JPH0353767A - Picture data compression and restoration system for print - Google Patents

Abstract

PURPOSE:To improve the transmission efficiency and storage efficiency by providing a restoration color coordinate conversion section applying calculation of inverse matrix with respect to a masking matrix to R, G, B data from a restoration look-up table section, converting them into Y(yellow), M(magenta) and C(cyan) data and outputting the data with a K(black) data. CONSTITUTION:A light part of a positive dot area differs from a pattern of a print original and a human sense, and in this embodiment, a positive dot area of 20-30% or less is defined as a light part and a positive dot area of 70-80% or over is defined as a shadow part respectively. Moreover, a compres sion color coordinate conversion section 2 separates Y', M', C', K' data from a compression look-up table section (LUT section) 1 into the Y', M', C' and the K' data and the calculation of masking matrix comprising [A]=[3Xn] matrix (n: a number depending on the combination of R, G, B) is applied to the Y', M', C' data to convert them into R, G, B data, which are inputted to a compression device 3 together with the K data.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention provides compression and restoration of image data for printing that can improve data transmission efficiency and storage efficiency when transmitting and storing image data for printing. It's about systems.

[Prior Art] With the recent advances in digital image processing technology, total scanners have been introduced in the printing field as well, which convert printing images into digital data and perform plate collection and retouching operations. In this case, it is necessary to transmit the image data for printing (for example, between a publisher and a factory, between factories, between a region and a factory), and store it (in a database). Various methods have been tried, but the best method is method has not yet been proposed. This is because the amount of image data for printing and printing is enormous compared to normal digital data. For example, one A4 size print image costs 15
In order to express it with four color components at the resolution of book/me,
As shown in the following equation, 56 MB of data is required. Here, one pixel is 4 Byte (-8b i
It is assumed that it is expressed by data of t (IBy te)×1 (color).

210+nmX297mmX (15 pieces/related) 2X4B
yte/point - 56MByte Therefore, when transmitting the data for one A4 size printing image, the normal public line is (9600bit/s).
If you use ec), it will take 13 hours, and even if you use a high-speed digital line, it will take just under 2 hours.

In addition, even when saving printing image data to construct a database, optical discs (1 to 20 By t
Even if you use e), the A4 size image data mentioned above is 1
Only 8 to 36 images can be stored, which is insufficient to construct an image database. As described above, printing image data is complicated to handle, and processing time is required when transmitting and storing the data.

On the other hand, in the fields of television and facsimile,
In order to improve the transmission efficiency and storage efficiency of monitor image data, compression of monitor image (still image) data is performed. As a compression device for still images of this type for television, there are recent excellent compression devices (for example, A
A compression device based on the DCT method has been proposed. This ADCT compression device compresses image data into 8x8
After performing two-dimensional discrete cosine transformation on a block-by-block basis, and performing threshold processing and quantization processing on the transform coefficients,
The coefficients are variable-length coded to obtain coded data. Therefore, it is possible to reduce the processing time during data transmission and storage by compressing print image data using such a compression device.

[Problem to be solved by the invention] However, in this case, although the monitor image data is R, G, and B luminance data, that is, the electric signal value (current or voltage) is 8 bit (0 to 255) data. On the other hand, the image data for printing is 8-bit (0-255) data indicating halftone dot percentage (0-100%) on top of Y, M, C, K data. Therefore, if such a compression device is directly applied to compress image data for printing, differences in color reproduction (especially,
In the highlight area (for example, 0 to 20% in bright halftone dots)
occurs. That is, in a certain printed matter, the human eye is sensitive to light parts, but less sensitive to shadow parts. Therefore, there is a problem in that a step, for example, a pattern such as a contour line, is formed in the gradation, and the gradation characteristics are deteriorated (the gradation is not smooth and the tone is poor).

Further, while the monitor image data is R, G, and B data, the print image data is Y, M, C, and K data. Therefore, such a compression device is used for printing image data. If applied directly to compression, there is a problem that a difference in color reproduction occurs, that is, image quality deteriorates during color reproduction (colors differ from original colors, etc.).

The present invention was made to solve the above-mentioned problems, and it is possible to compress and compress image data for printing without degrading image quality during color reproduction and without degrading gradation characteristics.
An object of the present invention is to provide a highly reliable print image data compression/restoration system that can perform restoration and improve transmission efficiency and storage efficiency when transmitting and storing print image data.

[Means for Solving the Problems] In order to achieve the above object, the first invention provides a method for compressing print image data using a compression device and restoring the compressed print image data using a decompression device. Now, as a preprocessing means for the compression device, each Y, M, C, and K data that constitutes the image data for printing is separated into Y, M, and C data, and each Y, M, and C data is A correction operation is performed using a quantization function so as to weight the light part of the halftone dot area, and the data is converted into Y', M', and C'' data having the gradation of the monitor image system. It is provided with a compression look-up table section that outputs each of the data together with the K data, and outputs each data of Y'M'' C-K from the compression look-up table section as Y',
Separate data into M', C-data and Y', M',
For each data of C-, [3×n] matrix (n:
Masking consisting of (number determined by the combination of R, G, B)
It is equipped with a compression color coordinate conversion section that performs matrix calculations to convert each data into R, G, and B data, and manually inputs each data to the compression device together with the K data. The R, G, B, and K data constituting the print image data output from the printer are separated into R, G, and B data, and a masking matrix is applied to each R, G, and B data. It includes a restoration color coordinate conversion section that calculates the inverse matrix of , converts it into each data of Y', M', and C-, and outputs each data together with K data. Y', M',
Separate each data of C', K into Y', M', C-data and data, and perform a correction operation using an inverse function of the quantization function for each data of Y''M', C-. Y, M, C
, and outputs each data together with the K data.

Furthermore, in the second invention, as a pre-processing means of the compression device,
Separate the Y, M, C, and K data that makes up the print image data into Y, M, and C data, and
[3×nl matrix (n:R,
A compression color coordinate conversion unit that calculates a masking matrix consisting of a number determined by a combination of G and B, converts it into R, G, and B data, and outputs each data together with K data, R from the compression color coordinate conversion section
, G, B, and K data into R, G, and B data, and a quantum Compression in which the data is converted into R', G', and B- data having the gradation of the monitor image system by performing a positive operation using a conversion function, and each data is input to the compression device together with the K data. The restoring device also has a look-up table section for processing the R', G', B', and K data that constitutes the printing image data output from the restoring device as a post-processing means for the restoring device. ′
, B = data and separated into data, R', G', B-
A restoring look-up table unit is provided which performs a correction operation on each data using an inverse function of a quantization function to convert it into R, G, and B data, and outputs each data together with K data. , Y,
A restoring color coordinate converter is provided for converting each data into M and C data and outputting each data together with K data.

[Function] Therefore, in the printing image data compression and restoration system of the first aspect of the present invention, weighting is applied to the light portion of the halftone dot area for each data of Y, M, and C. Increase bit allocation to the write section. A correction calculation using a quantization function is performed to allocate a small bit to the shadow area, and Y″M with the Fe tone of the monitor image system is created.
Convert to each data of ``C'', and further convert
[R, G, B
Each data is input to the compression device together with the K data, and the inverse matrix of the masking matrix is multiplied for each R, G, and B data output from the decompression device. Do Y', M', C
-, and then performs a correction operation using the inverse function of the quantization function on each data of Y''M''C'' to convert it to each data of Y, M, and C. It is output together with K data.

In addition, in the printing image data compression/restoration system of the second aspect of the present invention, [3×
n] Perform calculations on a masking matrix consisting of a matrix to convert it into R, G, and B data, and then weight the light part of the halftone/dot area for each R, G, and B data. By performing a correction operation using a quantization function, R-G with the gradation of the monitor image system is
- B- is converted into each data, and each data is converted into K.
The R', G', and B data that is input to the compression device together with the data and output from the decompression device is corrected by an inverse function of the quantization function. The inverse matrix of the masking matrix is calculated on each of the R, G, and B data to convert it into Y, M, and C data, and each of the data is output together with the K data.

This makes it possible to compress and restore image data for printing without degrading image quality during color reproduction and without degrading gradation characteristics.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an example of the configuration of an image data compression system for stamp IJIJ according to the present invention. As shown in FIG. 1, the print image data compression system of this embodiment includes a compression look-up table section (hereinafter referred to as compression LUT section) 1 as a preprocessing means of the compression device, and a compression color It consists of a coordinate conversion section (hereinafter referred to as a compression color coordinate conversion section) 2 and a compression device 3 based on the ADCT method.

Here, the compression LUT unit 1 stores Y (yellow), M (magenta), C (cyan), which constitute the image data of the print document.
, K (black) and convert them into Y,
The data is separated into M and C data, and only for each data of Y, M and C, weighting is given to the light part of the positive halftone dot area, that is, the bit allocation is increased to the light part. , In order to allocate a small bit to the shadow part,
A correction term is performed using a quantization function to convert each data of Y', M=, and C- having the gradation of the monitor image system, and output each data together with the unconverted K data. It is something. In this case, the quantization function has quantization function characteristics as shown in FIG. 2, for example. In addition, the light portion of the positive halftone dot area varies depending on the pattern of the printing document and human perception, but in this example, the positive halftone dot area is 20~
30% or less is defined as a light part, and a positive halftone dot area of 70 to 80% or more is defined as a shadow part.

Further, the compression color coordinate conversion unit 2 converts each data of Y', M', C', and K from the compression LUTH 1 into Y', M', C=
For each data of Y', M', C-, [A]1[3×n] matrix (n:
Masking consisting of (number determined by the combination of R, G, B)
It performs matrix calculations to convert each data into R, G, and B data, and inputs each data to the compression device 3 together with K data. In this case, n is a number determined by the combination of R, G, and B. In this example, for example, as the primary color, a masking pattern consisting of a [3X3] matrix as shown below is used.
It has a matrix.

On the other hand, the compression device 3 includes a luminance/chrominance conversion section 31, a subsampling section 32, an 8x8 blocking section 33, and a DC
T section 34, quantization section 35, and zigzag scanning section 36
and an unequal length encoding section 37. Here, the brightness/color difference conversion section 31 inputs each R, G, and B data output from the compression color coordinate conversion section 2, and converts these into the brightness data Y and the color difference data CR, C. It is converted into and output. Further, the sub-sampling section 32 receives the luminance data Y and the chrominance data CR + from the luminance/color difference conversion section 31.
The K data output from the cas and the compression color coordinate conversion unit 2 are input, and the resolution is changed so that the resolution of the color difference data CR and CB is lower than that of the luminance data Y (
correction). Furthermore, the 8×8 blocking unit 3
3 manually inputs the data from the sub-sampling unit 32,
The amount of data is optimized (8×8 blocks) so that the amount of calculations when performing DCT does not increase.

On the other hand, the DCT section 34 has an orthogonal transform function,
y, cR, c. on the X, Y coordinate axes. Each data of frequency lllI (horizontal, vertical direction) Fx. Each data Y-. on FV. CR', CB-1: to be converted. Further, the quantization unit 35 inputs the data Y', CR', and Ca from the DCT unit 34, and assigns a large bit 1 to the low frequency component and a small bit assignment to the high frequency component. It allocates data as follows.

Further, the zigzag scanning section 36 manually inputs the data from the quantization section 35 and rearranges the data from the low frequency data side to the high frequency data side. Furthermore, the unequal-length encoding section 37 encodes the rearranged data from the zigzag scanning section 36 into unequal-length encoding, and encodes the encoded data, that is, the compressed data Y', C, ゜゜,
It outputs c8°.

Next, the operation of the printing image data compression system configured as above will be explained.

In the figure, Y (yellow), M (magenta), C (cyan), and K (black) make up the image data of the printed original.
Each data is stored in the compression LU before being sent to the compression device 2.
The signal is input to T section 1. In the compression LUT unit 1, these data Y, M, C, and K are separated into Y, M, and C data.

Then, for each of the Y, M, and C data, weighting is given to the light part of the bright dot area, that is, a large bit is assigned to the light part, and a small bit is assigned to the shadow part. As shown in Figure 2, the correction calculation using the quantization function is performed to convert the Y' and M-C' data to have the gradation of the monitor image system, and the K data is converted to nothing. The data is input to the compression color coordinate conversion section 2 as it is without any conversion.

In addition, the compression color coordinate conversion section 2 converts these data Y-
M″C″K is separated into Y″M″C′ data and data. Of these, Y', M', C-
Each data is converted into R, G, and B data by performing a masking matrix operation consisting of [3 x 33 matrices as described above, and the K data is converted into R, G, and B data without any conversion. The data is input to the compression device 3 as is.

On the other hand, in the compression device 3, each of the R, G, and B data from the compression color coordinate conversion section 2 is converted into luminance data Y and color difference data CR, CB in a luminance/color difference conversion section 31, and the subsampling section converts the R, G, and B data from the compression color coordinate conversion section 2. 32. The sub-sampling section 32 uses the luminance data Y and the chrominance data c. from the luminance/color difference conversion section 31. The resolution of the K data from the l ca s and the compression color coordinate conversion unit 2 is changed so that the resolution of the color difference data CR and CB is lower than that of the luminance data Y, and then input to the 8×8 blocking unit 33. Ru. And 8
The X8 blocking unit 33 optimizes the amount of data (converts it into 8×8 blocks) so that the amount of calculations when performing DCT on the data from the sub-sampling unit 32 does not increase, and the DCT unit 34 is manually operated. be done.

On the other hand, in the DCT section 34, Y, which is on the X, Y coordinate axes,
Each data C R and C B corresponds to each data Y゛C. , CB- and input to the quantization unit 35. The quantization unit 35 receives data Y + . from the DCT unit 34 . cR′,
For Co-, data is allocated such that a large bit is allocated to the low frequency and a small bit is allocated to the high frequency. Then, in the zigzag scanning section 36, the data from the quantization section 35 is rearranged from the low frequency data side to the high frequency data side. The data from the scanning unit 36 is unequal-length encoded, and the encoded data is output as compressed data Y"CR', CI+".

As described above, the printing image data compression system of this embodiment uses Y, M, C, K, which constitutes the image data of the printing original, as a preprocessing means of the compression device 3 using the ADCT method.
Separate each data into Y, M, C data and data,
For each Y, M, and C data, a correction operation is performed using a quantization function so as to weight the light part of the positive halftone dot area, and Y″M with the gradation of the monitor image system is performed. ′
, C- and outputs each data together with K data, and converts each data of y', M', ', KC from this compression LUT part 1 into Y', Separate data into M', C'' data and [3X3] for each data of Y=, M', C-.
It is equipped with a compression color coordinate conversion unit 2 that calculates a masking matrix consisting of a matrix, converts it into R, G, and B data, and inputs each data to a compression device 3 together with K data. It is.

Therefore, Y. which constitutes the image data of the printed manuscript. Each data of M and C is corrected by a quantization function so that a large bit is allocated to the light part of the positive halftone dot area, and a small bit is allocated to the shadow part. It is converted into Y''M', C- data with gradation, and then masking consisting of a [3X3] matrix is performed on each Y-M''C- data. , B, and this data is compressed by the ADCT method in the compression device 3, so that a large amount of data is assigned to the light area, which varies depending on the image of the printed document and human senses, and smoothness of the gradation characteristics is achieved. At the same time, it increases the calculation speed during compression processing, and ensures acceptable image quality when reproducing gray color (without causing problems such as colors differing from the original color of the printed original). It becomes possible to compress image data. Thereby, it is possible to significantly improve the transmission efficiency and storage efficiency when transmitting and storing print image data.

In the above embodiment, as pre-processing for compression processing, the compression look-up table section performs a correction operation using a quantization function, and then the compression color coordinate conversion section performs masking and
Although we have described the case where matrix calculations are performed, the preprocessing means of the compression device is not limited to this.
[3×n'l? For compression, it operates on a masking matrix consisting of trixes (n: a number determined by the combination of R, G, and B), converts it into R, G, and H data, and outputs each data together with K data. It is equipped with a color coordinate conversion section, and R, G, B from the compression color coordinate conversion section.
, K, R. G, B data are separated into data, and a correction calculation is performed using a quantization function to weight the light part of the halftone dot area for each R, G, B data, and a monitor image is created. R゛G′ with the gradation of the system
, B-, and inputs each data to the compression device together with the K data.

Furthermore, in the above embodiments, the case where the present invention is applied to a print image data compression system has been described, but the present invention is not limited to this, and the present invention can be similarly applied to a print image data restoration system.

In this case, the printing image data restoration system is A.
As a post-processing means of the restoration device using the DCT method, R, G,
Separate each data of B and K into R, G, and B data, and apply the inverse matrix [A-'] of the above-mentioned masking matrix [A] to each data of R, G, and H. It is equipped with a color coordinate conversion unit for restoration that performs the calculation, that is, the calculation, converts it into each data of Y', M', and C-, and outputs each data together with the K data, and further includes a color coordinate conversion section for this restoration. Y from the department
−． Separate each data of M', C=K into Y', M', C-data and quantization of each data of Y=, M', C-1 as described above in Fig. 2. It may be configured to include a restoring LUT section that performs a correction calculation using an inverse function of the function, converts the data into Y, M, and C data, and outputs each data together with the K data.

In addition, the R', G', B', and K data constituting the printing image data output from the restoration device are R', G',
Separate into B=data and data, R=,G- IM
A lookup for restoring performs a correction operation on each data using the inverse function of the quantization function shown in FIG. - A table section is provided, and the inverse matrix [A-'] of the aforementioned masking matrix [A] is calculated for each of the R, G, and B data from this restoration look-up table section. Y. It may be configured to include a restoration color coordinate conversion unit that converts each data into M and C data and outputs each data together with K data.

On the other hand, in the above embodiment, the case is described in which only one compression LUT section and one compression color coordinate conversion section are provided, but the present invention is not limited to this, and the compression LUT section may have different quantization function characteristics. In addition to each having a plurality of compression LUT units, each of the compression color coordinate conversion units has a plurality of compression color coordinate conversion units having different masking matrices as compression color coordinate conversion units. The optimum compression LUT section may be selected and processed according to the ink hue condition of the printing document, and the optimum compression color coordinate conversion section may be selected and processed according to the ink hue condition of the printing document, resulting in even higher accuracy. This is something that can be obtained. Further, the restoring LUT section and the restoring color coordinate converting section can be configured in exactly the same way.

In addition, in the above embodiment, the area with a bright halftone dot area of 20 to 30% or less is defined as a light part, and the area of a bright halftone dot is 70 to 80% or more as a shadow part, but the light part with a bright halftone dot area is defined as a Since the area differs depending on the picture pattern and human senses, it is not limited to this range in any way; for example, a area with a positive halftone dot area of 15% or less is defined as a light part, and a area with a positive halftone dot area of 85% or more is defined as a shadow part. You can do it like this.

Further, in the above embodiment, a case was described in which a masking matrix consisting of a [3X3] matrix was calculated, but the present invention is not limited to this in any way; for example, a [3X9] matrix (in the case of secondary colors) or [3X6] matrix (for secondary color simple type) or [3X11] matrix (for tertiary color simple type)
It is also possible to perform calculations on a masking matrix consisting of .

[Effects of the Invention] As explained above, according to the present invention, image data for printing can be compressed and restored without deteriorating the image quality during color reproduction and without degrading the contrast properties. It is possible to provide an extremely reliable printing image data compression/restoration system that can improve transmission efficiency and storage efficiency when transmitting and preserving both image data.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the printing image data compression system according to the present invention, FIG. 2 is a diagram showing an example of the quantization function characteristics of the compression LUT section in the same embodiment, and FIG.
The figure is a block diagram showing an example of the configuration of a compression device in the same embodiment. DESCRIPTION OF SYMBOLS 1... LUT section for compression, 2... Color coordinate conversion section for compression, 3... Compression device, 31... Luminance/color difference conversion section, 3
2... Subsampling section, 33... 8x8 blocking section, 34... DCT section, 3 5... Quantization section, 3 6... Zigzag scanning section, 3 7... Unequal Long encoder.

Claims (2)

[Claims] (1) In a device in which print image data is compressed by a compression device and this compressed print image data is restored by a decompression device, Y, M, which constitutes the print image data, is used as a preprocessing means of the compression device. . A look-up table for compression that performs a correction calculation using a function, converts it into Y', M', and C' data having the gradation of the monitor image system, and outputs each data along with the K data. and Y', M', C' from the compression look-up table section.
, K are separated into Y', M', C' data and K data, and for each data of Y', M', C', [
3×n] matrix (n: the number determined by the combination of R, G, and B) to convert it into R, G, and B data, and convert each data into the above data together with the above data. A compression color coordinate conversion section input to a compression device is provided, and as a post-processing means of the restoration device, R, G, B, which constitutes the printing image data output from the restoration device,
Separate each K data into R, G, B data and K data, and apply the masking and masking to each of the R, G, and B data.
By calculating the inverse matrix of the matrix, Y′,
Convert each data into M' and C', and convert each data into the above K
It is provided with a color coordinate conversion section for restoration that outputs the data together with the data, and Y', M', C',
Separate each data of K into Y', M', C' data and perform a correction operation on each data of Y', M', C' by an inverse function of the quantization function. Y, M,
1. A printing image data compression/restoration system comprising a restoration look-up table section for converting each data into C data and outputting each data together with the K data. (2) In an apparatus in which print image data is compressed by a compression device and this compressed print image data is restored by a decompression device, Y and M constituting the print image data are used as preprocessing means of the compression device. , C, K data is separated into Y, M, C data and K data, and a [3×n] matrix (n: combination of R, G, B A compression color coordinate conversion unit is provided which calculates a masking matrix consisting of a number determined by , converts it into R, G, and B data, and outputs the data together with the compression color coordinate conversion unit. R from the coordinate transformation section,
The G, B, and K data are separated into R, G, B data, and K data, and a quantum A compression unit that performs a correction calculation using a conversion function to convert R', G', and B' data having the gradation of a monitor image system, and inputs each data to the compression device together with the K data. A look-up table unit is provided, and as post-processing means of the restoring device, R', G',
Separate each data of B', K into R', G', B' data and K data, and perform a correction operation on each data of R', G', B' by an inverse function of the quantization function. Do R
. The inverse matrix of the masking matrix is calculated for each data to obtain Y,
1. A printing image data compression/restoration system comprising a restoration color coordinate conversion unit that converts each data into M and C data and outputs each data together with the K data.