KR100543607B1 - Method for decoding of moving picture - Google Patents

Abstract

The present invention discloses a video decoding method that can efficiently decode in spite of an error caused by transmission of compressed video data in a wireless communication network such as IMT-2000. The disclosed invention includes determining a picture type from picture header information of a picture frame decoded from a compressed video bitstream; Determining the presence or absence of GOB header information in a decoded video frame to determine whether an error occurs in the picture type information; If there is the GOB header information, it is determined whether the GFID values in the GOB header information have the same GFID value, and if there is no GOB information, the information of the decoded macroblock type is determined to determine whether it is an intra mode. Determining whether it is in inter mode; Determining whether the GIFD value in the GOB of the decoded image frame has the same value as the GFID value in the GOB of the previous image frame; And if the GFID value in the GOB of the previous image frame and the GIFD value in the GOB of the decoded image frame have the same value, correcting and decoding the picture type of the image frame decoded into the picture type of the previous image frame. It is characterized by including.

Decoding, Compression, Movie, Error, GOB, Picture Information

Description

How to decode video {METHOD FOR DECODING OF MOVING PICTURE}

1 is a diagram illustrating a video decoder system according to the present invention.

2 illustrates a video decoder system according to the prior art.

3 is a diagram illustrating header information of a picture frame generally coming in a bitstream state.

4 is a diagram showing GOB header information of a picture frame used in the present invention.

5 is a flowchart for explaining a video decoding processing method according to the present invention;

* Description of the symbols for the main parts of the drawings *

1: DCT 3: Quantizer

5: Inverse quantization department 7: Inverse DCT

8: Image Memory 9: Motion Predictor (MP)

10: Variable Length Coding (VLC) 20: Motion Vector Compensator

* GSTUF {Group Stuffing (Variable length)}

* GBSC {Group of Block Start Code (17 bits)}

* GN {Group Number (5 bits)}

GSBI {GOB Sub-Bit stream Indicator (2 bits)}

* GFID {GOB Frame ID (2 bits)}

* GQUANT {Group of Quantizer Information (5 bits)}

The present invention relates to a video decoding method, and more particularly, to a video decoding method capable of improving decoding efficiency and preventing image quality degradation by determining and correcting a picture type error due to a channel error in wireless communication using IMT-2000. It is about.

The current and future communication environment is rapidly changing beyond the distinction between wired and wireless areas, and between regions and countries.In particular, future communication environments such as IMT-2000 require users as well as video and audio. The trend is to build an environment that provides a variety of information in real time, comprehensive.

In addition, the development of a personal mobile communication system is not only a voice communication even in a cellular phone or a PCS, but also transmits text information as well as wirelessly accesses the Internet using a personal mobile communication terminal to acquire various information. Reached to the extent possible.

Moreover, the digital television system that processes video into digital data, transmits it in real time, and receives and displays it is one step further to transmit only video (audio), and to maximize the advantages of digital information. A series of studies and their achievements are visible.

In particular, the research on compressing and processing video signals into digital data, and real-time transmission of high speed and high quality at higher speed and richer information volume has already been made, and the commercialization is expected to be made at a considerable level. In addition, the transmission and reception of such digital video information is being provided to terrestrial broadcasting systems (sometimes cable broadcasting systems) including satellite broadcasting.

1 is a diagram illustrating a video decoder system according to the present invention.

As shown in FIG. 1, when an analog video signal photographing a real object comes in, a digital conversion thereof is performed, and an encoding operation is performed to compress it. The digital video signal is discrete cosine transformed by DCT (Discrete Cosine Transform 1), and the image signal converted into frequency form is quantized by the quantization unit 3 and then entropyized by VLC (Variable Length Coding 10). Is sent to the outside.

In this case, when the encoded digital image is in the inter mode, a motion vector capable of predicting the previous image is generated from a motion predictor 9. An entropy process for transmitting the compressed image by decoding so as to estimate the motion from the image frames and the quantized image data are inversely quantized by the inverse quantization unit 5, and then inverse discrete cosine in inverse DCT 7. The reconstructed image is stored in the image memory 8 by the conversion.

The next input image is defined as a P picture as an inter mode, motion prediction and compensation are performed on blocks of each video frame, and then the error signal is encoded in the same manner as the encoding of macro blocks in the intra mode.

2a and 2b which are shown in the figure but are not described represent adders.

2 is a diagram illustrating a video decoder system according to the prior art.

As shown in FIG. 2, when the image data encoded by the encoder enters the decoder unit in the form of a bit stream, decoding is performed as opposed to encoding performed by the encoder in order to restore the original image.

Therefore, the decoder for decoding the compressed bitstream includes a variable length decoder (VLD 11) which temporarily stores a certain amount of data of a bitstream and performs decoding to reproduce the stored bitstream as an image. Error detection / restoration unit 13 for error detection, motion vector restoration and image reproduction in the bitstream decoded by the variable length decoder 11, and image for image reproduction in the error detection / restoration unit 13 Inverse Quantization (15), Inverse Discrete Cosine Transform (17), and Error Detection / Restoration (DCT), which perform inverse quantization and inverse discrete cosine transformation, in order to reverse the compression scheme performed by the encoder. A motion vector compensator (MC 20) for compensating the motion vector reconstructed in step 13) and a display unit 25 for displaying a video. It is composed of the video frame memory 21 for storing images.

The process of reproducing the compressed video by the decoder having the above structure is as follows.

When an analog video signal is received by a decoder in a compressed bit stream, a variable frame coded video frame signal is subjected to a discrete cosine transform and a quantization at an encoder. First, it is stored in the buffer to temporarily store it. The compressed video bitstream stored in the buffer converts a value, a length, and the like of the compressed bitstream by the variable length decoder 13 into two-dimensional encoding, and then the error detection / restoring unit 13 for error detection and countermeasure. ) Detects motion vector values and errors.

Then, the compressed image applied from the error detection / recovery unit 13 is reconstructed in the opposite direction as when the encoder compresses the image. The inverse quantization unit 15 performs inverse scanning and inverse quantization, and then performs inverse discrete cosine transformer (IDCT) 17 to perform IDCT conversion.

The video signal, which has been subjected to the Inverse Discrete Cosine Transform, is input to the adder 18. In the adder 18, the reconstructed image frame is an intra mode image that is first started, or the previous image frame. After determining whether the existing inter mode image exists, the display unit 25 directly reproduces the image because there is no motion vector in the case of the intra mode image.

In case of determining the inter mode, since the motion vector exists, the motion vector compensator 20 extracts the motion vector from the compressed bitstream and compensates for the motion vector. Send a vector.

The motion vector compensator 20 obtains and decodes the difference of the motion vectors extracted by the error detection / recovery unit 13 from the image frame memory 21 storing the previous image frame reproduced by the display unit 25. To compensate the video.

Therefore, when the motion vector compensator 20 transmits the difference obtained by subtracting the current motion vector value from the previous video frame to the adder 18, and the compressed bitstream is an inter mode in which the previous video frame exists. The image is reproduced by adding the inverse DCT value.

The reconstructed image frame is divided into macroblock units, and the data of each macroblock includes a motion vector while being decoded, and these motion vectors indicate a predetermined direction. Video compression technology uses the motion vector to take the macroblock of the previous video frame and restore the current video frame to reduce the compression rate.

3 is a diagram illustrating header information of a picture frame generally coming in a bitstream state.

As shown in FIG. 3, the video coded by the encoder has picture header information, GOB information, macro block information, and the like as follows.

PSC {Picture Start Code (22 bits)} as a code indicating the beginning of a compressed bitstream video, TR {Temporal Reference (8 bits)} indicating a time-series reference number of each video frame, and a decoded video. PTYPE {Picture Type Information (Variable Length)} indicating information that can identify a picture type of a frame, PQUANT {Quantizer Information (5 bits)} indicating information of a picture frame quantized by an encoder, and the current multi-point being continued. Continuous Presence Multipoint and Video Multiplex (1 bit)} for video multiplex, Picture Sub-Bitstream Indicator (2 bits)}, and TRB for time reference for B picture in P picture and B picture frames {Temporal Reference for B-pictures in PB-frames (3/5 bits)}, and DBQUANT {Quantization information for B-pictures in P-picture and B-picture frame] PB frames (2 bits)}, PEI {Extra Insertion Information (1 bit)} which specifies whether there is user data thereafter as a picture layer, and PSUPP {Supplemental Enhancement Information (0/8/16. .. bits)}, GOB (Group Of Black) and ESTUF {Stuffing (Variable length)}, EOS {End Of Sequence (22 bits)}, PSTUF {Stuffing (Variable length) )}.

Based on the picture header information of the compressed video, decoding is performed in the decoder system of FIG. 2 to reproduce the compressed video in the encoder.

However, in the real-time video transmission through the IMT2000 wireless network, this causes a problem that the information (PTYPE) of the picture type of the actual encoded video is changed.

In particular, PTYPE, which represents picture type information of picture header information, has a bit error, where 0 is an I frame picture and 1 is a P frame picture, where 0 is 1 or 1 is 0. The picture type of the input bitstream is judged differently.

As a result, there is a problem in the decoder that the wrong picture type causes the wrong processing result and the quality loss.

The present invention detects and corrects a picture type error by using a picture header information, GOB header information, and macro block information inserted into an image frame during decoding, by detecting a channel error occurring in a transmission / reception process of a video compression bitstream. The purpose is to provide a video decoding method that can improve efficiency and improve image quality.

In order to achieve the above object, a video decoder method according to the present invention,

Determining a picture type from picture header information of an image frame decoded from the compressed video bitstream;

Determining the presence or absence of GOB header information in a decoded video frame to determine whether an error occurs in the picture type information;

If there is the GOB header information, it is determined whether the GFID values in the GOB header information have the same GFID value, and if there is no GOB information, the information of the decoded macroblock type is determined to determine whether it is an intra mode. Determining whether the mode is inter mode;

Determining whether the GIFD value in the GOB of the decoded image frame has the same value as the GFID value in the GOB of the previous image frame; And

If the GFID value in the GOB of the previous image frame and the GIFD value in the GOB of the decoded image frame have the same value, correcting and decoding the picture type of the image frame decoded into the picture type of the previous image frame; Characterized in that.

Here, when the macro block type information is the inter mode, the decoded video frame is corrected to a P picture frame.

If the macroblock type information is an intra mode, the decoded video frame is corrected to an I picture frame, and if the GFID values of GOB information present in the decoded video frame are not the same mode, the same GFID value is many. The GFID value of the GOB information is corrected, and the GFID value in the GOB of the decoded image frame is compared with the GFID value in the GOB of the previous image frame. In other cases, the picture type of the picture header information is set to the picture type of the previous image frame. And decoding by correcting the picture type opposite to the information.

According to the present invention, in decoding processing due to a picture type error due to a channel error in an IMT-2000 wireless network, not only the picture type information in the picture header information but also the picture type information using the GFID value and the macro block information in the GOB header information. By correcting this, there is an advantage of improving image quality and minimizing errors that may occur during decoding.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

4 is a diagram illustrating GOB header information of a picture frame used in the present invention.

As shown in FIG. 4, in the video bitstream compressed by the encoder, a plurality of video frames continue in the time series direction, and picture header information is inserted in each video unit so that the video bitstream is constantly displayed in the compressed direction when decoded. .

In general, when an image comes in a bitstream form, one unit video includes a plurality of image frames, and picture header information, as described with reference to FIG. 3, is inserted into each image frame.

Here, the image frame is divided into a plurality of GOBs in the vertical direction, respectively. GOB information of the image frame is composed as follows.

A plurality of macro blocks in the direction are constituted by another small unit of information of picture header information of an image frame. GSTUF {Group Stuffing (Variable length)}, GBSC {Group of Block Start Code (17 bits)}, GN {Group Number (5 bits)}, GSBI {GOB Sub-Bit stream Indicator (2 bits)} GFID {GOB Frame ID (2 bits)}, GQUANT {Group of Quantizer Information (5 bits)} and a macro block. Although not clearly shown in the drawing, the macro block is divided into four blocks again and has two color difference values in the longitudinal direction.

 In the present invention, a picture type is determined using picture header information in a decoding operation, and the GOB information of each image frame is referred to to determine whether the picture type is information changed by an error.

In other words, instead of using only the data (PTYPE) corresponding to one by using error correction, image quality by more accurately grasping and correcting an error generated by using data corresponding to each of two data (PTYPE, GFID information) To improve.

Therefore, among the picture header information included in the video compression bitstream, the GFID exists in the GOB. The GFID represents a frame ID in the GOB, and has a fixed length of 2 bits. The GFID in the GOBs in one image frame All the same.

Therefore, if the picture type {PTYPE: Picture Type Information (Variable Length)} of the previous video frame is the same as the current video frame to be decoded, the picture type of the picture header should be the same, and the GFID value in each GOB should be the same. If the picture type is the same but the GFID value in each GOB is different, a channel error occurs and the picture type is changed and it is determined based on the GFID and corrected.

If there is no GOB header information, since there is no data for comparing and determining picture errors, it is determined whether intra mode or inter mode is discriminated when decoding a macro block that follows.

5 is a flowchart for explaining a video decoding processing method according to the present invention.

As shown in FIG. 5, when the compressed bitstream is applied to the decoder and decoded, decoding is performed for each video unit. When a plurality of image frames to be decoded are consecutive in time series and decoding such image information, in addition to macro block information of an image frame including luminance and chrominance signals, picture type of each image frame, respective GOB information, macro block information, etc. Are inserted together.

Accordingly, the picture header information is determined to determine whether the video frame of the applied bitstream is an intra mode video frame that is initially started or whether the previous video frame is an image frame within a decoded unit (S501, S503).

However, since a video bit stream transmitted and received wirelessly is introduced into the decoder through multiple paths, loss of video data or channel bit error is inevitable. That is, if the picture type has a value of '0' in the encoder, and an error occurs during the multi-path, and is converted to '1', the image to be encoded is decoded and the image to be decoded causes image quality damage.

However, after the picture type information is obtained from the picture header information, it is determined whether there is GOB information (S505). If there is GOB header information, it is determined whether all the GFID information of the GOB of the video frame to be decoded are the same (S506, S507). . In principle, data is stored so that the GFID values of GOBs in the same image frame have the same value.

If the GFID values of the image frame GOB to be decoded are not the same, each GFID should be compared and corrected equally (S508). It is determined that having a large number of identical values among a plurality of GFIDs is an error-free GFID value.

Then, the GFID value of the previous image frame GOB stored in the image frame memory in the decoder and the GFID value of the currently decoded image frame GOB are compared to determine the same picture type even when the picture type value is different ( S510).

That is, in this case, it is determined that an error occurs in picture type information due to an error.

When the GFID value of the previous video frame GOB is compared with the GFID value of the currently decoded video frame GOB, it is determined that the picture types are different from each other. Therefore, even when the picture type information is the same, it is determined as another picture type (S511). ).

In addition, if there is no GOB header information in the image frame to be decoded after determining the picture type information, there is no data for judging whether there is an error (S520). If the macroblock is in the inter mode, it is determined that a previous video frame (P picture) exists and is corrected to the inter mode if the picture type is the intra mode (S523).

In addition, when the macroblock value is not the inter mode, it is determined as a picture type of the intra mode (B picture) and corrected even when the picture type is the inter mode (S525).

As described above, in the present invention, the picture type error due to the channel error can be minimized by not only determining the picture type by looking at the picture type of the picture header information, but by using the internal GOB information.

As described in detail above, the present invention provides picture header information in which a picture type error, which may occur due to a channel error, is inserted into each bitstream when decoding compressed video information in a field using a wireless channel such as IMT2000. The image quality can be prevented by compensating for the error by using the GOB information, the macro block information, and the like.

In addition, there is an advantage that can be stably decoded by diagnosing and correcting an error by comparing image information in a situation where an error frequently occurs randomly.

The present invention is not limited to the above-described embodiments, and various changes can be made by those skilled in the art without departing from the gist of the present invention as claimed in the following claims.

Claims (6)

Determining a picture type from picture header information of an image frame decoded from the compressed video bitstream; Determining the presence or absence of GOB header information in a decoded video frame to determine whether an error occurs in the picture type information; If there is the GOB header information, it is determined whether the GFID values in the GOB header information have the same GFID value, and if there is no GOB information, the information of the decoded macroblock type is determined to determine whether it is an intra mode. Determining whether an inter mode; Determining whether the GIFD value in the GOB of the decoded image frame has the same value as the GFID value in the GOB of the previous image frame; And If the GFID value in the GOB of the previous image frame and the GIFD value in the GOB of the decoded image frame have the same value, correcting and decoding the picture type of the image frame decoded into the picture type of the previous image frame; Video decoding method characterized in that. delete The method of claim 1, And when the macroblock type information is the inter mode, correcting the decoded video frame to a P picture frame. The method of claim 1, And when the macro block type information is an intra mode, correcting an image frame to be decoded to an I picture frame. The method of claim 1, If the GFID values of the GOB information present in the decoded image frame are not the same mode, the video decoding method characterized in that for correcting a large number of the same GFID value to the GFID value of the GOB information. The method of claim 1, Comparing the GFID value in the GOB of the decoded image frame with the GFID value in the GOB of the previous image frame, in another case, correcting the picture type of picture header information to a picture type opposite to the picture type information of the previous image frame and decoding the same. Video decoding method characterized in that.

Images