JP2002247386A - Color image-processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image-processing device having a frame memory for efficiently using a general-purpose memory chip. SOLUTION: The color image-processing device comprises a mode selection section 8 for selecting either a full-color mode or a black-and-white mode, a color conversion circuit for converting the color of RGB data, a compression/ decompression section 5 for compressing color conversion data in 2×2 pixel block units, a selector for selecting compression data or data before compression, a memory control circuit 6 for storing selected data at a frame memory 7, the compression/decompression section 5 for decompressing data read from the frame memory 7, a selector for selecting either the decompression data or data before decompression, and an inverse color conversion circuit for inversely converting the color of the selected data to RGB data. The two selectors select compression and decompression data when the full color mode is selected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image processing apparatus such as a digital color copying machine and a color facsimile for forming an image in a full-color and monochrome mode.

[0002]

2. Description of the Related Art Generally, in a digital copying machine, C
The original image is read pixel by pixel using a CD image sensor or the like, the analog electric signal obtained from the output of the image sensor is A / D-converted, and the obtained digital signal is subjected to various processes. Give the copy image. In a digital color copying machine, a document to be copied is separated into R (red), G (green), and B (blue) and read. Based on the image data, Y (yellow), M (magenta), (Cyan), K (Black)
A color image is reproduced by using the color materials.

One such color image forming apparatus is disclosed in Japanese Patent Application Laid-Open No. 63-107274.
The apparatus disclosed in this publication includes a step of winding a recording sheet around a transfer drum, forming images corresponding to a plurality of color materials used for recording on a photosensitive drum in a face-to-side sequence, and transferring the image to the recording sheet.
Repeat for each type of color material used. That is, when a color image is reproduced with four colors of YMCK, four image forming steps are repeated.

On the other hand, the performance of color copiers has also been improved, and the pixel density is about 600 dpi, and the copy speed is about 50 cpm (copy / min) for a single color even in the aforementioned one-drum type image forming apparatus (printer). Has been put to practical use. In order to improve the copy speed and the pixel density of the image forming apparatus, a method of securing an exposure amount by switching a beam used for each line by using a plurality of writing laser beams has been put to practical use. .

On the other hand, in a document reading apparatus (scanner), the pixel size is reduced due to an increase in the pixel density, so that the amount of light is insufficient, and image data of an appropriate noise level (S / N ratio) is obtained. Cannot increase the reading speed too much. Therefore, there is a problem that the speed of the scanner is lower than that of the printer, and the copy speed is limited by the speed of the scanner.

In order to solve this problem, a method has been proposed in which document read data is temporarily stored in a frame memory, and image data is read from the memory at a high speed in accordance with the writing speed of a printer, thereby forming a high-speed image. . In this case as well, the copy speed decreases when one copy is made per document, but when multiple copies (N copies) are made per document, the effect increases as N increases. .

Further, when a plurality of printers are connected, data stored in the frame memory can be read out and transmitted to the plurality of printers to make copies in parallel. However, storing a large amount of image data for one page requires a large amount of memory, which causes a problem of high cost.

[0008] The data compression technique is used there, but there is a trade-off relationship between the compression ratio and the quality of the compressed / decompressed image data. In the case of a color document, even if a certain amount of compression is necessary, even for a black and white document which originally requires a small amount of data, there is a problem that image deterioration occurs due to compression / expansion.

For this reason, according to the technique disclosed in Japanese Patent Application Laid-Open No. 9-102878, compression / expansion processing is performed for a color original, and black / white data is stored in a memory without compression for a monochrome original. Reading is performed.

[0010]

However, according to the compression method disclosed in the publication, there is a problem that a large waste is generated in the memory chip, and the effect of cost reduction cannot be sufficiently obtained. This problem will be described. The maximum image data size is set to 12 inches × 17 inches in consideration of A3 size and DLT (double letter) size. Assuming that the pixel density is 600 dpi and the RGB colors are 8 bits / pixel, the data amount is (12 · 600) · (17.6)
00) .8 (bit / color) .3 (color) = 1,7
62,560,000 bits = 1680.9 M bits. When this is compressed with a fixed length of 1/3 compression ratio, 16
80.9 Mbit / 3 = 560.3 Mbit of memory is required.

In Japanese Patent Application Laid-Open No. 9-102878, since compression is performed at a compression ratio of 1/3 for each block of 4 × 4 = 16 pixels, 4.4.8 (bits / color). ) / 3 = 128-bit data amount. Therefore, assuming that a memory chip having a word length of 16 bits is used, a memory chip having a multiple of 128/16 = 8 is required.

In this case, since the total capacity is 560.3 Mbits, if a memory chip having a 64 Mbits capacity is used, 560.3M / 8 / 64M = 1.09, which is rounded up to 2 sets × 8 = 16. Chips are required. 12
Using an 8 Mbit memory chip, 560.3
Since M / 8 / 128M = 0.54, round up to 1
Set × 8 = 8 chips are required. Therefore, the mounted memory capacity is 64M · 16 = 1024Mbit,
128 M · 8 = 1024 M bits, and in each case, as much as 45% of the required amount of 560.3 M bits is wasted.

An object of the present invention is to provide an image processing apparatus having a frame memory which can efficiently use a general-purpose memory chip.

[0014]

In order to achieve the above object, according to the first aspect of the present invention, an original is color-separated into RGB and digitally read, and Y is converted from the read image data.
In a color image processing apparatus that digitally reproduces a color image using MCK color materials, a mode selection unit that selects a full color mode or a black and white mode, a color conversion unit that performs color conversion of RGB data, and color conversion data Is 2 × 2
Compression means for compressing each pixel block, first data selection means for selecting compressed data or data before compression,
Memory control means for storing the selected data in the memory, decompression means for decompressing the data read from the memory, second data selection means for selecting between decompressed data and data before decompression, Reverse color conversion means for performing reverse color conversion of data into RGB data, wherein the first and second data selection means include:
The present invention relates to a color image processing apparatus that selects compressed or decompressed data.

According to a second aspect of the present invention, the compression ratio in the color mode is set to 1/3.

According to the third aspect of the present invention, the color conversion means converts the RGB data to YUV data [Y = (R + 2G
+ B) / 4, U = RG, V = BG].

Further, in the invention described in claim 4, the data stored in the memory when the monochrome mode is selected is Y data.

According to the fifth aspect of the present invention, the data stored in the memory when the monochrome mode is selected is G data.

According to the first, second, and third aspects of the present invention, a frame memory that can efficiently use a general-purpose memory chip is configured. In addition, a black-and-white copy image with visually favorable gradation characteristics can be obtained at low cost. In addition, a black-and-white copy image with sharp image quality can be obtained at low cost.

According to the present invention, the uncompressed data is efficiently stored in the memory in the black and white mode. Also, a compression / decompression circuit with low cost and high image quality is configured.

[0021]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a color image processing apparatus according to an embodiment of the present invention. First, the overall configuration and operation will be described. The original is C in the scanner section 1.
The data is read by a CD (image sensor) 2, converted into an electric signal, subjected to shading correction by a shading correction unit 3, and then output as 8-bit data for each of RGB. The RGB data is usually temporarily stored in the frame memory 7 of the frame memory unit 4.

In the frame memory section 4, when the color mode selection signal is in the full color mode in synchronization with the reading operation of the scanner section 1, the compression / expansion section 5 performs a compression process. 7 is stored. When the color mode selection signal is in the monochrome (or single color) mode, the frame memory 7 remains uncompressed.
Is stored. Here, the color mode selection signal is output from the mode selection unit (actually, the operation panel is used) 8.

When the original reading operation is completed and one page of image data is stored in the frame memory 7, the operation proceeds to an image forming operation. Image data is read out from the frame memory 7 in synchronization with the image forming operation of the printer, and in the full color mode, the compression / expansion unit 5 performs expansion processing in accordance with the color mode selection signal specified at the time of storage. The data is converted into RGB data and output. In the monochrome (or single color) mode, the data is output as it is as RGB data. However, since the uncompressed data stored in the black and white mode is data for one color such as luminance data or G, the data is output as R = G = B (read data) as RGB color data common data.

The image processing section 9 receives the RGB data read from the frame memory 7 by the synthesizing circuit 10. The synthesizing circuit 10 is a circuit for synthesizing data from the frame memory 7 and data of another document read by the cascade unit 1 at the time of reading. However, the scanner unit 1 does not normally perform a reading operation at the time of reading data from the frame memory 7. , And selects data from the frame memory 7 and sends it to the spatial filter circuit 11 at the next stage.

The spatial filter circuit 11 enhances the sharpness of the character by applying edge enhancement to the character on the image, or performs the smoothing process on the halftone image to perform the dither pattern processing in the subsequent gradation processing. Used to prevent interference with
Edge enhancement / smoothing processing may be adaptively switched within an image for one page using the result of an image area separation circuit (not shown).

The next color correction circuit 12 is a circuit for performing a masking process for correcting the distortion of the color separation characteristics of the scanner unit 1 and the distortion of the spectral characteristics of the YMC color material used in the printer. The next UCR / UCA circuit 13 is a circuit for calculating an amount of replacing a common part of CMY with a K color material. The following gamma correction circuit 14 is a correction circuit that corrects fluctuations such as differences between printers, changes over time, and the like so that predetermined gradation characteristics can be obtained. The last gradation correction circuit 15 is a circuit for performing predetermined area gradation processing such as dither processing as necessary to secure a predetermined gradation.

The KCMY data after the gradation processing is sequentially sent to the printer 16 to form a full-color copy image.
However, in the monochrome mode, only the K data is
And the image formation is completed in a single color. In the other single color mode, only the necessary color plane data is sent to the printer 16, and the image formation is completed with only the corresponding color.

Next, the compression / decompression circuit 5 will be described.
FIG. 2 is a block diagram of a general encoding circuit (compression circuit). The RGB data read by the scanner unit 1 is subjected to encoding processing in units of 2 × 2 pixel blocks 21. In order to improve the quality of the compressed image, the RGB data is converted into Y data by the color conversion circuit 22 in pixel units according to the procedure shown in FIG.
Color conversion to UV data. Here, Y is a luminance component,
UV is a color difference component.

Since the RGB signals have a large correlation between the color components, the encoding efficiency is poor if each color is compressed independently. By performing color conversion to luminance / color difference YUV data having low correlation between components and compressing the converted YUV data, image data with little deterioration can be obtained at the same compression ratio. In the next S conversion circuit 23, the low frequency component LL and the high frequency component H
(HL, LH, HH). Ab in FIG.
cd is an identification symbol for four pixels of a 2 × 2 block which is a coding unit. ABCD represents the value of the abcd pixel of each of the three components of YUV.

In the quantization circuit 24, the data after the S-transform is vector-quantized to reduce the data amount. Here, in order to further improve the quality of the compressed image, the area is separated by the area separation circuit 25 using the Y high-frequency component (Y_H). In a region (non-edge portion) such as a photograph portion where a high number of bits are allocated to the
It is configured so that many bits are allocated to the low frequency component.

FIG. 5 shows bit allocation after quantization. An identification signal of an edge portion or a non-edge portion is assigned to the head, and the data is coded in a total of 32 bits in both regions. Since the original data is 8 bits × 3 colors × 4 pixels = 96 bits, the compression ratio is 32/96 = 1/3.
In FIG. 5, the number in parentheses is the number of bits. The first one bit is an area signal, which is 1 in a non-edge area and 0 in an edge area.

FIG. 6 is a block diagram of a general decoding circuit (decompression circuit). The coded data read from the frame memory 7 is divided for each component by the dividing circuit 3 in accordance with the area signal, and the inverse S-transformation procedure shown in FIG.
The data is decoded by the conversion circuit 32 into YUV data corresponding to each 2 × 2 pixel. The YUV data is returned to RGB data by the reverse color conversion circuit 33 according to the reverse color conversion procedure of FIG.

FIG. 7 is a block diagram showing a first example of the encoding circuit of the present invention. In the present invention, data compression is not performed when the color mode is the monochrome mode. Therefore, in the example of FIG. 7, in the monochrome mode, Y (luminance) data is used as the data for the monochrome mode. By using luminance data as data for a black-and-white (single color) image, it is possible to obtain a natural-looking black-and-white copy with good consistency with visual density sensitivity when copying a color document as a black-and-white image.

In the circuit shown in FIG. 7, the color mode selection signal is converted by the selector 26 (first data selection means) into Y after color conversion.
Data is selected as data to be stored in the frame memory 7. In the full-color mode, the fixed-length coded data compressed to the ３ data amount is stored in the frame memory 7.

FIGS. 8A and 8B are diagrams showing the bit allocation of uncompressed data in the black and white mode. FIG. 8A shows the bit allocation when Y (luminance) data is used as the black and white data, and FIG. The bit allocation when G (green) data is used as data is shown. Both 2 × 2
= Data representing a total of 4 pixels, 8 bits x 4 = 32
Is a bit.

In FIG. 8A, the luminance data corresponding to the pixel position abcd is stored in the frame memory 7 as 8 bits × 4 pixels = 32 bits data. The timing of the non-compressed data is matched with the timing of the encoding process by the delay circuit 27. This eliminates the need for the image processing unit 9 at the subsequent stage to perform position correction for the difference in delay amount between the color mode and the monochrome mode when reading memory data.

FIG. 9 is a block diagram showing an example of the decoding circuit of the present invention. The data read from the frame memory 7 is also subjected to a decoding process in parallel regardless of the color mode. In the black-and-white mode, the decoded data is meaningless as image data and is discarded, and the selector 34 (second data selection means) selects the data converted to RGB as the uncompressed data.

As described above, in the monochrome mode, uncompressed RGB data is used as common data as it is. That is, Ra = Ga = B_a = Y_a, R_b = G_
b = B_b = Y_b, R_c = G_c = B_c = Y_
c, R_d = G_d = B_d = Y_d. Also,
The uncompressed data division circuit 35 also performs a delay process for adjusting the timing when the decoding process is performed.

By adopting a compression ratio of 1/3 as the encoding process, the amount of uncompressed data for the black and white mode becomes the same, so that the memory is used together with the delay circuit 27 of the preceding encoding circuit. It is possible to support both full-color and black-and-white color modes without changing the operation of the control circuit 6 at all.

FIG. 10 is a block diagram showing a second example of the encoding circuit of the present invention. Here, G data before color conversion is used as data in the monochrome mode. Compared to the luminance data, the sensitivity to R and B is lower, and the matching with the visual shading sensitivity is slightly deteriorated. However, it is affected by the displacement of the pixel data of three channels of RGB and the difference of the MTF characteristic. Therefore, there is an advantage that bleeding or blurring of an edge portion of a black-and-white document hardly occurs.

This is particularly advantageous when a large number of originals including images having many edges such as characters are copied. The configuration may be such that the user can select whether to use the luminance data or the G data in the monochrome mode according to his / her preference. In the circuit of FIG. 10, the G data before color conversion is stored in the frame memory 7 as it is in the monochrome mode.

FIG. 8B shows the bit allocation of the uncompressed data. As in the case of FIG. 7, the uncompressed data is delayed by the delay circuit 27 and stored in the frame memory 7 in order to match the timing with the encoding process. When reading data from the frame memory 7, the circuit shown in FIG. 9 can be used as in the case where a luminance signal is used. In this case, not only the G data but also the RB data is output as RGB data instead of the original G data.

In the present invention, a 2 × 2 pixel size is adopted as a pixel unit for the compression / expansion processing. In this case, the amount of memory chips required as the frame memory 7 is calculated. Since the data amount per unit block is 32 bits,
If a 16-bit word-structured memory chip is used, 32/16 = 2 chips are used. Since the required memory amount is 560.3 Mbits, a 64 Mbit chip requires 560.3 M / 2/6.
4M = 4.37, which means that five sets and two = 10 memory chips are rounded up.

When using a 128 Mbit chip,
560.3M / 2 / 128M = 2.18, and three sets of 2 = 6 memory chips are required after rounding up. 4
In the case of a × 4 block unit, 64M bits, 128M
When 16-bit chips are used, 16 and 8 memory chips are required, respectively, and it can be seen that 6 and 2 chips can be saved respectively. This is because, as the number of bits of the unit block increases, the unit of use of the chip increases, and thus an extra amount exceeding the necessary amount increases.

Therefore, the memory efficiency can be improved by reducing the amount of data that accesses the frame memory 7 at one time. For this purpose, a method of dividing the unit data and accessing the frame memory 7 in a time-division manner will be considered. 2x
Since 2 is 32 bits and 4 × 4 is 128 bits, the frame memory 7 can be mounted with the same memory efficiency by dividing it into 128/32 = 4 times.

The access time to the frame memory 7 is 4 pixel clocks when the block length is 4 × 4 and 2 pixel clocks when the block length is 2 × 2. In order to access the memory by dividing it into four in the case of 4 × 4, it is necessary to access 32-bit data with 4/4 = 1 pixel clock. This is twice as fast as accessing 32 bits of data with two pixel clocks in the 2 × 2 case.

Therefore, even if the number of memory chips to be used is the same, a high-speed chip corresponding to double-rate access, that is, an expensive chip is required. Therefore, the frame memory 7 can be configured using a small amount of inexpensive memory chips by employing the compression processing of 2 × 2 pixel size.

[0048]

According to the first, second and third aspects of the present invention, it is possible to configure a frame memory that can efficiently use a general-purpose memory chip. In addition, a black-and-white copy image having visually favorable gradation characteristics can be obtained at low cost.
Further, a black-and-white copy image with sharp image quality can be obtained at low cost.

According to the fourth and fifth aspects of the present invention, uncompressed data can be efficiently stored in the memory in the monochrome mode. In addition, low cost, high quality image compression
An expansion circuit can be configured.

[Brief description of the drawings]

FIG. 1 is a block diagram of a color image processing apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram of a general encoding circuit.

FIG. 3 is an explanatory diagram of color conversion.

FIG. 4 is an explanatory diagram of S conversion.

FIG. 5 is an explanatory diagram of a quantization method with a compression ratio of 1/3.

FIG. 6 is a block diagram of a general decoding circuit.

FIG. 7 is a block diagram illustrating a first example of an encoding circuit according to the present invention.

FIG. 8 is an explanatory diagram of bit assignment of uncompressed data in a monochrome mode.

FIG. 9 is a block diagram showing an example of a decoding circuit of the present invention.

FIG. 10 is a block diagram showing a second example of the encoding circuit of the present invention.

[Explanation of symbols]

 Reference Signs List 1 scanner unit 2 CCD 3 shading correction circuit 4 frame memory unit 5 compression / expansion unit 6 memory control circuit 7 frame memory 8 mode selection unit 9 image processing unit 10 synthesis circuit 11 spatial filter circuit 12 color correction circuit 13 UCR / UCA circuit 14 gamma correction circuit 15 gradation processing circuit 16 printer

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) H04N 1/60 B41J 3/00 B 5C079 1/41 H04N 1/40 DF term (reference) 2C262 AA24 AA26 AA27 AB20 AC02 AC04 BA02 CA09 DA17 EA17 GA14 2H027 EE08 EE10 EF09 FA28 FB06 2H030 AA02 AD07 AD12 AD13 5C077 MP08 PP32 PP33 PQ08 PQ22 RR21 TT06 5C078 AA09 BA44 CA02 CA27 DA01 DA02 5C079 HA13 HB01 NA03 H03 HB12 H03 HAB

Claims (5)

[Claims] 1. A color image processing apparatus for color-separating a manuscript into RGB and digitally reading the same, and digitally reproducing a color image from the read image data using a YMCK color material. A mode selecting means for selecting a mode, a color converting means for performing color conversion on RGB data, a compressing means for compressing the color converted data in units of 2 × 2 pixel blocks, and a second means for selecting compressed data or data before compression. 1 data selection means, memory control means for storing the selected data in the memory, decompression means for decompressing the data read from the memory, and second data for selecting between decompressed data and data before decompression Selecting means, and reverse color converting means for performing reverse color conversion of the selected data into RGB data,
A color image processing apparatus, wherein the first and second data selection means select compressed or decompressed data when the full color mode is selected. 2. The color image processing apparatus according to claim 1, wherein the compression ratio in the color mode is set to 1/3. 3. The color conversion means converts RGB data into YUV data [Y = (R + 2G + B) / 4, U = RG, V = B
-G]. The color image processing apparatus according to claim 1, wherein 4. The color image processing apparatus according to claim 1, wherein the data stored in the memory when the monochrome mode is selected is Y data. 5. The color image processing apparatus according to claim 1, wherein the data stored in the memory when the monochrome mode is selected is G data.