KR20040062110A - Moving picture encoder and method for coding using the same - Google Patents

Abstract

PURPOSE: A motion picture encoding system and a coding method are provided to obtain natural screen quality even when carrying out skipping of image signals for keeping a transmission bit rate in a communication system having a low bit transmission rate. CONSTITUTION: A motion picture encoding system includes an image signal receiving part(33), an image signal conversion part(35), a temporal filtering processor(36), an encoder(37), and a video encoder(39). The image signal receiving part receives image signals photographed by a camera. The image signal converting part converts the image signals of the image signal receiving part to luminance and color difference signal. The temporal filtering processor calculates and accumulates mean values for skipped image frames among image signals input from the image signal conversion part. The encoder encodes the image signals applied from the temporal filtering processor. The video encoder mixes the encoded images with other coding signals.

Description

Video encoding system and coding method {MOVING PICTURE ENCODER AND METHOD FOR CODING USING THE SAME}

The present invention relates to a video encoding system and a coding method, and more particularly, to a video encoding system and a coding method capable of obtaining a natural video in spite of a low bitstream transmission rate in a bidirectional video transmission / reception system such as IMT-2000. .

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 makes it possible to send and receive rich amounts of information, greatly contributed.

When the video signal is encoded and transmitted in the bitstream state, the transmission rate improves the efficiency encoded by the encoder and also the efficiency when decoded by the decoder.

Therefore, the bit rate of each image frame is very important factor in controlling the transmission bit rate when encoding, and has emerged as an important issue in improving image quality in encoding bit rate and transmission bit rate control.

1 is a diagram illustrating a process of signal processing in a video encoder region according to the prior art.

As shown in FIG. 1, a signal photographed through the camera 1 is an analog video signal, and the analog video signal is received by the video signal receiver 3 and then converted into a digital video signal.

The image signal receiving unit 3 converts the image signal photographed by the camera 1 into a digital image signal and then applies the image signal to the image signal converting unit 5 so that the pixels of each image frame can obtain a luminance value and a color difference value. Switch to the form that has. The analog image signals then have one digital value for each pixel of the image frame.

As described above, when the image signal photographed by the camera 1 is converted into a digital signal having a luminance value and a color difference value, it is input to the encoder 7 to perform encoding.

The encoding of the video signal is performed through discrete cosine transform (DCT), quantization, variable length encoding, etc. into one bitstream, and the audio signal included in the captured video signal is also subjected to the audio encoding process. Is compressed.

As described above, the video signal photographed by the camera is encoded for each image or voice, and then integrated in the video encoder 9 to be transmitted in the form of a bitstream to the outside.

2 is a block diagram illustrating a structure of a video encoder for controlling a transmission bit rate according to the prior art.

As shown in FIG. 2, when an image signal photographed by a camera is converted into a luminance and chrominance signal and applied to an encoder, the encoder encodes and compresses the image signal and transmits it to an external space in the form of a bitstream. .

The video signal applied from the outside performs encoding for external transmission, and also reconstructs and stores the encoded image to continuously encode the video signal applied from the outside, thereby increasing the amount of calculation and encoding efficiency.

Such a video encoder includes a DCT (Discrete Cosine Transform) 21 for discrete cosine transforming a video signal transmitted from the outside, a quantization unit 23 for quantizing a discrete cosine transformed video signal in DCT in frequency form, and Inverse quantization unit 25, inverse DCT 27, which performs inverse quantization and inverse DCT transformation, and variable length (entropy) encoding of quantized image blocks to restore the quantized image frame VLC (Variable Length Coding) 20 for transmitting a signal, a motion predictor 29 for motion prediction and compensation for a motion generated image from external video signals, and a reference picture for motion prediction And a bit rate controller 21 for adjusting and transmitting the bit rate of the image frame quantized by the quantization unit 23. .

Although shown in the figure, 22a and 22b, which are not described, represent adders for summing signals.

The video encoder having the structure described above may perform discrete cosine transform and quantization when the video signal is the first image frame, and then predict the image block within its own image frame because the reference image is not encoded. The mode encoding process is performed.

However, when the video signal to be input is a P picture having motion based on the intra-mode encoded video frame, whether to perform intra coding on each block in which motion occurs or to obtain a motion prediction value for the first coding The encoding performance is improved by appropriately selecting whether to reproduce an image using blocks in the received image frame.

Since the video frame initially encoded by the intra mode does not exist in the previous video frame, the encoding is performed after the prediction is performed in itself. Therefore, it is common that the bit rate of one image frame is very large.

An image frame following the intra image frame (I-frame) is called a P picture. Since the encoding method of the P picture is performed by motion estimation, the image frame is encoded with a smaller bit amount than the I frame.

The encoder sends the encoded bitstream at an appropriate transmission bit rate according to the channel, which is adjusted by the bit rate controller 11 to adjust this. That is, when the bit rate of the encoded video frame is large, the bit rate controller 11 reduces the encoding by increasing the quantized Q value in the unit of the image frame or the macroblock, or skips the encoded image frames. To match.

In particular, bidirectional video systems such as IMT-2000 have increased bandwidth compared to existing wireless communication networks, and video telephony can be implemented using the increased bandwidth.

When there is no motion in the video frame encoded by the encoder, the compression rate is very high and fast encoding is possible. However, when there is a lot of motion in the video frame, the coding compression rate decreases and the video bitstream is adapted to the transmission bit rate (64 kbps to 384 kbps). Will not be able to transmit.

However, video terminals such as IMT-2000, which are generally used, have a narrower bandwidth and lower bit rate of video bitstream than the Internet or LAN, so that the quantization (Q) value is increased to compensate for these drawbacks. Although the data compression ratio can be improved due to the high Q value during encoding, there is a problem in that the image quality deterioration is large.

In addition, the method of reducing the transmission frame rate by skipping the transmitted image frame has a disadvantage in that the screen quality is degraded because the displayed image is displayed in a continuous and disconnected form.

An object of the present invention is to provide a video encoding system and a coding method capable of obtaining natural image quality even if a video frame skipping process is performed to maintain a transmission bit rate in a communication system having a low bit rate such as IMT-2000. have.

1 is a diagram illustrating a process of signal processing in a video encoder region according to the prior art.

2 is a block diagram illustrating a structure of a video encoder for adjusting a transmission bit rate according to the related art.

3 is a diagram illustrating a signal processing procedure in a video encoder region according to the present invention.

4 illustrates an input / output structure of a temporal filtering processor according to the present invention.

5 is a flowchart for explaining signal processing of a temporal filtering processor according to the present invention;

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

1: camera 3: video signal receiver

5: video signal switching unit 7: encoder

9: Video Encoder 21: DCT

23: quantization unit 25: inverse quantization unit

27: reverse DCT 28: image memory

29: Motion predictor (MP)

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

An image signal receiver configured to receive an image signal photographed by a camera;

An image signal conversion unit for converting an image signal into a luminance and color difference signal from the image signal receiving unit;

A temporal filtering processor that calculates and accumulates an average value of skipped image frames among the image signals input from the image signal conversion unit;

An encoder receiving and encoding an image signal from the temporal filtering processor; And

And a video encoder for mixing the image coded by the encoder together with other coded signals.

In this case, the temporal filtering processor includes an internal frame buffer for storing a previous video frame encoded by an encoder, and the internal frame buffer uses a skipped video frame and a stored previous video frame. The average value is accumulated to reduce the high frequency component.

In addition, the video coding method according to another embodiment of the present invention,

Converting an input video signal into a luminance signal and a color difference signal;

Inputting a signal for determining whether to encode or skip the converted image signal and the image frame of the image signal to a temporal filtering processor;

When the skip signal for each image frame is applied, calculating an average value with respect to a previous image frame stored in an internal frame buffer of the temporal filtering processor for the skipped image frame; And

And in the temporal filtering processor, transmitting the accumulated average value to the encoder when an image signal for encoding is applied without being skipped next.

Here, an average value of each skipped image frame input to the temporal filtering processor is obtained and summed, and when the temporal filtering processor is initially driven, an image frame signal to be encoded is input. It is done.

In particular, when an image signal input to the temporal filtering processor is an encoded signal, the encoding is performed by directly transmitting to an encoder.

According to the present invention, a high frequency component is reduced for an image having a lot of motion in an input image signal, and a natural screen quality can be obtained despite the image skip by accumulating the previous image frame and the average value for each image frame being skipped. There is an advantage.

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

3 is a diagram illustrating a signal processing procedure in a video encoder region according to the present invention.

As shown in FIG. 3, the signal photographed by the camera 31 is an analog video signal, and the analog video signal is applied to the video signal receiver 33 and converted into a digital video signal.

The image signal receiving unit converts the digital image signal into a digital image signal and inputs the same to the image signal switching unit 35 to convert the pixels of each image frame into luminance and color difference values. The analog image signals then have one digital value for each pixel of the image frame.

In the present invention, when the image signal photographed by the camera is converted into a digital signal of the luminance value and the color difference value by the image signal conversion unit, the temporal filtering processor 36 is input before inputting it to the encoder 37. The process of reducing the high frequency component of the skipped image frame is performed and inputted to the encoder 37 to perform encoding.

The encoding of a video signal is performed through a discrete cosine transform (DCT), quantization, variable length encoding, and the like into one bitstream, and similarly, an audio signal included in the captured video signal is compressed through a speech encoding process. .

The temporal filtering processor 36 is operated before the encoder 37 operates so that the skipped image is stored in an internal frame buffer disposed therein according to whether a previous video frame is skipped. Calculate the average value of the frames.

Video frames that are not skipped are immediately input to the encoder 37 to encode the video signals. Since the first starting video frame is an intra mode, the first video frame is treated as an encoded video frame and stored in the internal frame buffer.

The video bitstream encoded by the encoder 37 is combined with other audio compression signals by the video encoder 39 and then transmitted in the form of a bitstream.

4 is a diagram illustrating an input / output structure of a temporal filtering processor according to the present invention.

As shown in FIG. 4, an internal frame buffer for storing skipped video frames is disposed inside a temporal filtering processor, and a high frequency component is generated for an encoded previous video frame that is skipped. The mean value is accumulated cumulatively to reduce.

Is Previous Frame Coded has a value of "1" if coded in the encoder and "0" if not coded in the encoder.

The temporal filtering processor determines whether the Is Previous Frame Coded value has a value of "1" or "0" while receiving a signal obtained by converting an image signal into a luminance signal and a color difference signal.

If the input image is an image skipped as "0", the image frame skipped in the internal frame buffer disposed in the temporal filtering processor and the image previously stored in the internal frame buffer The cumulative average value of frames is obtained.

However, when the input image is an image frame encoded as "1", the temporal filtering processor immediately inputs the input image to the encoder and transmits the input image to the outside in the form of a bitstream.

In addition, when the encoder first starts, since the video signal is an intra mode, that is, the previous encoded video does not exist, the encoder is set to "1" so that the encoder performs the encoding.

5 is a flowchart for explaining signal processing of a temporal filtering processor according to the present invention.

As illustrated in FIG. 5, an Is Previous Frame Coded value is determined by a temporal filtering processor to adjust a target frame rate or to store a target bit rate (S501).

If the value of the Is Previous Frame Coded value "0" indicates a value, the video frame is skipped, and thus the video frame skipped from the internal frame buffer disposed in the temporal filtering processor and the previously coded image are skipped. The average value between the frames is obtained (S502).

An average value of a video frame previously stored in the internal frame buffer and a new video frame skipped is obtained, and then stored in the internal frame buffer and transmitted to the encoder.

For example, when the Nth video frame is encoded and k video frames are skipped to encode the N + k + 1th video frame, the average value is calculated from the N + 1th video frame to the N + k + 1th video frame. Input when accumulating the N + k + 1st image frames.

The method of accumulating an average value of image frames is as follows.

(N + 1 th picture frame) / (2 ^{k + 1} ) + (N + 2 th picture frame) / (2 ^k ) + (N + 3 th picture frame) / (2 ^k-1 ) + .... The average value is accumulated in the order of (N + k th video frame) / (2 ² ) + (N + k + 1 th video frame) / (2 ¹ ).

Therefore, the pixel data of skipped image frames are added with a weight based on the image frame to be encoded.

If there is little motion in skipped image frames, the pixels of a certain screen area are hardly changed even when the average value is calculated.

However, when there are many movements of the images in the image frame, the pixels in the screen area are spread as much as they move on the screen by a temporal filtering processor.

By filtering the pixels in the portion where the motion is generated, less high frequency components are generated in a region having a lot of motion during video encoding, thus reducing the amount of bits used for encoding.

If the bit rate for the video frame is reduced, the skipped video frame will be reduced as a whole. Thus, when decoding the encoded bitstream, a much smoother image can be reproduced.

When the value of the Is Previous Frame Coded value "1" is shown, since it is determined that there is no skip for the video frame, since the first video frame or the previous video frame of the video signal is coded, it is immediately transmitted to the encoder for encoding. Proceed to step S503.

As described in detail above, the present invention has the effect of obtaining a natural image by reducing the high frequency component of the skipped image frame and then transmitting and coding the image signal to the encoder.

In addition, by reducing the high frequency component of the skipped video frame in the internal buffer of the temporal filtering processor, it is possible to reduce the amount of bits used for encoding, thereby reducing the number of skipped video frames to improve image quality. have.

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 (7)

An image signal receiver configured to receive an image signal photographed by a camera; An image signal conversion unit for converting an image signal into a luminance and color difference signal from the image signal receiving unit; A temporal filtering processor that calculates and accumulates an average value of skipped image frames among the image signals input from the image signal conversion unit; An encoder receiving and encoding an image signal from the temporal filtering processor; And A video encoder which mixes the image coded by the encoder with other coded signals; Video encoding system comprising a. The method of claim 1, The temporal filtering processor is a video encoding system, characterized in that an internal frame buffer for storing the previous video frame encoded by the encoder. The method of claim 2, The internal frame buffer is a video encoding system, characterized in that to reduce the high-frequency components by accumulating the average value by using the skipped video frame and the stored previous video frame. Converting an input video signal into a luminance signal and a color difference signal; Inputting a signal for determining whether to encode or skip the converted image signal and the image frame of the image signal to a temporal filtering processor; When the skip signal for each image frame is applied, calculating an average value with respect to the previous image frame stored in the internal frame buffer of the temporal filtering processor for the skipped image frame; And And transmitting, at the temporal filtering processor, the accumulated average value to an encoder when a video signal for encoding is applied without being skipped next. The method of claim 4, wherein And calculating an average value for each skipped image frame input to the temporal filtering processor. The method of claim 4, wherein And a video frame signal to be encoded when first driving the temporal filtering processor. The method of claim 4, wherein And if the video signal input to the temporal filtering processor is a coded signal, the video signal is directly transmitted to an encoder to perform encoding.