JPH10304363A - Image signal processing unit and its method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain transmission and/or recording of image data with less image quality deterioration due to quantization. SOLUTION: This method employs; a division means 103 that divides a received image signal into a plurality of blocks consisting of a prescribed number of pixels; a transformation means 104 that applies orthogonal transformation to image data of each block division by the division means 103; a quantization means 106 that quantizes the image data orthogonally transformed by the transformation means 104; a coding means 107 that applies variable length coding to the quantized image data; and a means that adjusts the code after quantization and variable length coding to have a code in a range consisting of a prescribed number of blocks. Then the method for rounding an arithmetic operation of the quantization means 106 is made unequal within one block consisting of pixels of a prescribed number so as to attain image compression with high image quality and to reduce a maximum value of a quantization error.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image signal processing apparatus and method, and more particularly, to a method for compressing analog-to-digital converted image data, decompressing the same after passing through a transmission medium or a recording medium. The present invention also relates to an image signal processing apparatus and method for outputting an image signal by performing digital-to-analog conversion.

[0002]

2. Description of the Related Art An image signal is divided into blocks each composed of a plurality of pixels, and the data is subjected to an orthogonal transformation. An image signal processing device that performs variable-length encoding and transmits or records the encoded data has been considered. Hereinafter, an example of a conventional image signal processing device will be briefly described.

FIG. 4 shows a configuration diagram of a conventional image signal processing apparatus. In FIG. 4, reference numeral 201 denotes an input signal of an interlaced analog image signal, 202 denotes an analog-to-digital converter for converting an analog image signal into a digital image signal, and 203 denotes a block (for example, a quadrature transform of the digital image signal). , 8 pixels vertical x 8 horizontal
Block dividing unit for dividing the image into pixels.

[0004] Reference numeral 204 denotes a DCT processing unit for orthogonally transforming a block-divided digital image signal. 205
Means that a plurality of blocks (for example, 30
This is a code amount estimating unit for estimating a quantization characteristic so as to obtain a predetermined code amount when collectively quantizing and performing variable length coding.

[0005] Reference numeral 206 denotes a quantization unit which quantizes the data orthogonally transformed by the DCT processing unit 204 by using the quantization characteristic estimated by the code amount estimation unit 205. A variable-length coding unit that performs variable-length coding on the converted data. A data processing unit 208 transmits or records an image signal encoded by the variable length encoding unit, and outputs a signal 209 to be transmitted or recorded.

FIGS. 5A and 5B show the input of a quantization block on the X-axis and the output of the quantization block on the Y-axis, and the quantization characteristic is Y = X ÷ N. FIG. 5 (a)
5 shows a case where truncation is performed by calculation, and FIG. 5B shows a case where rounding is performed by calculation.

In FIG. 5A and FIG. 5B, the broken lines indicate the calculation result of Y = X ÷ N, and the solid line indicates the result of rounding. , Rounding such as truncation and rounding was performed.

[0008]

In the conventional image signal processing apparatus, when the same data is quantized with the same quantization characteristics, the output is (a) smaller when rounding down than when rounding off. More zeros in the data. Also,
(B) Since the data value is reduced, the compression efficiency of the output data of the variable length coding unit 207 may be improved.

Therefore, when the truncation is performed, the N can be made smaller in the quantization characteristic Y = X ÷ N than when the rounding is performed, so that the quantization step becomes smaller. There is an advantage that high-quality image compression can be performed.

[0010] However, there is a demerit that the maximum value of the quantization error increases when truncation is performed. As described above, the conventional rounding method in the quantization operation has advantages and disadvantages, and there is a problem that high-quality image compression cannot be performed by using one kind of rounding method. Was.

SUMMARY OF THE INVENTION In view of the above-mentioned problems, it is an object of the present invention to perform high-quality image compression and to perform quantization with little image quality deterioration.

[0012]

According to the present invention, there is provided an image signal processing apparatus comprising: a dividing means for dividing an input image signal into a plurality of blocks each having a predetermined number of pixels; and an image of each block divided by the dividing means. Transforming means for orthogonally transforming data, quantizing means for quantizing image data orthogonally transformed by the transforming means, encoding means for variable-length encoding the quantized image data, and the quantization and variable length Means for making the encoded code have a code amount within a range of a predetermined number of blocks, wherein the method of rounding the operation performed by the quantization means comprises the predetermined number of pixels. It is characterized in that it differs within one block.

Another feature of the present invention is that
The quantization means, when making the rounding method of the operation different in the one block, the error of the rounding of the operation is smaller in the lower frequency component of the orthogonally transformed image data than in the higher frequency component. It is characterized by doing so.

According to another feature of the present invention, the quantizing means rounds the orthogonally transformed image data to a lower frequency component by a quantization operation, and vice versa. In addition, the method is characterized in that rounding of truncation is performed by quantization operation on the higher frequency component.

According to another feature of the present invention, the encoding means rearranges the orthogonally-transformed data from low frequency components, so that the number of continuous zeros of the data and the non-zero number of zeros are determined. Huffman coding is performed by combining data.

Further, according to the image signal processing method of the present invention, a dividing process of dividing an input image signal into a plurality of blocks each composed of a predetermined number of pixels, and an image data of each block divided by the dividing process are orthogonalized. A conversion process for converting;
A quantization process for quantizing the orthogonally transformed image data, an encoding process for performing variable length encoding on the quantized image data, and a range in which the quantized and variable length encoded code is a predetermined number of blocks. In the image signal processing method for performing the process of setting the code amount of
It is characterized in that it is different in each block.

Another feature of the present invention is that
In the quantization process, when differentiating the rounding method within the one block, the rounding error of the operation is such that the lower one of the frequency components of the orthogonally transformed image data is smaller than the higher one. It is characterized by doing.

According to another feature of the present invention, in the quantization process, the orthogonally transformed image data is rounded to a lower frequency component by a rounding operation using a quantization operation. In addition, the method is characterized in that rounding of truncation is performed by quantization operation on the higher frequency component.

According to another feature of the present invention, in the encoding process, the orthogonally transformed data is rearranged from the one having a low frequency component, and the number of consecutive zeros of the data and the non-zero data are determined. Huffman coding is performed by combining data.

Since the present invention comprises the above technical means, the rounding method in the operation for performing quantization is selected according to the frequency component of the orthogonally transformed data, so that the compression efficiency can be improved and the quantization efficiency can be improved. Image quality deterioration due to image formation can be reduced.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) An embodiment of an image signal processing apparatus and method according to the present invention will be described below with reference to the drawings. FIG.
FIG. 2 is a block diagram illustrating a configuration of an image signal processing device according to the present embodiment. In FIG. 1, reference numeral 101 denotes an input signal of an interlaced analog image signal, and reference numeral 102 denotes an analog-digital converter for converting the input analog image signal into a digital image signal.

Reference numeral 103 denotes a block division processing unit that divides the digital image signal into blocks (for example, 8 pixels vertically × 8 pixels horizontally) that are orthogonally transformed. 104 denotes a DCT processing unit that orthogonally transforms the divided digital image signal. , 10
5 denotes a plurality of orthogonally transformed blocks (for example, 30
This is a code amount estimating unit for estimating a quantization characteristic such that a predetermined code amount is obtained when the data is collected and quantized and subjected to variable-length coding.

Reference numeral 106 denotes a quantization unit that quantizes the image data orthogonally transformed by the DCT processing unit 104 with the quantization characteristics estimated by the code amount estimation unit 105. 110
Is a quantization control unit that controls the quantization characteristics of the quantization unit 106, 107 is a variable length coding unit that performs variable length coding on the quantized image data, and 108 is a unit that transmits the coded image data. Alternatively, it is a data processing unit for recording, and outputs an output signal 109 to be transmitted or recorded.

The variable-length coding unit 107 performs the Huffman coding by rearranging the orthogonally transformed data in ascending order of frequency components and combining the continuous number of zeros of data with non-zero data.

At this time, in the Huffman coding, the data amount tends to be highly compressed as the number of zeros of the data increases or as the value of the data decreases. In the quantization unit 106, the rounding of rounding down and rounding is selected and performed by the quantization control unit 110 in the quantization operation.

For example, as shown in FIG. 2, the orthogonally transformed data has a lower frequency component as an A region and a higher frequency component as a B region.
Area. Then, the operation is performed so that the rounding of rounding is performed in the A region by the quantization operation, and the rounding of the truncation is performed in the B region.

The reason for this is as follows: (a) The frequency of the A region (low band) becomes zero by quantization as compared with the B region (high band). Therefore, even if the data is truncated, it does not become zero, and high compression in Huffman coding cannot be expected. Therefore, rounding with less quantization error is performed. Also, (b) the A region (low band) is more likely to affect the image quality than the B region (high band), so it is more advantageous to perform rounding with less quantization error. .

On the other hand, (c) the frequency of the B area (high frequency) becomes zero by quantizing more frequently than the A area (low frequency). Therefore, by performing the truncation, high compression in the Huffman coding can be expected. Therefore, the truncation is rounded.

(Second Embodiment) In the first embodiment described above, the rounding of the quantization operation in the quantization unit 106 is divided into two regions, an A region and a B region shown in FIG. However, in the present embodiment, as shown in FIG. 3, the frequency components are divided into three regions, a C region, a D region, and an E region.
In the D area, rounding is performed to the nearest whole number. Further, in the E region, the operation is performed so as to perform rounding.

Here, the number of divided areas and the boundaries of the areas are not limited to those described in the above embodiment. Further, the rounding method in the quantization operation is not limited to the truncation or the rounding and the rounding described in the embodiment.

(Other Embodiments of the Present Invention) The present invention is applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), and is applied to one device (for example, a copying machine). , A facsimile machine).

Further, in order to realize the functions of the above-described embodiment, a device connected to the above-described various devices or a computer in a system is operated so as to operate various devices so as to realize the functions of the above-described embodiment. The present invention also includes a program that is implemented by supplying the program code of the software described above and operating the various devices according to a program stored in a computer (CPU or MPU) of the system or apparatus.

In this case, the program code of the software implements the functions of the above-described embodiment, and the program code itself and means for supplying the program code to the computer are provided.
For example, a storage medium storing such a program code constitutes the present invention. As a storage medium for storing such a program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM,
A magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the supplied program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) or other operating system running on the computer. Needless to say, the program code is also included in the embodiment of the present invention when the functions of the above-described embodiment are realized in cooperation with application software or the like.

Further, after the supplied program code is stored in a memory provided in a function expansion board of a computer or a function expansion unit connected to the computer, the function expansion board or the function expansion unit is specified based on the instruction of the program code. It is needless to say that the present invention also includes a case where a CPU or the like provided in the first embodiment performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0036]

As described above, according to the present invention, the rounding method in the operation for performing the quantization is selected according to the frequency component of the orthogonally transformed data. Or image quality degradation due to quantization can be reduced, and high-quality image transmission and / or recording with little image quality degradation can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image signal processing device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating an area of orthogonally transformed data according to the first embodiment.

FIG. 3 is a diagram illustrating an area of orthogonally transformed data in a second embodiment.

FIG. 4 is a block diagram showing an example of a conventional image signal processing device.

FIG. 5 is a diagram showing a state of rounding in an operation for performing quantization.

[Explanation of symbols]

 Reference Signs List 101 input signal of analog image signal 102 analog-digital conversion unit 103 block division processing unit 104 DCT processing unit 105 code amount estimation unit 106 quantization control unit 107 variable length coding unit 108 data processing unit 109 output signal 110 quantization control unit

Claims (8)

[Claims] A dividing unit that divides an input image signal into a plurality of blocks each including a predetermined number of pixels; a transforming unit that orthogonally transforms image data of each block divided by the dividing unit; A quantizing unit for quantizing the orthogonally transformed image data; an encoding unit for performing variable-length encoding on the quantized image data; and a range in which the quantized and variable-length encoded code is a predetermined number of blocks. An image signal processing apparatus having means for obtaining a code amount, wherein a method of rounding an operation performed by the quantization means is made different within one block composed of the predetermined number of pixels. Image signal processing device. 2. The method according to claim 1, wherein the quantizing means is configured such that, when differentiating a rounding method in the one block, a rounding error of the operation is higher when the frequency component of the orthogonally transformed image data is lower. 2. The image signal processing device according to claim 1, wherein the number of the image signal processing devices is smaller than that of the image signal processing device. 3. The quantization means rounds off the orthogonally transformed image data by a quantization operation in a lower frequency component, and conversely, a quantization operation in a higher frequency component. The image signal processing apparatus according to claim 1, wherein rounding of truncation is performed by: 4. The encoding means according to claim 1, wherein said orthogonally transformed data is rearranged in descending order of frequency components, and Huffman encoding is performed by combining a continuous number of zeros of data and non-zero data. Claim 1 to claim
4. The image signal processing device according to claim 3. 5. A division process for dividing an input image signal into a plurality of blocks each including a predetermined number of pixels; a transformation process for orthogonally transforming image data of each block divided by the division process; Quantization processing for quantizing the quantized image data, encoding processing for variable-length encoding the quantized image data, and a code amount in a range where the quantized and variable-length encoded code is a predetermined number of blocks. An image signal processing method for performing the following processing, wherein the method of rounding the operation performed in the quantization processing is made different within one block composed of the predetermined number of pixels. Processing method. 6. In the quantization process, when a rounding method is made different in the one block, a rounding error of an operation is such that a lower frequency component of the orthogonally transformed image data has a higher frequency component than a higher frequency component. 6. The image signal processing method according to claim 5, wherein the image signal is reduced. 7. The quantization processing includes performing the rounding of the orthogonally transformed image data by a quantization operation in a lower frequency component and a quantization operation in a higher frequency component. 7. The image signal processing method according to claim 5, wherein the rounding is performed by truncation. 8. The encoding process according to claim 1, wherein the Huffman encoding is performed by rearranging the orthogonally-transformed data in ascending order of frequency components and combining a continuous number of zeros of data with non-zero data. Claim 5
8. The image signal processing method according to claim 7.