KR0119126B1 - 45 Mb/s VIDEO CODER/DECODER FOR SUPPORTING SEVERAL MODES - Google Patents

Abstract

A method for improving the image quality of an image signal processing apparatus using an MPEG format is disclosed. The image signal processing apparatus comprises a memory(7) for storing the image data compressed by an MPEG type; an MPEG image decoder(1) for decoding the compressed image bit stream data stored at the memory(7) to provide each of 8-bit RGB image data; a synchronization signal generator(2) for generating a horizontal and a vertical synchronization signals and a pixel clock signal; a digital signal processor(3) for processing a high speed operation of the decoded RGB image data according to the synchronization signals provided from the synchronization generator(2); a digital-to-analog converting unit(6) for converting the output from the digital signal processor(3) to an analog signal; and a ROM(4) and a RAM(5) for storing a number of predetermined operating programs. According to one of the predetermined operating programs stored at ROM(4), in case that the similarity of one frame is larger than a predetermined level, the bypass is processed, and otherwise, a lowpass filtering is performed at the horizontal and the vertical directions of boundaries of the macro-block. Since, thereby, the similarity between blocks is calculated, the blocking effect can be reduced.

Description

Image Quality Improvement Method in Video Equipment Using MPEG Method

1 is a schematic block diagram of a video apparatus using a conventional MPEG method.

2 is a block diagram of a video device using the MPEG method according to the present invention.

3 is a flowchart for explaining the image quality improving method according to the present invention.

4 shows a format in which moving pictures are compressed on a recording medium such as a video compact disc.

* Explanation of symbols for main parts of the drawings

1: MPEG image decoder 2: Synchronization generator

3: digital signal processor 4: ROM

5: RAM 6: Video D / A Converter

7: recording medium 10: pixel

20: Video macroblock

The present invention relates to a method for improving image quality in a video device using an MPEG method. In particular, the present invention relates to a method for improving image quality in an video device using an MPEG method, by providing a correlation calculation unit to significantly improve the image quality of a video device monitor. It is about.

Recently, the MPEG (Movig Picture Expert Group) method has been proposed as a standard for video high-efficiency coding method for digital storage media. In this standard, not only forward playback but also special functions such as random access and high speed playback are considered.

The storage medium targeted by the MPEG has a continuous transmission speed of about 1.5 Mb / s like a compact disc. The storage medium is not only directly connected to a decoder, but also via a transmission medium such as a computer bus and a telecommunication network. May be connected.

In the MPEG type high efficiency coding method, DCT (Discrete Cosine Transform) and variable length coding are first used to perform motion compensation to reduce the redundancy in the time axis direction, to take the difference between the images, and to reduce redundancy in the spatial axis direction.

In other words, in a continuous moving picture, the image before and after and the current image are almost similar, so that the difference between the image to be encoded and the front image is different from the image of the front or the image is temporally forward or rearward or backward and forward in the bidirectional prediction image. The difference between the interpolated image and the interpolated image is encoded, and the difference is encoded and transmitted to reduce the amount of information transmitted by reducing redundancy in the time axis direction.

DCT processing is performed on a block basis without transmitting the difference of the image data obtained by the above method. The reason for this is that in the case of a natural background taken by a video camera, for example, a background color such as a sky, spatial redundancy is increased. If used, the amount of data can be further reduced.

Fig. 1 is a block diagram of a video apparatus using the conventional MPEG method, which is a recording medium such as a video compact disk (hereinafter abbreviated as video CD) in which video data is compressed in the MPEG method; An MPEG video decoder which decodes the video bitstream data compressed in block units on the recording medium and outputs 8-bit video data R, G, and B, a synchronization generator for generating horizontal and vertical synchronization signals and a pixel clock signal; And an image D / A changer for converting and outputting digital image data output from the MPEG image decoder into an analog signal.

Next, the operation of the apparatus configured as described above will be described.

First, as shown in FIG. 4, video data is compressed by using an MPEG method with 8x8 pixels as one block, and four blocks are defined as one macroblock (that is, 16x16 pixels). MPEG video decoder outputs the video data stored in macroblock units and decodes the horizontal and vertical synchronous signals generated by the synchronization generator and the pixel clock signal to create a moving picture. The / A converter converts the digital video data R, G, and B of the moving picture produced by the MPEG decoder into an analog video signal for output.

However, when decoding the compressed video in the block unit by simply applying MPEG as described above, some data loss occurs in the compression process, so that the chromaticity difference between the pixels of the boundary line between the macroblock and the macroblock becomes significant. There is a problem in that a boundary line can be seen on the screen by a blocking effect generated between the blocks, and more noticeable when the motion of the image is severe.

Accordingly, the present invention has been made in view of the above circumstances, and an object of the present invention is to provide a method for improving image quality in a video device using an MPEG method that provides a correlation calculation means to significantly improve the picture quality of a video device monitor. There is this.

According to the method for improving the image quality of a video device using the MPEG method of the present invention to achieve the above object, there is provided a recording medium (7), such as a video CD in which video data is compressed in the MPEG method, and the recording medium (7). MPEG video decoder 1, which decodes the compressed video bitstream data in block units and outputs 8-bit video data R, G, and B, and a synchronization generator 2 for generating horizontal and vertical synchronization signals and pixel clock signals. A digital signal processor (3) for performing high-speed operation according to a synchronization signal generated by the synchronization generator (2) by receiving image data R, G, and B decoded by the MPEG decoder (1). An image D / A converter 6 for converting and outputting digital image data output from the processor 3 into an analog signal, and a device having a ROM 4 and a RAM 5 in which a program for performing a predetermined operation is stored. In the ROM (4) The image quality improvement method is provided so that the correlation of one frame is larger than the reference level and bypassed. If the correlation is smaller than the reference level, the macroblock economy in the horizontal and vertical directions is low pass filtered.

Here, the low pass filter processing method multiplies the filter tap weights corresponding to the two pixels closest to the first block boundary among the two macroblocks and the first pixel of the next block, and then adds the multiplied three values. Compensation for the pixel closest to the original block boundary is performed by dividing by the total weight, and then multiplying the corresponding filter tap weights by the two pixels of the next block and the one pixel closest to the original block boundary, respectively. By multiplying the multiplied three values and dividing by the total weight, the pixel closest to the next block boundary is compensated.

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

2 is a block diagram of a video device using the MPEG method according to the present invention. In the drawing, reference numeral 1 denotes synchronization after receiving and decoding video bit stream data compressed in units of blocks. MPEG video decoders for outputting video data R, G, and B each having 8 bits according to the synchronization of the unit 2, 2 is a synchronization generating unit for generating a horizontal synchronization signal, a vertical synchronization signal, and a pixel clock signal; A digital signal processor which receives RGB data decoded by the MPEG decoder 1 and performs high-speed operation according to the synchronization generated by the synchronization generator 2, 4 is a ROM in which a correlation processing method is programmed, 5 is a RAM, 6 is an image D / A converter which converts and outputs digital image data calculated and output from the digital signal processing unit 3 into an analog signal, and 7 is a recording medium such as a video CD in which image data is compressed. (7) it shows, respectively.

Next, the operation of the present invention will be described in more detail with reference to the flowchart of FIG. 3 and FIG.

First, as shown in FIG. 4, when moving picture data compressed and stored in the recording medium 7 is inputted to the MPEG video decoder 1 using a pixel having a ratio of 16 x 16 as a width and height as one macroblock, The MPEG decoder 1 decodes the moving image data and outputs the digital signal processing unit 3 as 8-bit moving image data R, G, and B in accordance with the vertical / horizontal synchronization signal and the pixel clock signal applied by the synchronization generating unit 2. To be sent. Then, the digital signal processing section 3 is stored in the ROM 4 on the basis of the moving picture data R, G, and B of each macroblock that are continuously input in accordance with the synchronization signal of the synchronization generating section 2, High-speed computation is performed according to the correlation calculation program. If the correlation is large when the correlation of one frame is calculated, the chromaticity difference between pixels is small when the correlation is large, and when the correlation is small, between pixels Since the chromaticity difference is large, the correlation is calculated to determine which chromaticity difference between the pixels, that is, one frame, is large, and therefore, the correlation can be calculated to predict what chromaticity difference between the pixels, that is, one frame of the image, is large. That is, when the vertical synchronization signal is applied from the synchronization generating unit 2, the correlation of one frame is calculated to bypass the correlation when the correlation is greater than the reference level, and perform the low pass filter processing when the correlation is smaller than the reference level. Here, the low pass filter processing method includes a filter tap weight (for example, 1: 3: 1) for each pixel of the first macroblock and the first pixel of the next macroblock, each of which is one pixel increment closest to the two macroblock boundaries. ) And multiplying each value by three to divide the total weight (e.g., the sum of the five filter tap weights) to compensate for one pixel of the boundary of the first macroblock stage, and then the method as described above. By multiplying the filter tap weights by two pixels of the next block boundary part adjacent to the first block and one pixel of the first block boundary part, each of the three values is added and divided by the total weight. Compensation is made for one pixel of the block. In this way, correlation processing for all blocks in the horizontal direction is continued.

Similarly, the inter-block boundary portion is correlated with respect to the macroblocks in the vertical direction using a low pass filter processing method in the horizontal direction.

Thereafter, the video D / A changing unit 6 converts the digitally compensated output video data R, G, and B calculated by the digital signal processing unit 3 into an analog video signal and outputs the converted analog video signal.

Therefore, according to the present invention operating as described above, by providing a correlation calculation means when the decoded compressed video in block units can significantly reduce the blocking effect between the block and the block caused by some data loss in the compression process It is possible to improve the image quality.

Claims (2)

Video media R, G, and B, each of which is composed of 8 bits by decoding a recording medium 7 such as a video CD in which video data is compressed by MPEG method, and video bit stream data compressed in block units on the recording medium 7. MPEG video decoder 1 for outputting the video signal, a synchronization generator 2 for generating horizontal and vertical synchronization signals and a pixel clock signal, and receiving image data R, G, and B decoded by the MPEG decoder 1 Digital signal processing section 3 that performs high-speed operation in accordance with the synchronization signal generated by the synchronization generating section 2, and image D / A conversion for converting and outputting digital video data output from the digital signal processing section 3 into analog signals. Part 6 In a device having a ROM (4) and a RAM (5) in which a program for performing a predetermined operation is stored, stored in the ROM (4) and bypassed when the correlation of one frame is greater than the reference level, And a low pass filter is processed in the horizontal and vertical macroblock boundaries when the reference level is smaller than the reference level. The method of claim 1, wherein the low-pass filter processing method multiplies two pixels closest to the first block boundary among two macroblocks and a filter tap weight corresponding to the first pixel of the next block, respectively. Compensation for the pixel closest to the first block boundary is performed by adding three values and dividing by the total weight, and then a filter corresponding to each pixel of the next block and one pixel closest to the boundary of the first block. A method of improving image quality in a video apparatus using an MPEG method, comprising: multiplying a tap weight and adding the multiplied three values to divide the total weight to divide the pixel closest to the next block boundary.