the writing of the image data, and returns to the normal decoding process of outputting the written image data to the format conversion means 7.

In the first embodiment, the data comparator 1
3, when the error information is output, the image data of the previous frame stored in the image memory 6 is output to the format conversion means 7. However, as shown by the broken line in FIG. The error information may be output from the means 20 to the format conversion means 7, and the image data stored in the format conversion means 7 may be output to the D / A converter 9. This is because the format conversion means 7 usually has a memory, and the image data after the format conversion exists in this memory. Even in this case, the same effect as in the case of using the image memory 6 is obtained. can get.

Embodiment 2 FIG. 2 is a block diagram showing a bit stream data decoding apparatus according to a second embodiment of the present invention, wherein the same reference numerals as those in FIG. 1 denote the same or corresponding parts, respectively, and a packet header detector 11, a packet header length counter. An error determination means 20 comprising the data comparator 14 and the data comparator 13 is provided. Hereinafter, a configuration of a portion different from the first embodiment and an operation thereof will be described.

The coded data is input to the input buffer 1 and input to the packet header detector 11 and the packet header length counter 14. The packet header detector 11 detects the pack start code 80 shown in FIG. 9 to know the start of the pack header 71, then reads the information of the pack stuffing length 81 written in the pack header 71, and That is, the value of the data length of the packet header 74 and the detection information of the packet start code 76 written in the packet header 74 are output to the packet header length counter 14.

The packet header detector 11 further includes a PES packet header length 7 in the packet header 74.
9 and outputs the value of the PES packet header length to the data comparator 13, and outputs information indicating that the packet start code 76, the PES packet header length 79, and the video data start code 83 have been detected to the packet. Output to the header length counter 14.

The packet header length counter 14 starts counting the data length in synchronization with the transmission timing of the input data at the same time when the packet start code 76 is detected, and continues until the detection information of the video start code 83 is input. The count value is output to the data comparator 13.

The data comparator 13 has a PES packet header length 7 input from the packet header detector 11.
The value of 9 is compared with the count value input from the packet header length counter 14, which is the actual packet length. If they do not match, there is some missing data in the input coded data, or mixed error data due to noise. And outputs the error information to the image memory 6.

The image memory 6 receiving the error information prohibits the writing of the image data input from the adder 5 and outputs the stored image data of one frame before to the format conversion means 7.

The data comparator 13 compares the value of the PES packet header length 79 detected by the packet header detector 11 and the count value input from the packet header length counter 14 for the coded data input next. If they do not match, error information is output to the image memory 6. The image memory 6 prohibits writing of the image data input from the adder 5, and converts the image data stored in the memory into the format conversion means 7. The operation of outputting error information is stopped when they match. The image memory 6 from which the input of the error information is stopped resumes the writing of the image data, and returns to the normal decoding process of outputting the written image data to the format conversion means 7.

As shown by the broken line in FIG. 2, the error information is output from the error determination means 20 to the format conversion means 7 and the image data stored in the format conversion means 7 is converted into a D / A signal. It may be output to the container 9.

Embodiment 3 FIG. 3 is a block diagram showing a bit stream data decoding apparatus according to a third embodiment of the present invention. The same reference numerals as those in FIG. 1 denote the same or corresponding parts, respectively. 15 and an error determining means 20 comprising a data comparator 13. The format information recording means 15 is written in a parameter of a data protocol which can be decoded by the bit stream data decoding device, for example, a header of encoded data. Are written in a memory, a magnetic medium, or the like. Hereinafter, as an example of a parameter of a data protocol that can be decoded by the bit stream data decoding device, an example using a PCT (Picture Coding Type) 90 in the first part of each of the pictures 57, 58, and 59 shown in FIG. An operation of a portion different from the first embodiment will be described.

When the packet header detector 11 detects a PCT 90 from the input coded data,
The address information in which the information 90 is written is output to the format information storage unit 15.

The format information storage means 15 outputs format data corresponding to the input address information, that is, information of the PCT 90, to the data comparator 13.
PCT 90 is 3-bit information and indicates the type of picture. For example, PCT = “001” is an I picture, “0”
"10" is defined as a P picture, and "011" is defined as a B picture.The use of other values is prohibited by the MPEG standard.The data comparator 13 is input from the packet header detector 11. The information of the PCT 90 is compared with the information of the PCT 90 read from the format information storage means 15.
Since data that should not be used is detected, it is determined that some noise has been added to the input coded data, and error information is output to the image memory 6.

The image memory 6 receiving the error information prohibits the writing of the image data input from the adder 5 and outputs the stored image data one frame before to the format conversion means 7.

The data comparator 13 stores information of the PCT 90 detected by the packet header detector 11 with respect to the coded data inputted next and the format information storage means 15.
The information of the PCT 90 read from the PCT 90 is compared. If they do not match, error information is output to the image memory 6. The image memory 6 prohibits the writing of the image data input from the adder 5 and stores the image data in the memory. The operation of outputting the current image data to the format conversion means 7 is repeated, and the output of the error information is stopped when they match. The image memory 6 from which the input of the error information is stopped resumes the writing of the image data, and returns to the normal decoding process of outputting the written image data to the format conversion means 7.

As indicated by the broken line in FIG. 3, the error information is output from the error judging means 20 to the format converting means 7 and the image data stored in the format converting means 7 is converted by the D / A conversion. It may be output to the container 9.

In the above description, an example using the PCT 90 has been described. However, similar effects can be obtained by using other values as long as the values change depending on the encoding / decoding format.

Embodiment 4 FIG. 4 is a block diagram showing a bit stream data decoding apparatus according to a fourth embodiment of the present invention, wherein the same reference numerals as those in FIG. 1 denote the same or corresponding parts, respectively, and a sequence header detector 16, a bit rate calculator 17 and an error determining means 20 comprising a bit rate comparator 18. Also,
FIG. 5 is a diagram showing details of the sequence layer of the video ES of MPEG-2.

Hereinafter, the operation of a portion different from the first embodiment will be described. The encoded data is input to the input buffer 1 and to the sequence header detector 16.
The sequence header detector 16 includes a BRV (Bit Rate Value) 91 and a PALI (Profil) in the sequence layer of FIG.
e And Level Indication) 92 and BRE (Bit Rate)
Extension) 93 are detected.

The BRE 93 is the upper 12 bits of the bit rate value.
The bit BRV91 indicates the lower 18 bits of the bit rate value, and the bit rate is represented by a total of 30 bits in a binary number (hereinafter, the representation of these 30 bits in a decimal number is "B
RT ”). The PALI 92 is a value indicating the quality of an image according to the MPEG standard, and is composed of 3-bit profile identification data and a 4-bit level identification value. The profile identification data indicates the compression method of the image,
It takes a value from “000” to “111”. For example, "1
"00" can be identified by 3-bit data such as a main profile, "011" can be identified by an SNR scalable profile, etc. The level identification value is a value indicating the number of pixels of an image.
“1000” indicates a main level (720 horizontal pixels, 576 lines, 30 screens per second), and “0100” indicates a high level (1440 horizontal pixels, 1152 lines, 60 screens per second).

In the MPEG, the above profile and level values determine the maximum value of the bit rate of the transmission image. For example, the maximum bit rate of MPEG data encoded at the main profile main level is 15 Mpbs, and the main profile The highest bit rate of high-level encoded MPEG data is 80 Mpb
It is fixed as s.

The PALI 92 detected by the sequence header detector 16 is output to the bit rate calculator 17, where the bit rate T of the coded data input by the bit rate calculator 17 is calculated, and the bit rate comparator 1
8 is output. In addition, BRV (Bit Rate Value) 91, BR detected by the sequence header detector 16
E (Bit Rate Extension) 93 is also a bit rate comparator 1
8 is output. The bit rate comparator 18 compares the value of the input bit rate T with the BRT. If BRT> T, whether bit data encoded at a bit rate not expected by the bit stream data decoding device has been input, Or PALI92, BRV91 and BRE9
3 judges that an incorrect value is being transmitted due to noise or the like, and outputs error information to the image memory 6.

The image memory 6 receiving the error information inhibits the writing of the image data input from the adder 5 and outputs the stored image data of one frame before to the format conversion means 7.

The bit rate comparator 18 compares the value of the next input bit rate T with the BRT and finds that BRT> T
In the case of (1), the error information is output to the image memory 6, and the image memory 6 prohibits the writing of the image data input from the adder 5, and outputs the image data stored in the memory to the format conversion means 7. The output of the error information is stopped when BRT ≦ T. Image memory 6 with no error information input
Restarts the writing of the image data, and returns to the normal decoding process of outputting the written image data to the format conversion means 7.

As shown by the broken line in FIG. 4, error information is output from the error determination means 20 to the format conversion means 7, and the image data stored in the format conversion means 7 is converted into a D / A signal. It may be output to the container 9.

In each of the above embodiments, the image memory 6 and the format conversion means 7 have different configurations. However, since both perform processing using memories, it is not always necessary to provide memories independently. Even if the area is divided on one memory or the same memory area is shared and used, the same effect can be obtained.

Although the processing steps in each of the above embodiments are realized using hardware, it is not always necessary to perform processing using hardware, but means capable of performing arithmetic processing, such as a microcomputer, etc. The same effect can be obtained by performing processing by software using.

[0063]

Since the present invention is configured as described above, it has the following effects.

A packet header detector for detecting the packet length information of the input bit stream data, a packet length counter for measuring the packet length of the input bit stream data, the packet length information and the packet length The data length of the PES packet of the input bit stream data is monitored by the error determination means composed of a data comparator for comparing the output value of the counter, and when the data is lost due to noise or the like and the PES packet length changes, Since it is determined that there is an error and the image data stored in the image memory or the format conversion means is replaced with the error image data and output, the unsightly error image can be prevented from being output as it is.

The error determination means includes a packet header detector for detecting information on the packet header length of the input bit stream data, a packet header length counter for measuring the packet header length of the input bit stream data, A data comparator for comparing the packet header length information with the output value of the packet header length counter, and
The data length of the ES packet is monitored, and if data is lost due to noise or the like and the PES packet header length changes, it is determined that there is an error, and the image data stored in the image memory or the format conversion means is replaced with the error image. Since it is configured to output by replacing with data,
It is possible to prevent an unsightly error image from being output as it is.

The error determining means includes a packet header detector for detecting packet header information of the input bit stream data, and a format information storage means for storing an initial value of the header information of the input bit stream data. And a data comparator for comparing the packet header information with an initial value stored in the format information storage means. The header information accompanying the input encoded data of the input bit stream data is The irregular data is monitored by comparing with the data of the bit stream data stored in advance, and when the data is destroyed due to noise or the like during transmission and changed from the data at the time of encoding, it is determined that there is an error. Image data stored in the image memory or format conversion means And then, is output by replacing the error image data, it is possible to prevent the unsightly error image is output as it is.

The error judging means is provided by a sequence header detector for detecting sequence header information of the input bit stream data, and profile information and level information written in the sequence header.
A bit rate calculator for calculating the maximum bit rate of the input bit stream data, and a bit rate comparator for comparing the bit rate information detected by the sequence header detector and the output of the bit rate calculator, Monitors bit rate information in the header information associated with the input coded data of the input bit stream data, and when coded data having a bit rate that cannot be decoded by the bit stream data decoding device is input. If the bit rate information is destroyed due to noise or the like during transmission and the bit rate information at the time of encoding changes, it is determined that there is an error, and the image data stored in the image memory or the format conversion means is replaced with the error. It is configured to output by replacing with image data, There error image can be prevented from being output as it is.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a bitstream data decoding device according to Embodiment 1 of the present invention.

FIG. 2 is a block diagram showing a bitstream data decoding device according to Embodiment 2 of the present invention.

FIG. 3 is a block diagram showing a bitstream data decoding device according to Embodiment 3 of the present invention.

FIG. 4 is a block diagram showing a bitstream data decoding device according to Embodiment 4 of the present invention.

FIG. 5 is a configuration diagram showing header information in a sequence layer of a video ES of MPEG-2.

FIG. 6 is a block diagram showing a conventional bit stream data decoding device.

FIG. 7 is a conceptual diagram showing a conventional motion compensation prediction method.

FIG. 8 is a configuration diagram showing a hierarchical structure of an MPEG-2 video ES.

FIG. 9 is a configuration diagram showing a relationship between a GOP and a PES in a DVD.

FIG. 10 is a configuration diagram showing a PES packet length.

[Explanation of symbols]

1 input buffer, 2 variable length decoding means, 3 inverse quantization means, 4 inverse DCT converter, 5 adder, 6 image memory, 7 format conversion means, 9 D / A converter,
10 error determination means, 11 packet header detector,
12 packet length counter, 13 data comparator, 14
Packet header length counter, 15 format information storage means, 16 sequence header detector, 17 bit rate calculator, 18 bit rate comparator, 20 error determination means.

Claims (4)

[Claims] An input buffer to which a quantizing process is performed after performing a discrete cosine transform, and further, a variable length coded bit stream data is input, and the bit stream data input from the input buffer is variable. Variable-length decoding means for outputting a prediction error encoded by long decoding, a prediction mode, and a motion vector; an inverse quantization means and an inverse DCT transformer for decoding the encoded prediction error; An adder that adds the prediction mode and the motion vector, the prediction error, and the predicted image to output a decoded image, an image memory that stores the decoded image, rearranges and outputs the input image, and an image memory. Format conversion means for converting the decoded image output from the device into a predetermined format; and D / A conversion of the format-converted decoded image. A packet header detector for detecting information on the packet length of the input bit stream data, and a packet for measuring the packet length of the input bit stream data. A length counter, and a data comparator for comparing the information of the packet length with the output value of the packet length counter, and configured to determine whether there is an error in the input bit stream data based on the comparison result. An error determining unit; and a processing unit that replaces the image data stored in the image memory or the format converting unit with the error image data and outputs the error image data when it is determined that there is an error. Bit stream data decoding device. 2. A packet header detector for detecting information of a packet header length of input bit stream data, a packet header length counter for measuring a packet header length of input bit stream data, and And a data comparator for comparing the packet length information with an output value of the packet header length counter, and configured to determine whether or not there is an error in the input bit stream data based on the comparison result. 2. The bit stream data decoding device according to 1. 3. The error determination means stores a packet header detector for detecting packet header information of input bit stream data, and stores initial values of parameters of a data protocol that can be decoded by the bit stream data decoding device. And a data comparator for comparing the packet header information with the initial value stored in the format information storage means. The bit stream data decoding device according to claim 1, configured to determine presence / absence. 4. An apparatus according to claim 1, wherein said error determining means detects a sequence header information of the input bit stream data, and a profile header and level information written in the sequence header. And a bit rate comparator for comparing the bit rate information detected by the sequence header detector with the output of the bit rate calculator. 2. The bit stream data decoding apparatus according to claim 1, wherein the bit stream data decoding apparatus is configured to determine whether there is an error in the bit stream data.