KR0178197B1 - Apparatus for encoding an image signal using region segmentation - Google Patents

Abstract

The present invention relates to a hybrid image compression device which compresses and transmits an image signal of a predetermined block unit using motion compensation DPCM (differential pulse code modulation) and two-dimensional discrete cosine transform (DCT). A quantization unit for quantizing the DCT coefficients; A coefficient unit for sequentially scanning the quantization values in the block to count non-zero quantization values; Dividing the block into a plurality of diagonal areas, counting the number of non-zero quantization values in diagonal order from a diagonal area having a DC value to detect a diagonal area whose coefficient value is less than a predetermined ratio with respect to the count value of the counting part. A detection unit for transmitting a divided signal; The quantization unit receives the quantization value and sequentially outputs the quantization value from the DC value of the block to the diagonal region corresponding to the divided signal through the first terminal, and outputs the quantization value of the quantization unit after the diagonal region corresponding to the divided signal. A selective output unit for outputting to two terminals; A first encoder for encoding and transmitting the quantized value applied from the first terminal of the selective output unit; And a second encoder for zigzag-scanning the quantization value of the quantization unit applied from the second terminal of the selective output unit and performing line length encoding, and variable-length encoding the transmitted line length encoded value.

Description

Image signal encoding apparatus using region segmentation

1 is a block diagram of a conventional video signal encoding apparatus.

2 is a block diagram of an image signal encoding apparatus using region division according to the present invention.

3 is a diagram illustrating a region segmentation method of an image signal encoding apparatus using region segmentation according to the present invention.

4 is a diagram illustrating an example of intra-block quantization values in an image signal encoding apparatus using region division according to the present invention.

* Explanation of symbols for main parts of the drawings

10: subtractor 20: DCT unit

30: quantization unit 40: inverse quantization unit

50: IDCT unit 60: addition unit

70: frame memory 80: motion estimation unit

90: motion compensation unit 100: encoder

110: counter 120: detector

130: selective output unit 140,150: encoder

The present invention relates to an apparatus and method for encoding an image signal. More particularly, the present invention relates to an apparatus and method for encoding an image signal. More particularly, the present invention relates to an apparatus and method for encoding an image signal. A video signal encoding apparatus.

In general, a video signal can maintain a much better picture quality in digital than an analog picture, but the amount of data transmitted increases in response to the improved picture quality.

However, since a usable frequency range is limited in a conventional transmission channel, in order to transmit a large amount of digital data, it is necessary to compress the transmission data and reduce the amount thereof. In addition, hybrid coding schemes combining spatial compression techniques are known to be the most efficient.

Most hybrid coding techniques use motion compensated DPCM (Differential Pulse Code Modulation), 2-D Discrete Cosine Transform (DCT), quantization of DCT coefficients, run length coding, variable length coding, etc. The motion compensation DPCM determines a motion of an object between a current frame and a previous frame, and predicts the current frame according to the motion of the object to generate a differential signal representing a difference between the current frame and the prediction value. Staffan Ericsson's Fixed and Adaptive Predictors for Hybrid Predictive / Transform Coding, IEEE Transmissions on Communication, COM-33, NO.12 (1985, December), or A motion Compensated Interframe Coding Scheme for Television Pictures, by Ninomiy and Ohtsuka, It is well described in IEEE Transactions on Communication, COM-30, NO.1 (1982, January).

In addition, the two-dimensional DCT reduces or eliminates spatial redundancy between image data by converting one block of digital image data, for example, a block of 8x8 pixels into a group of transform coefficient data.

This two-dimensional DCT technique is described in Chen and Pratt's Scene Adaptive Coder, IEEE Transaction on Communications, COM-32, No. 3 (March 1984), and quantizer and zigzag scanning of the transform coefficient data described above. By processing by scanning, VLC, etc., data to be transmitted can be efficiently compressed.

In the motion compensation DPCM, the motion between the current frame and the previous frame is estimated to predict the current frame from the previous frame. The motion estimation is represented as a two-dimensional vector representing the displacement between the previous frame and the current frame.

1 is a block diagram of a conventional hybrid video encoding apparatus using the above-described method.

As shown, a conventional hybrid image encoding apparatus includes a subtractor 10, a DCT unit 20, a quantizer 30, an inverse quantizer 40, an IDCT unit 50, an adder 60, and a frame memory ( 70), a motion estimator 80, a motion compensator 90, and an encoder 100.

First, the subtractor 10 subtracts the current frame signal predicted by the motion compensator 90 to the current frame signal input from the input side. As a result, the difference signal representing the difference pixel value is converted into the DCT unit 20. And the quantization unit 30 is encoded into a series of quantized transform coefficients. Then, the quantized transform coefficients are input to the encoder 100 and the inverse quantizer 40.

More specifically, the DCT unit 20 converts an image signal in a time domain with respect to a differential signal based on input motion estimation and prediction into a transform coefficient in a frequency domain using a discrete cosine function. In addition, the quantization unit 30 quantizes the transform coefficient of the DCT unit 20 to a finite number of values through nonlinear operations, and quantizes the difference signal between the frame to be encoded and the predicted frame.

The encoder 100 performs data compression on the quantized transform coefficients through entropy encoding using line length coding and variable length coding. That is, the encoder 100 performs variable length encoding for encoding the quantized video signal by the run and the coefficient through zigzag scanning or the like, and then uses one codebook according to the generation frequency of each code. Variable length, that is, a code having a large frequency of occurrence of a code, is a short length code, and a variable length coding is performed for a code having a low frequency of code generation, a code of a long length. The reason for allocating different lengths to all codes through the encoder 100 is to increase the coding efficiency by substantially reducing the average value of the code lengths. The present invention substantially improves the efficiency of such variable length coding. It relates to the improvement of variable length coding techniques that can be improved.

On the other hand, the inverse quantization unit 40 and the IDCT unit 50 employed in a typical encoder as a means for performing the motion prediction differential coding as described above are compressed by the DCT unit 10 and the quantization unit 30 described above. The encoded video signal is restored to the original signal before encoding for motion estimation and compensation, and provided to the adder 60, after which the adder 60 is provided from the IDCT unit 50. By adding the difference signal and the predicted current frame signal provided from the motion compensator 90 to the frame memory 70, the frame memory 70 has a restored current frame signal, that is, the current encoding The image signal of the frame immediately preceding the frame signal is stored.

Accordingly, the motion estimator 80 and the motion compensator 90 estimate the motion vector using the current frame from the input side and the previous frame from the frame memory 70 to move the signal of the previous frame by the motion vector. By doing so, the motion compensation is performed and provided to the subtractor 10.

On the other hand, according to the conventional method of the typical encoder for performing the compression encoding of the video signal through the above-described process, only one variable length codebook is used in one frame or one system when variable length coding the quantized signal By using zigzag scanning, short codes are assigned when there are few runs, and long codes are assigned when the runs are long. This is because statistically, many short runs usually occur in the image and long runs are less frequent. It also depends only on the statistics. Where run is the number of zeros before the nonzero coefficient.

In practice, however, the length of these runs varies substantially depending on a number of factors, for example, the type of the picture, the type of encoding of the image, or the coefficient at any position within one DCT block. In the related art, since this point is not considered at all, there is a problem in that the amount of bit generation is unnecessarily large and the coding efficiency is lowered.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and an object of the present invention is to divide a block of quantization values diagonally and scan the quantization values in the divided diagonals so that the number of quantization values in the zero state is less than or equal to a predetermined threshold. The present invention provides an image signal encoding apparatus using region division for transmitting a quantized value of a scanned diagonal column and variable length encoding the remaining quantized values.

In order to achieve the above object, the present invention provides a two-dimensional hybrid image compression apparatus that compresses and transmits an image signal of a predetermined block unit by using motion compensation DPCM (differential pulse code modulation) and two-dimensional discrete cosine transform (DCT). A quantizer for quantizing the DCT coefficients converted by the DCT scheme;

A coefficient unit for sequentially scanning the quantization values in the block to count non-zero quantization values; The block is divided into a plurality of diagonal areas, and the number of non-zero quantization values is counted in the order of the diagonal from the diagonal area having the DC value so that the count value is less than or equal to a predetermined ratio with respect to the count value of the counting part. A quantization value is detected from a detection unit and a quantization unit which detects a diagonal region to be transmitted as a divided signal, and sequentially outputs a quantization value from the DC value of the block to the diagonal region corresponding to the divided signal through the first terminal, and then divides the divided signal. A selection output unit configured to output a quantization value of the quantization unit after the diagonal region corresponding to to the second terminal; A first encoder for encoding and transmitting the quantized value applied from the first terminal of the selective output unit; And a second coder configured to perform zigzag scanning of the quantization value of the quantization unit applied from the second terminal of the selective output unit by line length encoding, and to variable length code the transmitted line length encoded value.

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

2 is a block diagram of an image signal encoding apparatus using region division according to the present invention, in which the encoding unit 100 of the conventional encoding apparatus shown in FIG. 1 is omitted, and the coefficient unit 110 and the detection unit 120 are omitted. The select output unit 130 and the first and second encoding units 140 and 150 are further configured.

Specifically, the coefficient unit 110 connected to the quantization unit 30 sequentially scans the quantization value in the block output from the quantization unit 30 starting with the DC value, thereby quantizing the non-zero quantization value. It is configured to count and output.

In addition, the detector 120 divides the block B output from the quantization unit 30 into diagonal areas R1-R15 (hereinafter, referred to as a diagonal area) as shown in FIG. The number of non-zero quantization values is counted by sequentially scanning the quantization values in the diagonal regions R1-R15. Here, the block is an 8 * 8 block as an example, and the quantization value is located in the small block (b) shown by a dotted line. In addition, the detection unit 120 includes a diagonal region R1-1 that includes a number of quantization values whose counted values are equal to or less than a predetermined ratio, that is, equal to or less than a predetermined threshold ratio with respect to the value counted by the counter 110. One of R15) is detected and transmitted as a divided signal.

For example, if the quantization values of the quantization unit 30 are the same as those of FIG. 4, and the threshold ratio is 60%, the detection unit 120 scans the quantization values in the regions R1, R2, and R3 and then divides the space R3. Will output a signal. This is because, as can be easily seen, the total number of non-zero quantization values in the block is five, whereas the third non-zero quantization value, which is 60%, exists in the region R1-R3. .

The selection output unit 130 receiving the divided signal sequentially applies the quantization value from the DC value of the block to the diagonal region corresponding to the divided signal to the encoder 140, and the diagonal region corresponding to the divided signal. The quantized value is then applied to the encoder 150. That is, in the case of FIG. 4, the selection output unit 130 sequentially transmits the quantization values 6, 0, 2, 3,-2, 0 constituting the regions R1, R2, and R3 to the encoder 140. The quantization values constituting the regions R4-R15 are sequentially output to the encoder 150.

The incubator 140 receiving the quantization values from the selection output unit 130 encodes the input quantization values ((6,0,2,3, -2,0 in the example of FIG. 4) with a predetermined code and transmits them. Meanwhile, the encoder 150 receiving the quantization values from the selection output unit 130 performs zigzag scanning of the input quantization values (which form the regions R4-R15 in the example of FIG. 4) and performs line length encoding. In addition, the variable length coded values of the line length coded data are transmitted again.

That is, the present invention uses non-zero quantization values in a block of quantization values to be coded and transmitted, centered on a DC value, and more than a predetermined ratio is concentrated with respect to the total number of non-zero protons and values. By encoding and transmitting the information on the diagonal region and the quantization values constituting the diagonal region, and transmitting the variable length encoding on the remaining regions, the amount of data having variable length coding can be relatively reduced, thereby greatly reducing the amount of data to be transmitted. There is an effect that can be reduced.

Claims (1)

A hybrid image compression apparatus for compressing and transmitting a video signal of a predetermined block unit by using motion compensation DPCM (Differential Pulse Code Modulation) and 2D Discrete Cosine Transform (DCT) scheme, the DCT coefficient converted by the 2D DCT scheme A quantization unit for quantizing them; A counting unit for sequentially scanning the quantization values in the block to count non-zero quantization values; Dividing the block into a plurality of diagonal regions, counting the number of non-zero quantization values from the diagonal region having a DC value in the order of the diagonal so that the coefficient value is less than a predetermined ratio with respect to the coefficient value of the counter portion. A detection unit which detects an area and transmits the divided signal; The quantization unit receives the quantization value from the block and sequentially outputs a quantization value from the DC value of the block to the diagonal region corresponding to the divided signal through a first terminal, and the quantization after the diagonal region corresponding to the divided signal. A selection output unit configured to output a negative quantization value to a second terminal; A first encoder for encoding and transmitting a quantized value applied from the first terminal of the selective output unit; Image signal encoding using a segmentation unit having a second encoding unit configured to perform zigzag scanning of the quantization unit of the quantization unit applied from the second terminal of the selective output unit by line length encoding, and variable length encoding of the line length encoded value. Device.