JP2907715B2 - Video signal processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal processing apparatus for encoding and decoding a sampled video signal.

[0002]

2. Description of the Related Art In recent years, D1 and D2 video tape recorders (hereinafter referred to as VT) for digitizing and recording / reproducing video signals.
Described as R. ) Has been developed. Also, video floppy capable of recording and reproducing digital still images and VTR capable of recording digital moving images have been developed as consumer equipment. Examples of development of consumer digital VTRs include the Journal of the Institute of Television Engineers of Japan (vol. 45, No. 7, pp. 813-8).
19, 1991). In this consumer digital VTR, the redundancy of the video signal is recorded by compressing the information amount to about 1/5 using various methods.

[0003] A conventional video signal processing apparatus will be described below. FIG. 4 is a block diagram of a conventional video signal processing device. FIG. 3 is an explanatory diagram of the video signal subjected to the orthogonal transformation in FIG. 4, (Table 1) is a table for explaining class discrimination, (Table 2) is a table for explaining quantization, and (Table 3) is a quantization number and a quantum. (Table 4) is a table for explaining the relationship with the classifying step, and (Table 4) is a table for explaining the offset to the class value.

[0004]

[Table 1]

[0005]

[Table 2]

[0006]

[Table 3]

[0007]

[Table 4]

In FIG. 4, reference numeral 41 denotes an adjacent 8 after sampling.
An input terminal for a video signal divided into × 8 pixels, 42
Is an orthogonal transformation circuit that orthogonally transforms the signal of each block input to the input terminal 41 and outputs orthogonal coefficients (DC coefficient and AC coefficient). 43 denotes each block from the 63 AC coefficients output from the orthogonal transformation circuit 42. A class discriminating circuit for deciding the means of quantization; 44, a quantizing circuit for quantizing the output signal of the orthogonal transform circuit 42 based on the class discriminating information output from the class discriminating circuit 43; 45, a quantized signal Output terminal for output.

An input terminal 41 receives a video signal obtained by dividing the screen into blocks of 8 × 8 pixels. The blocked video signal is linearly converted by the orthogonal conversion circuit 42 using mutually independent conversion axes. As a result of the conversion, 64 coefficients as shown in FIG. 3 are obtained. In FIG. 3, the DC coefficient represents an average value (DC component) of 8 × 8 pixels. Remaining 63 coefficients 1
63 are AC coefficients, which include higher frequency components in the horizontal and vertical directions. Here, AC1 to AC5 are Area 0, AC6 to AC20 are Area 1, AC21 to AC
42 is defined as area 2, and AC43 to AC63 are defined as area 3.

As described above, since the power is concentrated on a specific component due to the statistical properties of the video signal by orthogonally transforming the video signal, it is possible to reduce the number of bits per block by devising the bit allocation. it can.

The result of the orthogonal transformation of the input signal is supplied to a class discrimination circuit 43 by a DC converter having low horizontal and vertical frequency components.
AC value of higher frequency from the value (AC1 → AC2 → AC3 →
・ ・) The class discriminating circuit 43 classifies the signals in block units output from the orthogonal transform circuit 42 based on the rules described in (Table 1). For example, when the maximum value of the AC coefficient in the block obtained by orthogonally transforming the input luminance signal is 30, the class is determined to be area 2.

When the class is the area 3, the quantization circuit 44 shifts down the AC coefficient in the block by 1 bit (1 bit).
/ 2), and after estimating the amount of data after quantization using 30 blocks as one unit, the AC coefficient of each area is converted into bits by nine types of coarse to fine quantization methods shown in (Table 2). The quantization is performed by shifting. An estimate of the data amount after quantization is obtained for each of the 16 quantization steps (see (Table 3)). When the quantization step is 0, each block is quantized with the quantization number indicated by the new quantization step obtained by adding the offset value (see (Table 4)) based on the class value of each block to the quantization step 0. Further, for the signal after quantization, the cumulative number of codewords at the time of variable length encoding is compared with a predetermined target value,
The quantization is performed based on the maximum quantization step that does not exceed the target. That is, the offset value of each block is added to the quantization step value, and quantization is performed using the quantization number corresponding to the new quantization step. The quantized signal is output to a signal output terminal 45.

[0013]

However, the above-mentioned conventional configuration has the following problems. When performing quantization, the class is determined based on the maximum value of the AC coefficient. Therefore, when a high-frequency AC coefficient is included in a block having a high proportion of a low-frequency AC coefficient, for example, a character appearing in an image In a visually conspicuous portion such as an edge portion, a large class value is selected, and the image quality is degraded.

An object of the present invention is to solve the above-mentioned conventional problems, and an object of the present invention is to provide a video signal processing device capable of reducing distortion generated in a visually conspicuous portion of an image.

[0015]

In order to solve the above-mentioned problems, a video signal processing apparatus according to the present invention receives a plurality of adjacent sample values and inputs a block-shaped video signal as an input.
Level determining means for determining a level value of the blocked video signal; orthogonal transform means for orthogonally transforming the blocked video signal to output a DC coefficient and an AC coefficient; DC detection means for detecting the obtained DC coefficient, AC detection means for detecting the AC coefficient obtained by the orthogonal transformation means, quantization means for quantizing the coefficient output from the orthogonal transformation means, Quantization control means for controlling the quantization means, wherein the quantization control means includes a level value output from the level determination means, a DC coefficient output from the DC detection means, and an output from the AC detection means. The quantization means is configured to be controlled based on the obtained AC coefficient.

[0016]

According to the present invention, when quantizing a signal subjected to orthogonal transformation after blocking, the present invention provides a level value output from a level discriminating unit, a DC coefficient output from a DC detecting unit, and an AC detection signal. Classification is performed based on the AC coefficient output from the means, for example, the sum of absolute values of the AC coefficient, and quantization is performed to reduce distortion generated in a visually conspicuous portion of the image.

[0017]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows an embodiment of the present invention.
It is a block diagram of a video signal processing device. FIG. 2 is an explanatory diagram of the level determination, and (Table 5) is a table for explaining the classification based on the level determination.

[0018]

[Table 5]

In FIG. 1, reference numeral 11 denotes an adjacent 8 after sampling.
Input terminal of video signal blocked into × 8 pixels, 12
Is a circuit for judging the level value of the input block unit signal. In this embodiment, the level judging circuit outputs a rank value by ranking the amplitude level. 13 orthogonally transforms the input signal for each block and performs an orthogonal coefficient (DC
This is an orthogonal transformation circuit that outputs coefficients and AC coefficients. 14
Is the orthogonal coefficient output from the orthogonal transformation circuit 13,
An AC detection circuit that outputs a sum of 63 AC coefficient values,
5 is a DC from the orthogonal coefficient output from the orthogonal transformation circuit 13
DC detection circuit for detecting coefficients, 16 is a level discrimination circuit 1
2, the sum of the absolute values of the AC coefficients output from the AC detection circuit 14, and the DC detection circuit 1
5 is a class discrimination circuit as quantization control means for deciding quantization means based on the DC coefficient output from 5. Reference numeral 17 denotes a quantization circuit for quantizing an output signal of the orthogonal transform circuit 13 based on the class discrimination information output from the class discrimination circuit 16, and reference numeral 18 denotes an output terminal for outputting a quantized signal.

A video signal obtained by dividing the screen into blocks of 8 × 8 pixels is input to the input terminal 11. The video signal divided into blocks is ranked by the level discrimination circuit 12 according to the amplitude level for each block. For example, as shown in FIG. 2, when the amplitude range of the image signal is 0 to 255, when the minimum level in the block is 200 or more, or when the maximum level is 50 or less, the rank is 1 and all the amplitude levels in the block are rank 1. Rank 0 when 51 or more and 199 or less
Is determined. That is, a block in which the amplitude level of the video signal before the orthogonal transformation is very small or very large is determined as rank 1, and a block having an intermediate amplitude level is determined as rank 0.

The block-formed video signal is linearly converted by the orthogonal conversion circuit 13 using mutually independent conversion axes. AC
The detection circuit 14 detects the 64
From the orthogonal coefficients, the sum of the absolute values of the 63 AC coefficients excluding the DC coefficient is calculated and output. DC detection circuit 15
Detects a DC coefficient from the 64 orthogonal coefficients output from the orthogonal transformation circuit 13 and outputs the DC coefficient.

The class discriminating circuit 16 includes a rank value for each block output from the level discriminating circuit 12, a sum of absolute values of AC coefficients output from the AC detecting circuit 14, and D
The DC coefficient output from the C detection circuit 15 is input.
The class discrimination circuit 16 classifies the signals for each block based on the rules described in (Table 5). For example, if the input block signal is composed of a luminance signal (Y signal), the DC coefficient is 128 or more, the sum of the absolute values of the AC coefficients is 128 or more, and the rank is 0, the class is determined to be area 2; If the sum of the absolute values of the AC coefficients is less than 128 and the rank is 0, the class is determined to be area 3.

The quantizing circuit 17 receives the orthogonal coefficients of the block output from the orthogonal transform circuit 13 and the class value output from the class discriminating circuit 16 and performs quantization. The quantized signal is output to a signal output terminal 18.

As described above, according to the present embodiment, blocks in which the amplitude level of the video signal before the orthogonal transform is very small or very large are coarsely quantized, and blocks in the middle are quantized. By performing fine quantization on a block having a small sum of the absolute values of the AC coefficients and a block having a large sum of the absolute values of the AC coefficients and having a large DC coefficient value, the quantization in a portion where distortion is visually conspicuous is performed. It is possible to reduce the image formation error and prevent image quality deterioration.

The above embodiment shows an example of the ranking of signals on a block-by-block basis.
The region may be divided into smaller regions and ranked.

[0026]

As described above, according to the present invention,
For example, each block is ranked according to the amplitude level of the video signal, and the DC coefficient after the orthogonal transformation,
By controlling the quantization based on the AC coefficient, a quantization error generated in a block including a visually conspicuous portion of the image can be reduced, and the image quality can be prevented from deteriorating.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a video signal processing device according to an embodiment of the present invention.

FIG. 2 is a view for explaining ranking of video signals in the video signal processing device according to one embodiment of the present invention.

FIG. 3 is a diagram illustrating an orthogonally transformed video signal.

FIG. 4 is a block diagram showing a configuration of a conventional video signal processing device.

[Explanation of symbols]

 Reference Signs List 11 signal input terminal 12 level discrimination circuit 13 orthogonal transformation circuit 14 AC detection circuit 15 DC detection circuit 16 class discrimination circuit 17 quantization circuit 18 signal output terminal

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) H04N 7/24-7/68 H04N 1/41-1/419

Claims (4)

(57) [Claims] 1. A level discriminating means for collecting a plurality of adjacent sample values and inputting a blocked video signal as input and discriminating a level value of the blocked video signal; Are orthogonally transformed, and a DC coefficient and an AC coefficient are output. An orthogonal transformation means, a DC detection means for detecting a DC coefficient obtained by the orthogonal transformation means, and an AC coefficient obtained by the orthogonal transformation means are detected. AC
Detecting means, quantizing means for quantizing the coefficient output from the orthogonal transformation means, and quantization control means for controlling the quantization means, wherein the quantization control means outputs from the level discriminating means. A video signal processing device for controlling the quantization means based on the obtained level value, the DC coefficient output from the DC detection means, and the AC coefficient output from the AC detection means. 2. The quantization control unit according to claim 1, wherein the level determination unit determines that the level of the blocked video signal is higher than a predetermined first threshold value, 2. The video signal processing apparatus according to claim 1, wherein when it is determined that the threshold value is smaller than the threshold value, the quantization means is controlled to perform coarse quantization. 3. The video signal processing apparatus according to claim 1, wherein the AC detection means outputs a sum of absolute values of AC coefficients from the coefficients orthogonally transformed by the orthogonal transformation means. 4. The quantization control means includes: a block having a small sum of absolute values of AC coefficients output from the AC detection means;
Controlling the quantization means so as to perform fine quantization on a block having a large sum of absolute values of the AC coefficients output from the AC detection means and having a large DC coefficient output from the DC detection means. 4. The video signal processing device according to claim 3, wherein: