KR20040062109A - Method for moving picture coding - Google Patents

Abstract

PURPOSE: A method for coding motion pictures is provided to safely control transmission bit rate by advancing the picture signal coding while dividing the picture encoding bit rate to I and P frames. CONSTITUTION: A method for coding motion pictures includes the steps of carrying out intra coding of a first I frame and next P frames(S801,S802). Among macro blocks of the first I frame, those proper for transmission bit rate are intra coded in the spiral direction. In the P frames, the macro blocks for which coding is carried out centering the I frame are coded by motion compensation, and intra coding is carried out for the macro blocks extended spirally with respect to coded macro blocks. If the number of coded macro blocks is equal to that of all macro blocks in a single image frame, following image frames are coded by motion compensation by inter mode.

Description

Video encoding method {METHOD FOR MOVING PICTURE CODING}

The present invention relates to a video encoding method, and more particularly, to a video encoding method capable of maintaining a stable transmission bit rate by encoding and transmitting a video signal so as to be distributed for each video frame.

The current and future communication environments are rapidly changing beyond the distinction between wired and wireless areas and regional 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 various information in real time or comprehensively.

In addition, the development of the personal mobile communication system is not only a voice communication in the cellular phone or PCS, but also to transmit text information as well as to access the Internet wirelessly using a personal mobile communication terminal, or to watch videos on TV It is being developed to transmit the data.

In particular, a digital television system that processes a video into digital data, transmits the data in real time, receives the video, and displays the video, and the video transmitted in real time is becoming an essential element in a personal portable terminal using the IMT2000.

In the related art, the portable terminal transmits and receives only human voice. However, due to the development of multimedia and the development of digital information processing technology, various information such as voice and video can be transmitted.

Above all, this technology has been commercialized. First of all, the analog video signal is subjected to special digital processing such as quantization and variable length encoding, and then included in the digital information and transmitted. The development of video compression technology, which allows abundant information transmission and reception, contributed greatly.

Recently, with the development of digital signal processing technology, methods of compressing and transmitting a large amount of video information through a limited bandwidth transmission channel have been developed. However, if an error occurs on the transmission channel, the quality of the reconstructed image is greatly reduced. Occurs.

At this time, error concealment techniques have been studied to recover near to the original image by supplementing the lost information from the neighboring information that is normally restored without using an error correction code to maximize the limited bandwidth.

In particular, in the case of an MPEG (Moving Picture Experts Group), if an error occurs, all information until the next sync signal is found is lost. In addition, since the motion compensation encoding technique is used, the influence of the damaged portion is continued over the subsequent frames.

In MPEG or H.263, the compression method determines the motion from the previous encoded video and compensates it for encoding. In this way, the motion estimation has a large amount of computation and directly affects the image quality and the amount of bits. It is very important because it is crazy.

Therefore, many studies on motion estimation and motion compensation values have been conducted continuously. Recently, the motion estimation values are almost similar to full search using a diamond search pattern. The most widely used three step search method, which has image quality and bit rate, is used.

In particular, in a video telephone system for transmitting and receiving voice and video, such as IMT-2000, the camera mounted on the mobile terminal captures an image, encodes it, and sends it to another mobile terminal for display. The coding scheme for the circuit becomes an important signal processing scheme.

The video encoder classifies an input video signal into I frames, P frames, and B frames, and encodes the input video signals without using a time prediction method. Although there is an advantage in that encoding and decoding can be performed independently, the amount of bits for encoding one image frame is very large.

The P frame is capable of predicting a video frame currently encoded from a previously encoded video frame in time, i.e., encoding only a difference value between a macroblock to be encoded and a macroblock in a previous video frame.

Accordingly, the P frame cannot be encoded independently like an I frame, and can be normally restored only when the previous video frame is encoded or decoded.

However, P frames have a much smaller amount of encoded bits than I frames.

The B frame performs prediction from a video frame located before and a video frame located next in time. Although the B frame can compress the video more efficiently than the I frame or the P frame, it is generally not used because of the complexity and time potential in the portable terminal or real time video call.

Therefore, in general, for video telephony, the encoder starts with I frames and proceeds with encoding in order of P frames, compresses them, decodes them, and displays them.

1 is a view showing a video encoder system according to the prior art.

As shown in FIG. 1, when an analog video signal photographing a real object is received, a digital conversion is performed, and a coding operation is performed to encode the same. The video signal converted into a digital video signal is subjected to discrete cosine transform in DCT (Discrete Cosine Transform: 1), and the video signal converted into frequency form is quantized in quantization unit 3 and then VLC (Variable Length Coding: 10). ) Is transmitted to the outside through entropy process.

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. The dequantized image data is inversely quantized by the inverse quantization unit 5 together with an entropy process for transmitting an encoded image to the outside to estimate the motion from the image frames, and then inverse discrete cosine in the inverse DCT 7. The reconstructed image is stored in the image memory 8 by the conversion.

If the video to be encoded is not the first video and is not stored in the video memory 8, encoding is performed for intra prediction, and motion coding is performed for P frames by macroblock of the video frame. Proceed.

That is, the encoder determines the first input image as an intra image frame, performs a prediction by encoding in the image frame itself, and then encodes the intra image frame, and then the input image frame is a previously encoded intra image frame. The motion is encoded by estimating the motion from.

In this process, the decoder performs decoding in the same manner as the encoder to reproduce the screen.

2 is a diagram illustrating image sequences of an encoding according to the prior art.

As illustrated in FIG. 2, the encoded image frames are displayed as one sequence in which P frames are successively sequentially from I frames over time.

The video captured by the camera is transmitted to the outside in the form of a bitstream after encoding of each video frame is performed by the encoder. In this case, since an error may occur in the transmission channel, P frames may continue to come after the I frame, but an I frame image is inserted in the middle due to an error to be generated.

Since the I frame is decoded by the decoder and motion compensation is performed on the P frame later, image quality may be degraded due to error accumulation when an error occurs. Therefore, the error is not accumulated by inserting an I frame in the middle.

3 is a graph illustrating a transmission bit rate per frame when the QP value is 13 according to the related art.

As shown in FIG. 3, the screen size is QCIF (176 * 144) in a system for transmitting an image through a portable terminal. When QP is 13 without adjusting the transmission bit rate, the encoded image frames The transmission bit rate is shown.

That is, the transmission bit rate is 18000 bits / frame in the first I frame, and the transmission bit rate is constant between 4000 and 6000 bits / frame for the P frame.

As described above, in the case of an I frame, since the encoding is performed in itself without any other reference image frame, the amount of bits per image frame is large, and in the case of a subsequent P frame, the encoding information of the I frame is used. The amount is relatively small.

However, as described above, the encoder performs encoding on the I frame and encodes the video frame following the next P frame. However, in order to confirm an error during transmission, the encoder is periodically inserted to encode the I frame. Proceed.

In the video terminal, since the bit rate transmitted per channel is not large, transmission bit rate adjustment becomes an important factor for implementing a good quality image. As described above, transmission is performed with a large deviation from P frame transmission due to the insertion of an I frame. There is a problem that the transmission bit rate is unstable.

4A and 4B are graphs showing the transmission bit rate per frame when the QP value is 25 according to the prior art. As shown in FIG. 4A, when the QP value is increased, the high frequency component is reduced to I frame. Although the corresponding bit amount can be reduced, the image quality is greatly degraded due to the increase in QP.

When the QP value is increased from 13 to 25 for the first I image frame, it can be seen that the bit amount of the image frame is greatly reduced from 18232 bit to 1105 bit.

However, as shown in FIG. 4B, the Luminance PSNR value indicating the picture quality is significantly different from the Luminance PSNR value to 28.5 when the Luminance PSNR value is 31.5 to QP 25 when the QP is 13 for the first I frame. It can be seen that the image quality is significantly lower than that of the P frames.

FIG. 5 is a graph illustrating a transmission bit rate per frame when skipping an image frame according to the prior art. As shown in FIG. 5, an I frame is used to adjust a transmission bit rate while maintaining a high quality image quality while maintaining a low QP value. After encoding the video frame, the video frame is skipped without encoding the subsequent video frame until the amount of bits used in the I frame meets the required average transmission bit rate.

As described above, the method for adjusting the transmission bit rate can maintain the high quality of the I frame, but there is a problem in that the displayed screen is unnaturally continued due to skipped image frames.

As shown in the figure, since seven P frames are skipped after the first I frame and then P frames are output, the average transmission bit amount can be maintained, but it is difficult to obtain a natural image.

An object of the present invention is to provide a video encoding method capable of safely controlling a transmission bit rate by performing encoding while dividing an image encoding bit rate into I frames and P frames when encoding a video signal.

1 illustrates a video encoder system according to the prior art.

2 is a diagram illustrating image sequences encoded by a conventional technique.

3 is a graph showing a transmission bit rate per frame when the QP value is 13 according to the prior art.

4A and 4B are graphs showing a transmission bit rate per frame when the QP value is 25 according to the prior art.

5 is a graph illustrating a transmission bit rate per frame when skipping an image frame according to the related art.

6 is a diagram illustrating image sequences of an encoder according to the present invention.

FIG. 7 is a diagram illustrating an intra macroblock spirally spread within a P frame according to the present invention. FIG.

8 is a flowchart illustrating a process of encoding in a spiral direction 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

10: VLC

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

Performing intra encoding on the central macroblocks of the first I frame with respect to the input image signal;

Encoding macroblocks in a helical direction in sequence with respect to a P frame after the first I frame;

Determining whether the number of macro blocks to be encoded is equal to the total number of macro blocks in one image frame; And

If the number of macroblocks to be encoded is equal to the total number of macroblocks of one video frame, encoding the subsequent video frames with motion compensation according to an inter mode; Characterized in that it comprises a.

Here, the macroblocks of the P frames that are encoded after the first I frame are simultaneously subjected to the motion compensation by the inter mode and the prediction encoding by the intra mode. In the P frame, the macro blocks are encoded in the first I frame. The macroblock may be encoded by motion compensation.

In the P frame, encoding of macroblocks of a new portion extended from the macroblock encoded in the first I frame is performed by intra mode, and encoding of the first I frame and the P frame is not performed. The blocks are forcibly encoded in a single color so that they are not used as reference blocks when encoding is performed in a P frame.

According to the present invention, the bit rate of an I frame having the largest encoding bit rate is distributed to P frames that are subsequently encoded, thereby stabilizing the transmission bit rate to prevent a sudden decrease in bit amount after the I frame.

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

6 is a diagram illustrating encoding image sequences according to the present invention.

As shown in FIG. 6, when encoding an image frame, the first I frame and subsequent P frames are sequentially encoded to adjust the bit amount.

That is, among macroblocks constituting the first I frame, encoding is sequentially performed from the center to the spiral. Macro blocks that are sequentially encoded are extended along P frames and continue up to a certain number of P frames. Preferably, the image frames are sequentially extended from the center to the spiral direction until the number of macro blocks corresponding to one image frame is reached.

When encoding is performed as described above, a large amount of bits of an I frame is distributed to P frames, thereby reducing transmission bit rate variation between image frames as a whole to maintain stable transmission bit rate.

From the user's point of view, the screen is reproduced by expanding from the center of the image frame, so that visually natural images can be seen.

FIG. 7 is a diagram illustrating an intra macroblock spirally spread within a P frame according to the present invention.

As shown in FIG. 7, the encoding proceeds in a spiral shape sequentially from the center according to the P frames starting from the first I frame.

According to the transmission bit rate, the number of macro blocks that spirally expand in I frames and P frames can be adjusted.

In particular, in the P frame, intra encoding and inter encoding are performed at the same time. The encoding is performed at the center portion, and then the intra encoding is performed on the macroblock extended in the P frame, and the motion compensation is performed on the macroblock that has been previously encoded. Do it.

That is, intra coding continues to encode macro blocks extending in a spiral until the number of macro blocks included in one image frame is processed, and encoding is performed by motion compensation on macro blocks that have already been encoded. do.

That is, intra coding is performed on the number of macro blocks suitable for the transmission bit rate with respect to the first image frame (I frame), and then, in the subsequent P frame, the intra coding is extended in a spiral direction from the macro block that has been encoded in the previous I frame. Intra-encode macro blocks. However, encoding is performed by performing motion compensation on the macroblocks subjected to encoding in the I frame.

The number of macro blocks to be newly intra coded in the P frame is similarly determined to match the transmission bit rate.

It is most preferable that the number of coded macroblocks being sequentially expanded from P frames is equal to the total number of macroblocks existing in one displayed screen.

When encoding the I frame and the P frame in sequence, the encoding process is stopped when the encoding proceeds, that is, when the number of macroblocks spirally expanded exceeds the total number of macroblocks existing in one image frame. As in general P frame encoding, motion compensation is performed to perform encoding.

8 is a flowchart illustrating a process of encoding in a spiral direction according to the present invention.

As shown in FIG. 8, intra-encoding of macroblocks suitable for the transmission bit rate is performed in the spiral direction on macroblocks of the first I frame with respect to the input image signal (S801).

When intra encoding is performed on the central macroblocks of the first I frame, encoding is performed on the P frames, and then encoding is performed by motion compensation on the macroblock where the encoding of the I frame center is performed. Intra encoding is performed on the macroblocks spirally extended around the macroblock (S802).

In other words, intra-encoding is performed on macro blocks that are sequentially expanded, and encoding according to motion compensation is performed on macro blocks that have been previously encoded.

When the macroblocks of P frames are sequentially encoded from the first I frame, the number of encoded macroblocks and the total number of macroblocks in the screen are the same, or when the macroblocks are collectively expanded by a certain number in the screen. Intra coding and inter coding are simultaneously performed until the total number of macroblocks is greater (S803).

If the number of encoded macroblocks for the I frames and the P frames is equal to or larger than the total number of macroblocks in the screen, it means that the encoding process is started to begin. Therefore, the encoding is performed while the motion compensation is performed. (S804).

However, if the number of macroblocks coded for I frames and P frames is smaller than the total number of macroblocks in the picture, this indicates that the encoding has not been performed for one picture yet. In the region, intra encoding is performed, and motion compensation is performed on the macroblock in which encoding has been performed (S802).

In addition, when the encoding is performed on the gray part while performing intra encoding inter encoding in the P frame at the same time, the encoding bit is forcibly encoded in the region other than the macro block where encoding is performed at the center of the I frame. This is because the amount is increased.

When encoding the video signal, there is an advantage in that a natural image can be obtained by distributing the encoding bit amount to the I frame and the subsequent P frames in order to reduce the transmission bit amount variation generated due to the large bit amount of the I frame.

As described in detail above, the present invention has an effect that a stable transmission bit amount can be realized by sequentially encoding a plurality of P frames in a spiral direction in order to distribute the amount of encoding I frames. .

In addition, the size of the transmitted bit amount does not vary greatly between the I frame and the P frame, which is an efficient advantage in a circuit switch network system requiring a constant transmission bit rate.

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 (5)

Performing intra encoding on the central macroblocks of the first I frame with respect to the input image signal; Encoding macroblocks in a helical direction in sequence with respect to a P frame after the first I frame; Determining whether the number of macro blocks to be encoded is equal to the total number of macro blocks in one image frame; And And if the number of macroblocks to be encoded is equal to the total number of macroblocks in one video frame, encoding subsequent motion picture frames with motion compensation in inter mode. Way. The method of claim 1, The macroblocks of the P frames that are encoded after the first I frame are simultaneously subjected to the motion compensation in the inter mode and the prediction encoding in the intra mode. The method of claim 1, And the P frame is encoded by motion compensation on the macro block encoded in the first I frame. The method of claim 1, In the P frame, the video encoding method according to the intra mode is performed on the macro blocks of the new portion extended from the macro block encoded in the first I frame. The method of claim 1, The video encoding method of claim 1, wherein the first I frame and the P frame that are not encoded are forcibly encoded in a single color so that they are not used as reference blocks when encoding is performed in the P frame. .