JPS60214159A - Picture signal processing system - Google Patents

Abstract

PURPOSE:To satisfy both resoluting power and gradation performance by using a specific threshold value at a region where a spatial frequency is a prescribed value or over and using the dither method in other regions so as to apply binary-coding. CONSTITUTION:An analog picture signal from a scanner 1 is converted into a digital picture signal, which enters a matrix circuit 3, signals X1-X4 extracting picture information of 2X2 picture matrix are inputted to a binary-coding circuit 4 and a mean value circuit 5 and inverting signals X'1-X'4 are inputted to a spatial frequency discriminating section 7. The binary-coding circuit 4 uses a specific threshold value so as to apply binary-coding to the signals X1-X4 and stores the result to a buffer memory 8. The mean value circuit 5 calculates a mean value Oij of the signals X1-X4 and inputs the result to a comparison section 6. The comparison section 6 compares the threshold value of a matrix with a mean density Oij and opens an AND gate 12 when the spatial frequency is a prescribed value or over, and opens an AND gate 13 when the frequency is less than the prescribed value.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image signal processing method applied to a case where a halftone image is output in binary black and white in a digital printer.

[Prior art]

When outputting a halftone image with a digital printer, the image signal must be binarized in advance. In that case, a dither method is generally employed. In a broad sense, this dithering method is like a systematic dithering method.

One pixel of the image corresponds to one component of the threshold matrix, and 2
It includes not only the most basic method of value conversion, but also its modified density pattern method and partial matrix method.

To put it simply, the density pattern method is a method that compares the average density of a matrix of images of the same size as the threshold matrix with each component (threshold) of the threshold matrix, and generates a binary image of the same size as the threshold matrix. be. The partial matrix method is an intermediate method between the basic dither method and the density pattern method, and compares the average density of an image matrix of the same size as the partial matrix of a threshold matrix with each component (threshold value) of the corresponding partial matrix. By doing this, a binary image of the same size as the partial matrix is generated.

Conventional image signal processing methods use either the basic dither method or the density pattern method or both described above, but as the gradation of intermediate tones increases, the resolution decreases2, and edges become blurred. However, the problem was that the reproducibility of characters and line drawings was poor. Five different improvements have been attempted to solve these drawbacks. Even if 1. (a) Adopted in the basic differential f the method 1. ．． 7. Depending on the spatial frequency of the image fi [7
．． A method of changing the threshold of the 111 matrix (Japanese Unexamined Patent Publication No. 111-78275); Method for changing the size of (JP-A-57-159173
5), (c) Basically use one density pattern -j
; f', , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,, Well, I'm in one.

Shi-, Kashi1. (a) Method cannot remove the harmonic interference 1 included in the original image information, and the size of the threshold value Ma]-RiX! 1st place - Koshika cannot express sharpness. The (eye) method is not expected to have much effect on contour enhancement and resolution improvement.
．． Also, since the processing power is less than +M, it is easy to speed up the processing. (C) Method (J, LQ by averaging the original image information
= 8-degree error in image information: Δ can be reduced, but in the case of line drawings, noise is generated due to the binarized output by the basic dither method.

〔the purpose〕

The present invention has been made in view of the above points! The turtle is a tamono,
The objective is to provide an image signal processing method that can satisfy resolution and gradation at the same time and is capable of high-speed processing.

〔Example〕

An embodiment of the present invention will be described in detail below.

The first numeral 7 shows the block configuration of an apparatus according to an embodiment of the present invention. In the figure, numeral 1 is a scanner that reads the original image. The scanner 1 serially outputs an analog image signal indicating density information of the document that has been decomposed. This analog image signal is converted by the A/[) converter 2 into a 6-hit (65 gradations with white as the third) digital image signal (hereinafter simply referred to as image hX), which is then manually input to the matrix circuit 3. . 1 This 71 helix circuit 3 is
2x2 pixels 71-Rikus on the original image (part 71-
This is a circuit for extracting image information (same size as the image data), and its configuration will be explained with reference to Fig. 2.

Figure 2 shows JE-C120 and 21, each with 12 lines worth of stroke f+'.
It is M. The image signal output from the A/D converter 2 is delayed by one line by one RAM 20, and then input to the RAM 2], and (l) σ) R
It is further delayed by one line by AM:21 and is input to the latch 25, tL. The output of latch 23.25 is then coupled to latch 24.26. These latches latch inputs in synchronization with pixel transfer and lock of the image signal, or at the timing of L1 clock CLK.

1, the output value of the latch 23124+2512G 5'X1+X7+Xi+Xa is as shown in FIG. 3:! x2 pixel 71-x, the f factor σ of the pixels with the same sign is reduced by [-1]. This signal X1.・・、
x4tt, the binarization circuit 11 in FIG.

In addition, the inversion of the characters is ￥fX'1. ..., X'4 is manually input to the spatial frequency determination section 7. 22 is a scanning counter, and the address A stored in RAM 20.21
One R occurs.

Figure 4 shows the addresses A l) R, Kurotsuta C, I-K,
and RAM 20, 21
- timing diagram of signal WE;

The binarization circuit 4 uses a specific threshold value -C, signal Xl.

..., binarize X4. The binarized signal is stored in the buffer memory 8.

The average value circuit 5 calculates the average value of the signals X, . . . , X4, -
)*lJ, 2X2+iM&'3? average of h'J7s is 78
M (ll'IO1, is calculated by '31. As shown in FIG. Imaka W ware that creates harmony 2
9, and a full adder 27 that manually calculates the output values thereof. Average value (-)4. If so, it is input to the comparison output section 6.

In this example, the part 71 in the broad sense de, r law
-The risk method is adopted. Threshold 7 trins and L C
, (8×8 matrix 7 shown in Figure 5 is set in the comparison output section. This matrix is 2×2 matrix 7
1 helix Dll+ DH+ DI3+...
D4), [) is divided into 44 parts. fL, the comparison output unit 6 calculates each component (threshold value) of one partial matrix D, J corresponding to the 2×2 pixel matrix extracted by the matrix circuit 3, and the average density value 0.fL of the two same pixel matrices. J and generates an output cuff 1-rix of 2×2 pixels. This output matrix has an average density value ≧
The pixel corresponding to the component (threshold value) is L I II
, the pixel corresponding to the component that becomes the average density value is ``0''
′. The information of this output matrix, that is, the partial matrix I.
The binarized signal obtained by the RIX method is output from the comparison output section 6 serially, pixel by pixel. A binarized signal from the binarization circuit 4 is output from the buffer memory 8 in synchronization with the output.

The spatial frequency determination section 7 receives the signal X'1. ......,X
'4. Based on the average density values o and J, it is determined whether the spatial frequency of the 2×2 pixels 71 to RIX extracted by the matrix circuit 3 is greater than or equal to a predetermined value, and an L level is output only when it is greater than or equal to the predetermined value. . FIG. 7 shows the configuration of the spatial frequency determination section 7. In the figure, 30. . . . , 33 is an adder, each of which receives the density signal Xl of each pixel of the 2×2 pixel matrix.
l..., the difference between X4 and its average density value oIJ is calculated. Each adder 30. . . . , an OR gate 34 .・・・・・・
, 37 are input to a NOR gate 38. This NOR gate 38 outputs the spatial frequency determination result.

The explanation continues with reference to FIG. The output signal of buffer memory 8 is applied to one input of AND gate 12, and the output signal of comparison output section 6 is applied to one input of AND gate 13. The output signal of the spatial frequency determining section 7 is applied to the other input of the AND gate 13. AND gate 1
A signal obtained by inverting the output signal of the spatial frequency determining section 7 by an inverter 11 is applied to the other input of the device 2. A signal obtained by ORing the output signals of the AND gates 12 and 13 with an OR gate 14 is sent to the outside as a final binary signal.

That is, in a 2×2 pixel matrix where the spatial frequency is equal to or higher than a predetermined value, the AND gate 12 side is opened, so the binarized signal by the binarization circuit 4 is selected and output to the outside. When the spatial frequency is less than a predetermined value, the AND gate 13 side is opened, and the output of the comparison output section 6, that is, the binary signal based on the partial matrix method, is sent to the outside.

Note that since the output from the OR gate 14 is not serial in the main scanning direction, it generally needs to be converted into a serial output in the main scanning direction at a later stage, but since the conversion is easy, a description thereof will be omitted.

In the above embodiment, the size of one portion 71 to 2
×2, but the size may be changed as appropriate. Considering the spatial frequency determination error, it is advantageous to have a large portion 71 to risk, but from the viewpoint of processing speed, it is preferable that the partial matrix be small. Therefore, in general, 2×2 or 4×4 is selected for portions 71 to RIS.

Furthermore, although the partial matrix method among the dither methods is employed in the above embodiment, it is also permitted to change to the density pattern method or the basic dither method.

Furthermore, although the above embodiment is composed of hardware,
It is also possible to realize this through program processing using a microprocessor or the like.

〔effect〕

As explained above in detail, the present invention performs binarization using a specific threshold in a region where the spatial frequency is higher than a predetermined value,
In other areas, binarization is performed using dithering methods (partial matrix method, density pattern method, basic dithering method), so when expressing halftone images with digital printers, both resolution and gradation are satisfied. This has advantages such as good reproducibility of characters and line drawings in photographs, and because the processing is simple, high-speed processing can be easily realized. In addition, according to the mode in which the partial matrix method is adopted as in the above embodiment, since the processing is changed for each small partial matrix, extra dots are eliminated as in the case of adding conventional edge enhancement and normal halftone processing. does not occur.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an apparatus according to an embodiment of the present invention, FIG. 2 is a block diagram of a matrix circuit, FIG. Figure 4 is a timing diagram related to the matrix circuit, Figure 5 is a cross-section diagram of the average value circuit, and Figure 6 is the threshold value 7.
1-Concepts for explaining risk [31, Figure 7 is a diagram of the spatial frequency determination section] [Toscanner/-62-A/D converter, 3/71-Risk circuit, 4... Binarization circuit, 5... Average value circuit, 6
Comparison output section, 7. Spatial frequency determination section, 8 Buffer memory +1.
・Inverter, 12゜13-ANrjkey 1-, 1
4 N 1 R game-1. First cause, Figure 2, Tiger [ Figure 3, Ward 5, 0.) House X! X2 ,,X5 X4 Fig. G Fig. 4-k 1-, J 1. J-'-1" □--Low 1-Shi"----1-"-L"□ 1[)RC2X2]Tsugu)σ Fig. 7

Claims (1)

[Claims] (1) A first means that binarizes the image signal obtained by photoelectrically converting the original image using a specific threshold value, a second means that binarizes the image signal using a dither method, and a second means that converts the image signal into an original image. a third means for determining the height of the spatial frequency of the region to be binarized;
If the third means determines that the spatial frequency is equal to or higher than a predetermined value, the binarized signal produced by the first means is selected and output, and if not, the binarized signal produced by the second means is selected and output. and a fourth means for outputting an image signal.