JPS62100884A - Picture processor - Google Patents

Abstract

PURPOSE:To detect an edge accurately even with a variable density picture of light colors of the same group and also synchronously with the reading of the picture data, by recognizing whether or not a noticed picture element is equal to the edge of the picture information from the difference of color information between the noticed picture element and a picture element in the vicinity of said noticed picture element. CONSTITUTION:A picture is scanned and separated (10) into three primary colors R, G and B to be written to buffers 12-14. The picture element selecting circuits 15-17 extract the data on a noticed picture element and the picture element adjacent to the noticed picture element and deliver them to a color difference arithmetic circuit 18. The circuit 18 squares the difference between the noticed picture element and its adjacent picture element for those color elements R, G and B respectively and adds these differences together. Then the output of the circuit 18 is compared with the threshold and a threshold processing circuit 19 by which '1' is delivered when the output of the circuit 18 is larger than the threshold and an OR circuit 20 which decides an edge when the outputs of each threshold processing circuit are equal to '1' are provided.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image processing apparatus having a function of detecting edges of color image data.

[Prior Art] Conventionally, edge detection in this type of device includes an image memory that stores all of one screen of a read image, and a synchronization signal is used when inputting and outputting image data by reading it. Detection processing was performed independently.

Other devices perform edge detection using only a specific color signal among the three color separation signals of color image data, or independently determine monochromatic edges between each color separation signal, and Based on the results, edges in the color image were detected.

[Problems to be Solved by the Invention] However, the former method does not have the immediacy of synchronizing with a scanning synchronization signal, and requires a large-capacity storage device, which limits its use. It was unsuitable for applications such as in terms of cost and equipment size. In addition, the latter has difficulty in detecting edges in images based on shading of similar colors or in detecting edges in images between light colors.

f stage for solving the problems] The present invention was made with the aim of solving the problems of the above-mentioned prior art, and as a means for solving the problems, for example, A calculation means for calculating a color difference (u'j) corresponding to the difference in each color information from a pixel in the vicinity of the pixel, a comparison means for comparing a value calculated by the calculation means with a predetermined value, and a comparison means in the lower stage of the comparison. and recognition means for recognizing whether or not the pixel of interest is an edge of image information based on the result.

[Operations] In such a configuration, the calculation means synchronizes with the scan synchronization signal of the image information and uses as a calculated value a value corresponding to the sum of the squares of the color information of the pixel of interest and pixels in the vicinity of the pixel of interest, thereby obtaining a good image. Performs data edge detection.

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

FIG. 1 is a block diagram of an embodiment according to the present invention,
In the figure, 10 scans the document surface and separates the read color image data into a red signal (hereinafter referred to as R) 22, a green signal (hereinafter referred to as G) 23, and an iY multiplied signal (hereinafter referred to as B) 24. Water mo synchronization signal (H3YNC) 25 and vertical synchronization signal (
VSYNC) 26 is a scanning data source that is output to the R buffer 12, G buffer 13, and B/header 14 in synchronization with each other. As shown in FIG. 2, the scanning data source 10 scans the read original 1 line by line and sequentially outputs image data together with each periodic signal. 11 is a buffer control circuit;
HSYNC25 and VSYNC from scanning data source 10
Control is performed to input the color separation signals R22, G23, and B24 into each buffer (12 to 14) in synchronization with 26, and control to read out the color separation signals written into the light from each buffer (12 to 14). 12 is an R buffer that stores 4 lines of R22 (4 lines in the main scanning direction);
3 is a G buffer that stores 4 lines of G23; 1
4 is a B buffer that stores 4 lines of B24. Also, 15-17 are buffers for each color (12-14)
Based on the pixel information from each pixel selection circuit (15 to 17), a pixel selection circuit extracts pixel data and pixel data adjacent to the pixel of interest from the readout data and outputs it to the color difference calculation circuit 18. A color difference calculation circuit 19 calculates the color difference between the pixel of interest and each pixel adjacent to the pixel of interest; reference numeral 19 compares the color difference information calculated by the color difference calculation circuit 18 with a threshold value added in advance, and outputs the comparison result; The value processing circuit 20 is an OR circuit for the comparison results from the threshold processing circuit 19, and the output from the OR circuit 20 is the edge detection signal (2) of the pixel of interest.
7).

Reference numeral 21 denotes an image processing section which can take in each color separation signal (28) from each sofa (12 to 14) and perform image processing for other purposes with reference to the edge detection signal 27.

Details of each of the above-mentioned buffers (12 to 14) are shown in Figure 753. Note that all the buffers (12 to 14) have the same configuration.

In FIG. 3, 32 is a buffer selector A.

Reference numerals 33 to 36 represent buffers A to D capable of storing color-separated image data for one life of each main scan, and reference numeral 37 represents a buffer selector B to 7. A color signal 38 (R22, G23, or B24) color-separated and output from the scanning data source 10 is input to a buffer selector A32. Buffer control circuit 11
is rotated by the VSYNC signal 26 to buffers A33 to /
One of the buffers D36 is selected and the color signal 38 sent in synchronization with the HSYNC signal 25 is sequentially written into the selected buffer. At this time, the color signals written in the light are sequentially read out from the remaining three buffers in synchronization with the H5YNC signal 25 and output to the buffer selector B37. The buffer selector B37 compares the color signals read from each buffer with the currently sent scan data, and selects the read data from the buffer storing the previous scan data as the color signal A39. The data read from the buffer storing the scan data two trees ago is output as the color signal B40, and the data read from the buffer storing the scan data three trees ago is output as the color signal C41. That is, while the current scan data is being input, the data at the corresponding main scan position of the three most recent scan data adjacent to each other are output as the respective color signals A to C (39 to 41).

Each color signal output in this way is transmitted to each pixel selection circuit (1
5 to 19). Each pixel selection circuit (15 to 17
) all have the same configuration, and the details of this pixel selection circuit are shown in FIG. In the figure, latches 42 to 50 each hold data for one pixel, and latch data for nine pixels (3×3). Here, the pixel of interest is indicated by Xo.

Eight pixels x1 to x8 adjacent to the pixel of interest Xo are simultaneously output and sent to the color difference calculation circuit 18.

FIG. 5 shows the states of each pixel selected by the pixel selection circuits 15 to 17. Each pixel of the original image data (I o −
Ia) are separated into each color (Ro~R8), (Go-
Ga) and (Bo”B8) are output simultaneously, and the pixel of interest I
Data of each of the eight pixels adjacent to each pixel of interest Ro, Go, and Bo, which have been separated into o color, is extracted.

In this embodiment, an example has been described in which each pixel adjacent to the pixel of interest Xo is extracted by the pixel selection circuit (15 to 17).
If the configuration is such that each of the 8 pixels adjacent to o can be read out simultaneously with one addressing, this pixel selection circuit (15 to 1
7) can be omitted.

Next, details of the color difference calculation circuit 18 are shown in FIG.

In the figure, 61 to 68 are color difference calculation circuits 1 to 8, 69 to 76 are threshold processing units 1 to 8, and 20 are logical sum circuits. Each of the color difference calculation circuits 1 to 8 (61 to 68) contains color-separated target pixels Ro, Go, and Bq, and adjacent pixels Rj, Gj, and Bj.
(j=1 to 8) are respectively input. The color difference calculation circuits (61 to 68) calculate the sum of squares of the difference between the input three-color separation color signal of the pixel of interest and the three-color separation color signal of the pixel to be calculated, and calculate the sum of squares of the difference between the three-color separation color signal of the input pixel of interest and the three-color separation color signal of the pixel to be calculated, and calculate the sum of squares of the difference between the input three-color separation color signal of the pixel of interest and the three-color separation color signal of the pixel to be calculated. 76). Each threshold processing section compares the square sum signal from the color difference calculation circuit with a predetermined threshold and outputs the result to the OR circuit 20.

The configurations of the color difference calculation circuit and the threshold value processing section are all the same, and the details of these circuits are shown in FIG. In the figure, 80 is a color difference calculation circuit, and 90 is a VIA value processing section.

Ro, Go, and Bo of the pixel of interest are input to the subtracters 81 to 83 of the color difference calculation circuit 80, respectively, and the adjacent pixel signals of Rj, Gj, and Bj are input to the other of each subtractor 81 to 83. It has been entered. The pixel signals input to each of the g calculators 81 to 83 are subjected to subtraction processing, and the pixel of interest is subtracted from the adjacent pixel. Next, multipliers 84 to 86 calculate the square value of the subtraction result, and all square values are sent to adder 87.8.
8, and the summed output is output to the threshold processing section 90. Here, the three color separation values (Ro, Go, Bo) of the pixel of interest and the three color separation values (Rj) of the pixels in the vicinity are output to the threshold processing unit 90.

The sum of squares (ΔEj) 2 of the differences of each decomposition value between Gj, Rj) (j=1.z,..., 8) is (ΔEj) 2 = (Rj-Ro) 2+
(G j-Go) 2 + (B j-Bo
) becomes 2.

In the threshold processing section 90, the comparator circuit 92 calculates this sum of squares (
ΔEj)2 is compared with a threshold THj (j=1, 2...8) preset in the threshold setting section 91, and (ΔE
j) If 2>THj, that is, (ΔEj)2 is larger, the output signal is set to "i" (high level).The threshold set in this threshold setting section 91 is set for each of the three color separation values. Also, the optimum value is selected for each adjacent pixel position.

In this way, the color difference calculation between the pixel of interest and the eight pixels adjacent to the pixel of interest is performed between each pixel, and the results of comparison with each threshold value THj are sent to the OR circuit 20, and all the threshold values are The output of processing units 1 to 8 (69 to 76) is 1'''
(high level), the output signal 2 of the OR circuit 20
7 is output 1 This becomes an edge detection signal indicating that this is the edge pixel of the pixel of interest image take.

In addition, in the following explanation, an example was described in which color difference calculation is performed between 8 pixels adjacent to the pixel of interest, but the number of pixels for which color difference calculation is performed is not limited to this. pixel, or furthermore, 5 pixels near the pixel of interest.
The configuration may be such that the calculation is performed for every pixel in the ×5 matrix, and the number of pixels for color difference calculation is not limited. moreover,
With the logical sum of all the outputs of the agony 16 processing units 1 to 8,
Instead of making edge detection reliable, it is quiet 1a processing parts 1 to 8.
It may be determined that an edge has been detected when two or more and seven or less of the outputs are High.

Further, the color difference calculation circuit is not limited to the configuration shown in FIG. 9, but may be configured with a look-up table, a gate array, or the like.

Furthermore, the read take obtained by scanning the halftone printed material may be smoothed before being input to the R buffer 12, G buffer, and B buffer 14, and then input to each buffer. .

As explained above, according to the present embodiment, by focusing on the value of the sum of 2 tones of the differences between the color separation signals, edge detection of a color image is determined in consideration of the difference in hue brightness of the input image. As a result, extremely accurate edge detection can be performed. In addition, this edge detection can be performed by scanning the read data from the scanning data source in one line, making real-time and accurate processing possible, and image memory that stores all read data in bit matter for edge detection. It is possible to create an image processing device that does not require the following.

[Effects of the Invention] As explained above, according to the present invention, with a simple configuration, very accurate image take edges can be detected even in image data of similar colors e-light and image data of light colors. I can do it.

Furthermore, edge detection can be performed substantially in synchronization with reading of image data, etc., making it possible to realize a device that achieves faster processing and smaller size.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment according to the present invention, FIG. 752 is a diagram showing an example of scanning image data in this embodiment, FIG. 3 is a detailed configuration diagram of the buffer shown in FIG. 1, and FIG. A detailed configuration diagram of the pixel selection circuit shown in FIG. 1, FIG. 5 is a diagram for explaining a pixel matrix that performs color difference calculation using color separation in this embodiment, and FIG. 6 is a color difference riit calculation circuit shown in FIG. 1. , and a detailed configuration diagram of the threshold value processing circuit. FIG. 7 is a detailed circuit diagram of the color difference calculation circuit and the threshold value processing section shown in FIG. 6. In the figure, 1...read original, 10...scan data source, 1
1...Buffer control circuit, 12-14.33-36.
... Buffer, 15-17... Pixel selection circuit. 18... Color difference calculation circuit, 19... Silent LA processing circuit, 20... OR circuit, 21... Image processing section, 32
．． 37... Buffer selector, 42-50... Latch, 61-68... Color difference calculation circuit, 69-76...
Threshold processing unit, 81-83... Subtractor, 84-86...
- Multiplier, 87.88... Adder, 91... Threshold value setting section, 92... Comparison circuit. Figure 4 Figure 5 FP series, i, 'ji series 1

Claims (4)

[Claims] (1) Calculating means for calculating a color difference value corresponding to the difference in color information between a pixel of interest and a pixel in the vicinity of the pixel of interest, a comparing means for comparing the value calculated by the calculating means with a predetermined value, and the comparing means An image processing apparatus comprising recognition means for recognizing whether or not a pixel of interest is an edge of image information based on a comparison result by the means. (2) The image processing apparatus according to claim 1, wherein the calculation means calculates a color difference value for each of the three primary colors. (3) The image processing apparatus according to claim 1, wherein the calculation means uses a value corresponding to the sum of squares of color information of the pixel of interest and pixels in the vicinity of the pixel of interest as the calculated value. (4) The calculation process by the calculation means and the edge recognition process by the recognition means are executed in synchronization with a scan synchronization signal of the image information, according to any one of claims 1 to 3. image processing device.