JPH10145790A - Method for forming picture and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a method for forming picture and a device therefor improved in compression ratio without making picture quality largely deteriorate even in the case of image picture data. SOLUTION: The picture data of an original picture is inputted (S1), and replaced by each coefficient by wavelength transformation (S2). Then, quantization (S3) processing is performed thereon and it is data-compressed by entropy coding (S4) like a QM-coder. The quantization processing averages and replaces for every eight coefficients having a high frequency component in each of the horizontal direction and the vertical direction, for example. The compressed data is temporarily held (S5), and then it is data-expanded and to be print data (S6). According to this processing, since plural adjacent coefficients mutually having a strong correlation are replaced by one coefficient, compression ratio is improved without making deterioration of picture be noticeable.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image forming apparatus and an image forming method provided with a high-efficiency image data compression processing system.

[0002]

2. Description of the Related Art Conventionally, in an image forming apparatus or an image forming method, generally, the cost of a memory portion for holding print data is very large. Therefore, in order to reduce the cost, a technique of data compression is used. As one of the methods, there is a compression method using a wavelet transform. The wavelet transform can be said to be a transform for dividing the original image into frequency components in the horizontal, vertical, and oblique directions.

FIG. 4 is a block diagram of a conversion processing unit to which conventional four-layer wavelet conversion processing is applied.
FIG. 5 shows coefficients after wavelet transform of the image data of the original image shown in FIG. The coefficients shown in FIG. 4 are assigned to the coefficients of the respective enclosed areas in FIG.

FIG. 6 shows a specific configuration example of an original image to be converted, which represents data for a conversion unit block (16 × 16 pixels) of the original image (8 bits / pixel). FIG. 7 shows an example of the configuration of coefficients generated from the original image in FIG. 6 by the wavelet transform processing unit in FIG.

HL1 is horizontal with respect to the original image, LH1
Is a coefficient representing a high frequency component in the vertical direction, HH1 is a coefficient representing a high frequency component in an oblique direction, and LL1 is a coefficient representing a low frequency component. The obtained LL1 (8 × 8) is considered as an original image, and the same conversion is performed to perform the second hierarchical coefficient (HL2, LH2, HH2, LL).
Obtain 2). Thereafter, the same processing is repeated to obtain the coefficients of the third and fourth layers.

Several proposals have been made to improve the compression ratio by using the wavelet transform method in order to increase the efficiency. For example, Japanese Patent Application Laid-Open No.
No. 8, “Image data compression processing method” is to apply a wavelet transform to image data at a plurality of hierarchies, and obtain a coefficient of “0” for a coefficient at a corresponding position of a lower hierarchy (high resolution side). Are all changed to "0" to improve the compression ratio.

[0007] Further, in the "Image data compression processing method" of Japanese Patent Application Laid-Open No. Hei 7-23229 of the prior art 2, all of the coefficients spatially adjacent to each other in the same layer are obtained for each coefficient obtained by the wavelet transform. If "0", the attention coefficient is also "0"
By doing so, the noise is removed and the compression ratio is improved.

[0008]

However, in each of the above-mentioned conventional examples, image images such as photographs are generally hard to compress as compared with characters, line images, computer graphic images and the like. Even in the data compression method using the wavelet transform, it is not practically easy to increase the compression ratio of an image image. This is because the image image is formed of a plurality of frequency components having an irregular distribution. Therefore, the coefficient representing the frequency component obtained by performing the wavelet transform on such an image usually includes many values other than “0”, and this is rarely “0”.

In the above-mentioned conventional example 1, since the effect of improving the compression ratio appears only when the coefficient contains "0", there is almost no effect on an image image which originally has a small number of "0". Includes problems.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an image forming apparatus and an image forming method capable of increasing the compression ratio of image data without deteriorating the image quality.

[0011]

In order to achieve the above object, an image forming apparatus according to the present invention is an image forming apparatus which obtains an image output by quantizing coefficients obtained by dividing an input signal into components. The coefficients are averaged and replaced for every N high-frequency components in the direction (N is a natural number of 2 or more), and the coefficients are averaged for every N high-frequency components (N is a natural number of 2 or more) in the vertical direction. An image data compression processing system for performing quantization and coefficient conversion by transforming the image data.

It is preferable that the number N is eight.

An image forming method according to another aspect of the present invention is an image forming method for obtaining an image output by quantizing coefficients obtained by performing component division coding of an input signal. The image forming method includes N high-frequency components in the horizontal direction (where N is 2). A replacement step of averaging and replacing coefficients for each of the above natural numbers, and a replacement step of averaging and replacing coefficients for every N (N is a natural number of 2 or more) high-frequency components in the vertical direction. Then, the quantization of the coefficient is performed.

The number N is preferably eight.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment of an image forming apparatus and an image forming method according to the present invention will be described in detail with reference to the accompanying drawings. 1 to 3 show an embodiment of an image forming apparatus and an image forming method according to the present invention.

FIG. 1 is a flowchart showing a processing procedure to which the image forming apparatus and the image forming method of the present invention are applied. 4 to 7 used in the description of the conventional example are applied to the four-layer wavelet transform process. FIG. 2 is a specific example when FIG. 7 is processed in the present embodiment. FIG. 3 shows a state in which the quantized coefficients of FIG. 2 are restored.

In general, there is a human visual characteristic that, even if the image quality deterioration is almost the same, the steep change in density is more conspicuous in the portion where the change in density is more noticeable, and conversely it is inconspicuous in the portion where the density changes gradually. Similarly, from the horizontal and vertical edges,
It is known that a visual characteristic that an edge in an oblique (45 degrees) direction is hard to perceive. The present invention is intended to use such characteristics to concentrate the deterioration of the image quality due to the improvement of the compression ratio in a portion inconspicuous to human eyes.

FIG. 1 is a flowchart showing an embodiment of the present invention. In FIG. 1, image data of an original image is input in step S1. The data of the original image input in step S2 is replaced with coefficients by wavelet transform. The magnitude of the absolute value of this coefficient indicates the magnitude of the steepness of the density change of the original image. afterwards,
The quantization (S3) processing based on the present embodiment is performed, and Q
Data is compressed by entropy coding (S4) such as M-coder. The compressed data is temporarily stored (S5) and then expanded to become print data (S6).

In the wavelet transform, the original image is divided into blocks of 16 pixels vertically and 16 pixels horizontally, and a transform (TS transform) is first performed in the horizontal direction based on the following equation.
6 horizontal 8 LPF outputs s (n), HPF output d
(N) is obtained. Subsequently, the same TS conversion is performed on each of the output values in the vertical direction to obtain a conversion coefficient of the first layer.

LPF: s (n) = | [x (2n) + x (2n + 1)] / 2 | HPF: d (n) = x (2n + 2) −x (2n + 3)] + | [−s (n)
+ S (n + 1) +2] / 4 |

The procedure for obtaining the transform coefficients of the second to fourth layers is the same as in the conventional example, considering the obtained LL1 (8 × 8) as the original image, performing the same transformation and performing the second layer coefficient ( HL2,
LH2, HH2, LL2). Thereafter, the same processing is repeated to obtain the coefficients of the third and fourth layers.

FIG. 2 shows how the coefficients of FIG. 7 are quantized based on the processing procedure of this embodiment. That is, for the HL1 coefficient representing the horizontal edge component, the eight coefficients in the horizontal direction are averaged, and for the LH1 coefficient representing the vertical edge component, the eight coefficients in the vertical direction are averaged. , Respectively. In this replacement, since the edges in the oblique direction have lower image quality importance than the edges in the horizontal and vertical directions, all the coefficients HH1 representing the edge components in the oblique direction are deleted.

FIG. 3 shows a state in which the wavelet transform coefficients are restored according to the present embodiment. That is, the coefficients quantized based on FIGS. 1 and 2 are restored as shown in FIG. 5 at the time of expansion. That is, with respect to the HL1 coefficient and the LH1 coefficient, eight coefficients are replaced with their average values, and all the HH1 coefficients are replaced with “0”. By performing an inverse wavelet transform on this coefficient, an image having no visually significant difference from the original image is restored.

According to the above-described embodiment, a plurality of proximity coefficients having strong correlation are replaced with one coefficient to reduce the data amount. Therefore, the quantization process reduces the data amount and entropy, and as a result, improves the compression ratio. Therefore, it is possible to increase the compression ratio without making the image quality deterioration noticeable.

[0025]

As is clear from the above description, the image forming apparatus and the image forming method of the present invention average and replace the coefficients for every N high-frequency components in the horizontal direction, and replace the coefficients in the vertical direction. The coefficient is averaged and replaced for every N number of band components. The coefficient is quantized by this substitution. Therefore, a plurality of proximity coefficients having high correlation are replaced with one coefficient to reduce the data amount, so that the compression ratio can be increased without conspicuous image quality deterioration.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an example of a processing procedure to which an image forming apparatus and an image forming method of the present invention are applied.

FIG. 2 is a diagram showing a specific example of coefficients after replacement and deletion.

3 is a diagram illustrating a configuration example of image data after restoration of the quantized data of FIG. 2;

FIG. 4 is a block diagram of a conventional four-layer wavelet transform process.

FIG. 5 is a diagram illustrating a configuration example of names of wavelet transform coefficients.

FIG. 6 is a diagram illustrating a specific example of an original image to be converted.

FIG. 7 is a diagram illustrating a configuration example of coefficients generated by wavelet transform.

[Explanation of symbols]

 S1 Original image S2 Wavelet transform S3 Quantization S4 Encoding S5 Storage S6 Decompression & printing

Claims (4)

[Claims] An image forming apparatus for obtaining an image output by quantizing a coefficient obtained by component-division encoding of an input signal, wherein said coefficient is determined for every N (N is a natural number of 2 or more) high-frequency components in a horizontal direction. The image data compression processing method of averaging and replacing the coefficients for every N (N is a natural number of 2 or more) vertical high-frequency components, and quantizing the coefficients. An image forming apparatus comprising: 2. The image forming apparatus according to claim 1, wherein the number of N is eight. 3. An image forming method for obtaining an image output by quantizing coefficients obtained by component-division encoding of an input signal, wherein the coefficient is set for every N (N is a natural number of 2 or more) high-frequency components in the horizontal direction. And a substitution step of averaging and replacing the coefficient for every N (N is a natural number of 2 or more) high-frequency components in the vertical direction. An image forming method comprising: 4. The image forming method according to claim 3, wherein the number of N is eight.