KR0134357B1 - Divided image coding method and apparatus - Google Patents

Abstract

A method of encoding a video signal, the method comprising: inputting a sampled video signal, storing the input video signal, and dividing an effective image of one frame displayed on a screen among the stored video signals into a plurality of The effective image dividing step of dividing into divided images, and the predetermined division at the same time while being present in an image signal of a predetermined divided image among the plurality of divided images and other divided images on the left or right side adjacent to the divided image. A secondary image forming step of forming a secondary image composed of a predetermined amount of image signals adjacent to the image, a volcanic signal output step of outputting image signals of the secondary images formed in the step, and the output secondary A second storage step of switching the image signals of the images to secondary image intervals and storing them in a plurality of storage means; and in the second storage step. According to a segmentation image encoding method comprising encoding a stored data and a third storage step of storing an encoded image signal, a high quality image signal is obtained by using a plurality of prediction encoders which are formed of an existing integrated circuit for signal processing. There is an effect that can be encoded. In addition, there is an effect that the degradation of the image data compression efficiency generated in each predictive encoder can be prevented.

Description

Split Image Coding Method and Apparatus

1 is a block diagram of a split picture coding apparatus according to an embodiment of the present invention.

2 is a conceptual diagram showing an image segmentation method.

3 is a diagram showing a scanning method of an image signal.

4 is a block diagram showing a predictive encoder of the segmental coder according to an embodiment of the present invention.

5 is a conceptual diagram showing a motion vector estimation.

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

10: frame memory 11: demultiplex

12 to 16: 1/5 frame memory 17 to 21: Predictive encoder

22: buffer

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for segmentation encoding, and more particularly, to a segmentation encoding method and apparatus for encoding and transmitting a video signal having a wide transmission bandwidth, such as a video signal for a high quality TV.

In a digital imaging system, each frame of an image is sampled in units of pixels. The sample value representing the brightness of the pixel is quantized to 8 bits per pixel in the black and white image. In a color image, each pixel has three pieces of brightness information representing color information, for example, RGB, and is quantized to 8 bits. Since the image information thus constructed has a very large amount of information, an image compression (or encoding) technique is required to transmit or store the image information, which mainly uses a method of removing redundant information in the space and time domain.

In general, image compression methods include an interframe coding method for removing redundancy in a spatial domain and an interframe coding method for removing redundancy in a time domain.

Predictive encoders used for inter-frame encoding are means for orthogonal transform and quantization of input video signals, means for variable-length encoding quantized data, and inverse orthogonal transform for quantized data. It is provided with a means to. In addition, the predictive encoder includes a frame memory means for storing inverse quantized and inverse orthogonal transformed data in an image frame unit, and a predictive calculator for interframe motion estimation and compensation.

Such a predictive encoding device is widely applied in the field of digital video communication. In the conventional TV broadcasting system (NTSC, PAL, SECAM, etc.), the transmission bandwidth of the signal is less than 6MHz and the sampling frequency of the image does not exceed 15MHz. The predictive encoder is constructed in the above by using a circuit or integrated circuits dedicated to signal processing.

However, for high-definition television (HD-TV), the signal type (for example, 1125 lines / 60 Hz: Hi-vison, 1250 lines / 50 Hz: HD-MAC, 1050 lines / 59.94 Hz: US GI, ATRC) Although there are some differences, the sampling frequency of the pixel is about 50 ~ 75Hz. The high-definition TV video signal having such a sampling frequency has a large amount of data, which causes a problem that an existing signal processing integrated circuit cannot be used in a system for predictive encoding.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to process a high quality video signal having a large amount of data by using a predictive encoder composed of a conventional integrated signal processing circuit by dividing and parallel processing the image signal. The present invention provides a segmentation image coding method that can be used.

Another object of the present invention is to provide a segmentation image encoding apparatus capable of processing a high quality image signal having a large amount of data by an encoder composed of existing signal processing integrated circuits by dividing and parallel processing the image signals.

The object of the present invention is to provide a method of encoding a video signal, the method comprising: inputting a sampled video signal, storing the input video signal, and displaying one frame displayed on the screen among the stored video signals. The effective image separation step of dividing the effective image into multiple divided images, and the image signal present in a predetermined divided image among the divided images and in the other divided images on the left or right side adjacent to the divided image and simultaneously A secondary image forming step of forming a secondary image composed of a predetermined amount of image signals adjacent to the predetermined divided image, an image signal output step of outputting image signals of secondary images formed in the step, and A second storage step of converting the image signals of the output secondary images into secondary image intervals and dividing the image signals into multiple storage means And a third storage step of encoding the data stored in the second storage step and a third storage step of storing the encoded image signal.

Another object of the present invention is to provide a device for encoding a video signal, which receives a sampled video signal and stores it in units of frames. First storage means for matching and outputting, second storage means for storing the output signal of the first storage means in the predetermined divided image unit, means for encoding the image signal output from the second storage means, and This is achieved by a split image coded image device including third storage means for storing the encoded image signal output from the encoding means.

Hereinafter, with reference to the accompanying drawings, a preferred embodiment of a split image coding apparatus according to the present invention will be described in detail.

In the case of high quality TV, since the sampling frequency of the image is about 50-70 MHz, the conventional predictive encoder is used to process the video signal sampled at the sampling frequency not exceeding 15 MHz, and the parallel processing technique is used to process the video signal of the high quality TV. Parallel processing is to process signals simultaneously using several identical devices. The parallel processing is to speed up signal processing by dividing a signal and processing each divided part in several devices.

According to an embodiment of the present invention, an effective image is divided into five vertically for parallel processing. In the case of vertical five, the sampling frequency is f _{S, and} when the time corresponding to the frequency of f _S / 5 is predicted to be encoded. Signal processing is possible. In addition, since a normal video signal including a video signal of a high-definition TV includes a horizontal blanking period and a vertical blanking period, the effective scanning rate for displaying an effective image on the screen is about 75%.

Therefore, by using the remaining 25% of the time, data necessary to increase the accuracy of the motion estimation of each segmented image are additionally allocated to the left and right of each segmented image. The number of horizontal effective pixels of high-definition TV is about 1900. In general, 16 pixels (left and right) are needed for one horizontal scan line to estimate the movement to the left and right. Effective image division of the vertical five-division method that is described below is suitable for image division. The divisional image encoding apparatus uses the above-described parallel processing, and the structure and operation of the divisional image encoding apparatus that divides the effective image into five vertically will be described with reference to FIG.

1 is a block diagram showing an embodiment of a divided image encoding apparatus according to the present invention. The apparatus of FIG. 1 comprises image division storage means for dividing and storing a video signal of one frame according to a predetermined method, predictive encoding means for encoding the divided and stored image signal, and storage means for storing the encoded image signal.

The image division storing means includes a demultiplexer (DEMUX) 11 for switching and outputting the image signal output from the first frame memory 10 and the first frame memory 10 to five terminals, and the image signal for switching output. Five five-fifth frame memories 12 to 16 are provided. In addition, the predictive encoding means comprises five predictive encoders 17 to 21 that are individually connected to each of five 1/5 frame memories, and each of the predictive encoders 17 to 21 output terminals is a buffer 22 that is a storage means. ). The 1/5 frame memories shown in FIG. 1 are defined as the first 1/5 frame memory 12, the second 1/5 frame memory 13, etc. in order from the top, and the predictive encoders are first predicted from the top in order. The encoder 17, the second predictive encoder 18, and the like are determined.

2 is a conceptual diagram showing an image segmentation method. FIG. 2 (a) shows a general video signal composed of a horizontal blanking period (h), a vertical blanking period (v), and an effective image (e) in a block form, and is divided into one to five ruled lines. Indicates a split image. FIG. 2 (b) shows the data stored in the 1/5 frame memories 12 to 16 as secondary divided images A to E. As shown in FIG. Here, α, β, and γ, which are distinguished by a thick solid line and a thin solid line, are the same as the image blocks α, β, and γ, respectively, separated by a dotted line and a dotted line in FIG. 2 (a). It is included in the secondary split image. For example, the data in FIG. 2 (a) of the second divisional image A is composed of the data in the divisional image 1 and the data in the block β separated by a dotted line and a dotted line in the divisional image 2 on the right side thereof. The other secondary divided pictures are constructed in a similar manner to the secondary divided picture A described above.

FIG. 2C is a timing diagram for repeatedly outputting predetermined data from data stored in the first frame memory 10.

When the digital video signal for high-definition TV is input to the input terminal of the apparatus of FIG. 1, the digital video signal composed of data about the horizontal blanking period and the vertical blanking period and data about the effective image displayed on the TV screen is obtained by the first terminal. One frame size is stored in the frame memory 10. Among the data of FIG. 2 (a) stored in the first frame memory 10, data e corresponding to an effective image is output in accordance with the timing pulse shown in FIG.

For example, when the timing pulse value in FIG. 2 (c) is output from a predetermined control unit (not shown) and applied to the first frame memory 10, the data stored in the first frame memory 10 is stored. The data in the divided image 1 and the data in the block? Adjacent to the divided image 1 in the divided image 2 on the right side thereof, that is, the data about 16 pixels per horizontal scan line, are output. When the timing pulse of Fig. 2 (c) is applied to the first frame memory 10, the data in the divided image 2 among the data in the frame memory 10 and the divided image 2 in the divided image 1 on the left side thereof. The data of the adjacent block? And the data of the block? Adjacent to the divided picture 2 are output in the divided picture 3 on the right side thereof. In FIG. 2 (b), block α is the block α in the divided image 1 of FIG. 2 (a) and the block α in the divided image 1 of FIG. 2 (a) is the same as that of FIG. 2 (a) and (b). In the case of the block β in the same picture block. In this manner, the data in the divided images 1 to 5 of FIG. 2 (a) include 16 per horizontal scan line from adjacent blocks at the left and right edges of each divided image in accordance with the timing pulse of FIG. The data of the pixels are added and the data of the secondary divided images as shown in FIG. 2 (b) are output from the first frame memory 10. The data of the secondary divided images output in accordance with the timing pulse in the first frame memory 10 are switched in the demultiplex 11 and input to the 1/5 frame memory connected by switching. In the switching method of the demultiplex 11, the secondary divided image A is stored in the first 1/5 frame memory 12 among the secondary divided images A to E shown in FIG. B is stored in the 2 1/5 frame memory 13 and is sequentially stored in one 1/5 frame memory even in the case of the secondary divided images C to E. FIG. The picture signals of the secondary divided pictures A to E stored in the 1/5 frame memories are input to respective prediction encoders and encoded through a predetermined process.

The predictive encoders 17 to 21 all have the same configuration, and the data belonging to the secondary divided image A of FIG. 2 (b) output from the first 1/5 frame memory 12 are the first predictive encoder 17. Is encoded by The other predictive encoders 18 to 21 also encode the data of the secondary divided pictures B to E of FIG. 2 (b) in the same manner as the above-described predictive encoder 17, so that the predictive encoding in the first predictive encoder 17 is performed. Explain only.

FIG. 3 is a diagram showing a scanning method of an image signal. FIG. 3A shows a raster scan for the whole image. FIG. 3B shows a raster scan in units of N × N blocks. Block 1, 2,... , m, m + 1,... It shows the processing in the order of. A block-based data processing method as shown in FIG. 3 (b) is used in the predictive encoder, and the configuration and operation of the predictive encoder 17 used in the split image encoder according to the present invention will be described.

The apparatus of FIG. 4 is a first predictive encoder 17 shown in FIG. 1, which includes a discrete cosine transform unit 31 for orthogonally transforming a sampled image signal and a quantization unit 32 and a quantization unit for quantizing the output signal. And a variable length encoding unit 37 connected to the output terminal of the unit 32. In addition, the inverse quantization unit 33 connected to the output terminal of the quantization unit 32 is connected to transmit the output signal to the inverse discrete cosine transform unit 34.

The predictive operation unit 36 receiving the sampled image signal described above is connected to receive the signal output from the second frame memory 35, and its output terminal is connected to the first adder A1 and the second adder A2. . The first adder A1 adds the input sampled image signal and the predictive operator output signal and applies them to the discrete cosine transform unit 31 described above. On the other hand, the second adder A2 is placed between the inverse discrete cosine transform unit 34 and the second frame memory 35 to add the predictive operation unit 36 output signal and the inverse discrete cosine transform unit 34 output signal. It is connected to the second frame memory 35 so that it can be applied.

In FIG. 4, when the image signal sampled in the second divided image A in FIG. 2 (b) into a block of size N × N is input to the predictive encoder through the input terminal a, the discrete cosine transform unit 31 And the image signal quantized by the transform coefficient of the N × N block by the quantization unit 32 is inversely transformed and inversely quantized by the inverse quantization unit 33 and the inverse discrete cosine transform unit 34 to obtain the size of the secondary divided image A. The second frame memory 35 is stored in the second frame memory 35. When an image signal previous to one frame stored in the second frame memory 35 is input to the prediction operation unit 36, the prediction operation unit 36 inputs an image signal related to the secondary divided image A of the current frame input to the other one terminal. Motion estimation and motion compensation are performed by using the image signal relating to the second divided image A before the frame.

5 is a conceptual diagram illustrating a motion vector estimation, and shows a motion vector estimation in the prediction operation unit 36 using a block matching algorithm (BMA). The entire block of FIG. 5 is a search area composed of data on the right edge of the secondary divided image A. The left side of the straight line 1 is a 1/5 image when the effective image is divided into five, that is, a part of the divided image 1 The right side of the straight line 1 is a data block relating to a part of the picture block β formed by further reading data blocks adjacent to the right from the first frame memory 10 by 16 pixels horizontally. In the block matching method, assuming that all pixels in an N × N block move in the same direction, the current is within a given search area N + 2L × (N + 2L), where L is the maximum size of the allowed displacement (or vector). The motion vector minimizes the difference between the block in the frame and the block in the previous frame and uses the motion compensation. That is, if block a in the current frame and block b in the previous frame are blocks composed of the same data, the vector d is information used for motion estimation.

In the case where block b, which has the most similar information to block a in FIG. 5, is a data block that spans the boundary between the segmented image 1 and the image block β, an incorrect movement is made when the search area consists only of the right side of the dotted line 1 of FIG. Make an estimate.

On the other hand, if the search area is the entire block of FIG. 5, the data matching block b before one frame can be accurately found. Therefore, the predictive calculation unit 36 makes a motion estimation almost the same as when the screen is not divided, and performs motion compensation on the data of the previous frame by using the motion vector generated by the motion estimation.

When the motion compensated image signal output from the predictive calculation unit 36 is applied to the second adder A2, the second adder A2 adds the inversely quantized and inversely transformed image signal to the motion compensated image signal and obtains the result. The output data is stored in the second frame memory 35. The video signal inputted from the first adder A1 through the input terminal a and the motion compensated image signal output from the predictive calculator 36 are output. When the differential signal is generated and output, the differential signal data in the form of transform coefficients quantized by the discrete cosine transform unit 31 and the quantization unit 32 is converted by the variable length encoder 37 using a Huffman code or the like. Is encoded. The image signal encoded and output by the variable length encoder 37 of the first predictive encoder 17 is stored in the buffer 22. The output signals of the predictive encoders 18 to 21 that are encoded in the same manner, that is, the encoded video signals, are stored in the buffer 22. The output signals of the predictive encoders 17 to 21 are stored in the buffer unit in units of frames, and are output from the buffer 22 and sent to a system such as a decoder. Therefore, by dividing the high-definition TV image signal into five regions and performing predictive encoding on each region, only one fifth of the time required to encode the entire high-definition television image signal with one predictive encoder is obtained. Encode the whole image signal.

In the above-described embodiment, a system in which the effective image is divided into five vertical parts is implemented, but other effective image division methods are possible within the scope of the present invention.

As described above, the present invention has the effect of encoding a high quality image signal by using a plurality of predictive encoders configured as a conventional integrated signal processing circuit. In addition, there is an effect that can prevent the degradation of the image data compression effect rate generated in each predictive encoder.

Claims (15)

A method of encoding a video signal, comprising the steps of: inputting a sampled video signal; Storing the input video signal; An effective image division step of dividing an effective image of one frame displayed on the screen among the stored image signals into a plurality of divided images; 2 consisting of an image signal of a predetermined divided image among the plurality of divided images and a predetermined amount of image signal existing in other divided images on the left or right side adjacent to the divided image and adjacent to the predetermined divided image A secondary image forming step of forming a secondary image; An image signal output step of outputting image signals of secondary images formed in the step; A second storage step of converting the image signals of the output secondary images into secondary image intervals and storing the divided image signals in a plurality of storage means; Encoding the data stored in the second storage step; And a third storage step of storing the encoded image signal. The method of claim 1, wherein the effective image division step divides the effective image vertically. The method according to claim 1 or 2, wherein the effective image division step divides the effective image into five divided images. 2. The secondary image forming step according to claim 1, wherein the secondary image forming step is carried out by the 16 pixels per horizontal scan line while the amount of the image signal existing in the left or right other divided image adjacent to the predetermined divided image and adjacent to the predetermined divided image. Segmented image coding method, characterized in that. 2. The method of claim 1, wherein the second storing step divides and stores data for a plurality of secondary divided images into a plurality of storage means. The method of claim 1, wherein the encoding step is a motion compensation predictive encoding step. 1. An apparatus for encoding a video signal, comprising: first storage means for receiving a sampled video signal and storing it in units of frames and outputting an image signal of an effective image displayed on a screen from among the video signals in each frame according to a predetermined divided image; and; Second storage means for storing the output signal of the first storage means in the predetermined divided image unit; Means for encoding an image signal output from said second storage means; And a third storage means for storing the encoded image signal output from the encoding means. 8. The method of claim 7, wherein the first storage means divides the effective image displayed on the screen into multiple divided images, and constructs new secondary divided images in such a manner that predetermined regions adjacent to the divided images add to each other. A split image coding apparatus characterized by outputting an image signal of a secondary divided image. The divided image encoding apparatus according to claim 7 or 8, wherein divided images are generated by vertically dividing an effective image displayed on a screen. 9. The divided image encoding apparatus according to claim 7 or 8, wherein the first storing means generates divided images by dividing the effective image displayed on the screen vertically into five equal parts. 12. The storage device according to claim 10, wherein the first storage means is present in the data of the predetermined divided image and the left or right divided image adjacent to the predetermined divided image and corresponds to 16 pixels per horizontal scan line adjacent to the predetermined divided image. A divided image encoding apparatus, characterized in that a predetermined secondary divided image is generated from data. 8. The apparatus of claim 7, wherein the second storage means comprises: a switching unit for switching the image signal output from the first storage means; And a storage unit which receives the image signal output from the switching unit and stores the image signal in the secondary divided image unit. 13. The divided image encoding apparatus according to claim 12, wherein the second storage means includes five storage units. 8. The apparatus as claimed in claim 7, wherein the encoding means is prediction encoding means including a motion compensator and a differential pulse modulator. 8. The divided image encoding apparatus according to claim 7, wherein the third storage means stores the encoded image signal in units of frames.