JP2002335526A - Method for encoding picture and method for decoding picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily detect an error at a decoding device, and to reduce the influence on a decoded picture when an error is generated during transmission of encoded data. SOLUTION: This picture encoding method comprises an encoding step for generating encoded data, by encoding picture information and a synchronizing word insertion step for inserting a synchronizing word 32 indicating the boundary of the picture information into the boundary position of a macro-block in the encoded data. In this case, the boundary position of the macro-block into which the synchronizing word 32 should be inserted is set as the position before preliminarily decided prescribed data length N.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image encoding method for encoding image information and transmitting it to another device via a communication line, and decoding the transmitted encoded data to restore the original image information. The present invention relates to an image decoding method, and particularly to an image encoding method and an image decoding method for communicating encoded data in an environment with a high transmission error rate, such as a wireless communication network.

[0002]

2. Description of the Related Art In recent years, PHS (Personal Ha) has been developed.
An image communication service using a wireless communication network such as an ndyphone system or a digital cellular network is expected, and its practical use is being promoted.

At present, since these wireless communication networks do not have a large transmission capacity, when transmitting multimedia information having a large amount of information such as voice and image, the amount of data to be transmitted is reduced by encoding.

[0004] One of the problems in wireless communication is that the transmission error rate is high. Since the data encoded as described above has reduced various degrees of redundancy from the original information, the amount of information per bit is larger than that of the data before encoding, and a small amount of error that occurred during transmission is small. This can have a significant effect on the quality after restoration. Therefore, in image communication using these wireless communication networks, it is very important to take measures to prevent deterioration of information quality due to the transmission error.

FIG. 4 shows a configuration of a system including an encoder and a decoder used in conventional image communication. The input image signal 41 is encoded by the information source encoding unit 42 to reduce the data amount. The encoded image data 43 is
The data is sent to the transmission control unit 44 and is sequentially transmitted as encoded data 45 according to the transmission speed of the communication line.

On the other hand, the encoded data 45 received from the communication line is stored in a reception control unit 46, and the reception control unit 4
6 is sequentially transmitted to the information source decoding unit 48 for each encoded unit as an encoded signal 47. Then, the information is decoded by the information source decoding unit 48 and output as an image signal 49.

Hereinafter, the operation of the encoder shown in FIG. 4 (hereinafter referred to as an encoding device) will be described with reference to FIGS. FIG. 5 shows a configuration example of the information source coding unit 42 in FIG.

As shown in FIG. 5, an input image signal 41 is
The difference from the coded reproduction signal of the previous frame or the coded reproduction signal of the previous frame is sent to the conversion / quantization unit 51. Transform / quantization unit 51
Removes the spatial redundancy of the image signal, and transmits the coded signal 43 to an external buffer and the like, and also to the inverse quantization / inverse transform unit 52.

In the inverse quantization / inverse transforming section 52, the transform / inverse
A process opposite to that of the quantization unit 51 is performed to generate a decoded difference signal 53. The decoded difference signal 53 is added to the restored signal 55 of the previous frame stored in the frame memory 54,
The data is newly stored in the frame memory 54.

The restored image signal 55 stored in the frame memory 54 is used to reduce the temporal redundancy by taking the difference from the input image signal 41 of the next frame.

[0011] As shown in FIG. 4, a coded signal 43 from the information source coding unit 42 is sent to a transmission control unit 44. The transmission control unit 44 adds header information indicating encoding conditions and the like to the encoded signal, performs buffering according to the transmission speed of the communication line, and sequentially transmits the signals.

Next, the operation of the decoder shown in FIG. 4 (hereinafter referred to as a decoding device) will be described with reference to FIGS.
This will be described with reference to FIG. As shown in FIG. 4, the received encoded data sequence is divided into header information and image information by a reception control unit 46, and the image information is sent to an information source decoding unit 48.

FIG. 6 shows a configuration example of the information source decoding unit 48. The coded signal 47 of the received image information is decoded by the inverse quantization / inverse transforming unit 61. The decoded difference signal 62 generated here is added to the restoration signal 64 of the previous frame stored in the frame memory 63, is newly stored in the frame memory, and is transmitted to the outside. This output signal 49 is a restored image signal.

As described above, the image signal is encoded and transmitted to reduce the amount of transmission data, but is usually subjected to variable length encoding to further increase the encoding efficiency. When variable-length encoding is performed, the encoded data amount differs for each image frame, for example, and is not constant.

Therefore, in order to determine the start position of an image frame and the like on the decoding device side, the coding device inserts a synchronization word into the coded data sequence. This sync word is
A unique code, which is a fixed pattern and does not match the code of other image information, is allocated so that there is no detection error.

The presence of the synchronization word enables the decoding device to determine the boundary of the image information from the coded data sequence received continuously, and to correctly restore the image signal.

By the way, when encoding an image signal,
In order to reduce the delay in the encoding process, the image signal is divided into blocks and processed. FIG. 7 shows the block division unit. One frame of an image is divided into several GOBs (groups of blocks). The GOB is further divided into several 16-pixel × 16-line macroblocks.

The macro block is divided into four luminance signal blocks and two color difference signal blocks. Each block has a size of 8 pixels × 8 lines. In addition,
A set of macroblocks for one column in the horizontal direction is sometimes referred to as a macroblock line. In the case of FIG.
One GOB corresponds to one macro block line.

These coded data are sequentially transmitted from the upper left to the lower right of the image frame in GOB units, in the GOB in macro block units, and in the macro block in block units.

At this time, a synchronization word is inserted at the beginning of the frame and at the beginning of each GOB so that the decoding device can specify the positions of the frame and the GOB. The macro block and the block are identified by the macro block header following the GOB header and the data in the block.

As described above, since the synchronization word is set, if the decoding device is searching for the pattern,
Even if there is an error in the encoded data on the way, synchronization can always be achieved at the beginning of the GOB, so that the effect of the error can be prevented from spreading later.

[0022]

However, it is necessary for the decoding device to always search for a synchronization word in order to ensure synchronization, and the processing overhead increases accordingly. Furthermore, due to a transmission error, there is a possibility that the same pattern happens to appear even though it is not the beginning of the frame or GOB, or an error occurs in the synchronization word itself and it is recognized as another code. In such a case, the subsequent decoding of the code is affected, resulting in significant degradation of the restored image.

The present invention has been made in view of the above problems, and has been made in the case where an error occurs during the transmission of encoded data,
It is an object of the present invention to provide an image encoding method and an image decoding method capable of facilitating error detection on the decoding device side and reducing the influence on a restored image.

[0024]

According to a first aspect of the present invention, there is provided an encoding step for encoding image information to generate encoded data, and further comprising the steps of: encoding an image at a boundary position of a macroblock in the encoded data; A synchronization word inserting step of inserting a synchronization word indicating a boundary of information, wherein the boundary position of the macroblock in which the synchronization word is inserted is set to a position before a predetermined data length determined in advance. It is characterized in that it is determined.

According to a second aspect of the present invention, the predetermined data length is represented by the number of bits of the encoded data.

According to a third aspect of the present invention, there is provided a receiving step of receiving encoded data including a synchronization word indicating a boundary of image information, and a decoding step of decoding the encoded data to restore the original image information. Wherein the synchronization word is inserted at a boundary position of a macroblock in the encoded data, and the boundary position of the macroblock at which the synchronization word is inserted is more than a predetermined data length. It is characterized in that it is set at the previous position.

According to a fourth aspect of the present invention, the predetermined data length is represented by the number of bits of the encoded data.

[0028]

An embodiment of the present invention will be described below with reference to the drawings. In the following description, the same parts as those of the above-described conventional example will be denoted by the same reference numerals. In the following description, the encoding device and the decoding device are not limited to those that exist independently of each other, but include an encoder and a decoder that are provided in one encoding system.

FIG. 1 is a block diagram showing the configuration of an encoding device according to the present invention. 10 is an encoding control unit, 11 is a synchronization word insertion interval selection unit, 12 is a synchronization word insertion unit, 42
Is an information source coding unit, and 44 is a transmission control unit.

In the above-mentioned coding apparatus, at the start of the transmission of the coded data, the synchronization word insertion interval selecting section 11 determines the approximate size of one macroblock line from the format and frame rate of the input image signal and the transmission speed of the communication line. The data length N is determined and notified to the synchronization word insertion unit 12.

At this time, the data length N to be determined may be statistically obtained in advance and stored in a table, or may be calculated when an encoding condition is determined. The conditions necessary for the calculation are also transmitted to the decoding device at the start of communication.

The synchronization word insertion unit 12 notified of the data length N starts counting the data length from the beginning of each synchronization word, and places the data at the beginning of the nearest macroblock before and after the position where the data length N has elapsed. Insert sync word.

Here, the position where the synchronization word is inserted may be the head of the macroblock which appears first after the data length N has elapsed from each synchronization word, or the macroblock which appears last within the range of the data length N from each synchronization word. Decide on one of the top of the block.

FIG. 2 shows a configuration diagram of a decoding device according to the present invention. 20 is a decoding control unit, 21 is a synchronization word detection unit,
22 is a synchronization word insertion interval determination unit, 46 is a reception control unit,
Reference numeral 48 denotes an information source decoding unit.

In this decoding apparatus, the synchronization word insertion interval determination unit 22 determines the data length N of one macroblock line from the transmission speed of the communication line and the received encoded data, and detects the synchronization word. Notify section 21.

The data length N determined at this time may be statistically obtained in advance and stored in a table, or may be calculated from the coding condition notified from the coding apparatus at the communication destination. In any case, the same algorithm as that of the synchronization word insertion interval selection unit 11 on the encoding device side is adopted.

The synchronization word detecting section 21 notified of the data length N starts counting the data length from the beginning of each synchronization word, and searches for the synchronization word in the forward or reverse direction after the data length N has elapsed. To start.

Whether to start the search in the forward direction or in the reverse direction may be determined in advance in accordance with the operation of the encoding apparatus, or the search may be performed in both directions to a position closer to the data length N. It is also possible to adopt the synchronization word that has appeared.

In this decoding device, if a synchronization word cannot be detected at the above position, it is considered that a transmission error has occurred. At this time, the decoding device is expected to appear a synchronization word. By not using the preceding and succeeding macroblocks for decoding, it is possible to reduce the influence of a transmission error on a restored image.

FIG. 3 shows an example of an encoded data sequence by the encoding apparatus. The PSC 31 is a picture start code indicating the start of a frame. The SYNC 32 is a synchronization word, and the distance between the synchronization words is N ± α (α = 1 data length within one macroblock).

As described above, according to the encoding apparatus,
The synchronization word insertion interval selection unit 11 determines an appropriate data length interval of the synchronization word from the transmission speed of the communication line and the conditions of the input image signal, and notifies the synchronization word insertion unit 12 of the determination. The synchronization word insertion unit 12 counts the data length of the encoded data and inserts a synchronization word according to the data length N notified from the synchronization word insertion interval selection unit 11.

At this time, the data length N is a rough guide, and the position where the synchronization word is actually inserted is at the beginning of the macroblock. Whether the position where the synchronization word is inserted is before or after the data length N is determined in advance, or a condition is determined that the position is determined closer to the data length N.

Further, according to the decoding apparatus, the synchronization word insertion interval determination unit 22 determines the data length interval of the synchronization word by the same algorithm as the synchronization word insertion interval selection unit 11, and the synchronization word detection unit 21 Notify. The synchronization word detection unit 21 counts the data length of the encoded data, and detects a synchronization word according to the data length N notified from the synchronization word insertion interval determination unit 22.

[0044]

As described above, according to the image encoding method and the image decoding method of the present invention, the synchronization words are inserted into the encoded data sequence at semi-fixed intervals and always at the boundary positions of the macroblocks. By doing so, it is easy to specify the position of the synchronization word on the decoding device side without changing the processing procedure of the conventional block unit, and it is possible to reduce erroneous detection of the synchronization word due to a transmission error, and result,
Degradation of a restored image due to a transmission error can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of an image encoding device according to the present invention.

FIG. 2 is a block diagram showing an embodiment of an image decoding device according to the present invention.

FIG. 3 is a conceptual diagram illustrating a configuration of an encoded data sequence according to an embodiment of an image encoding device according to the present invention.

FIG. 4 is a block diagram illustrating a configuration of a conventional encoding device and a conventional decoding device.

FIG. 5 is a block diagram illustrating a configuration of an information source encoding unit in the encoding device.

FIG. 6 is a block diagram illustrating a configuration of an information source decoding unit in the decoding device.

FIG. 7 is a conceptual diagram illustrating a configuration of an image signal encoding block.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Synchronization word insertion interval selection part 12 Synchronization word insertion part 21 Synchronization word detection part 22 Synchronization word insertion interval judgment part 32 Synchronization word

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5C059 KK18 MA05 PP04 RB08 RC02 RF01 RF09 RF18 RF20 RF28 SS06 SS10 UA02 UA05 5K014 AA01 EA07 FA10 5K028 AA14 BB04 EE07 KK01 MM16 NN05 NN47

Claims (4)

[Claims] An encoding step of encoding image information to generate encoded data; and a synchronization word for inserting a synchronization word indicating a boundary of the image information into a boundary position of a macroblock in the encoded data. An image encoding method comprising: inserting a synchronization word into a boundary of the macroblock at a position before a predetermined data length. 2. The encoding apparatus according to claim 1, wherein the predetermined data length is represented by the number of bits of the encoded data. 3. An image decoding method, comprising: a receiving step of receiving encoded data including a synchronization word indicating a boundary of image information; and a decoding step of decoding the encoded data to restore the original image information. In the above, the synchronization word is inserted at a boundary position of a macroblock in the encoded data, and the boundary position of the macroblock at which the synchronization word is inserted is determined at a position before a predetermined data length. An image decoding method characterized in that: 4. The decoding device according to claim 3, wherein the predetermined data length is represented by the number of bits of the encoded data.