JPS6180971A - Picture signal processor - Google Patents

Abstract

PURPOSE:To obtain a picture signal processor which can execute satisfactorily smoothing with respect to a halftone dot image of a wide range, by executing a signal processing corresponding to the number of halftone dot lines with respect to a picture signal which has been read with regard to a halftone dot image. CONSTITUTION:An analog picture signal 23 from an image sensor 21 is A/D- converted by an A/D converter 27 and supplied simultaneously as a digital picture signal 28 of 6 bits and 64 gradations to a halftone dot line number/ character area deciding circuit 29, a smoothing circuit 31 and an MTF correcting circuit 32. The net point line number/character area deciding circuit 29 discriminates whether a notice picture element belongs to a character area or a halftone area, outputs the first deciding result signal 33, also discriminates the number of halftone dot lines with respect to an area which has been discriminated as a halftone image, and supplies the second discriminating result signal 35 to a coefficient table 36. Based on this second discriminating result signal 35, a coefficient of a space frequency filter is selected and supplied to the smoothing circuit 31, and smoothing of a halftone dot image of a wide range is executed satisfactorily.

Description

Detailed Description of the Invention "Industrial Application Field" The present invention relates to a reading device that reads an image from an image information source in which a halftone image is formed by halftone printing or the like. The present invention relates to an image signal processing device for smoothing.

``Prior Art'' In an image processing device that reads a document using an image sensor and records or displays it using a quantized image signal, when the document contains character images and halftone images, each part is Image processing is carried out by distinguishing between the two. This is because pixel-by-pixel binarization processing (simple binarization processing) only reproduces text images such as characters and line drawings well.
While the reproducibility of halftone image parts such as halftone images is impossible, halftone images can be reproduced well if condensed halftone processing such as dither processing is performed, but the reproducibility of character image parts becomes impossible. This is because it will lead to deterioration. Here, the halftone image is
This refers to an image part where gradation is expressed by halftone dot printing or an image part where so-called "shading" is performed by a printer.

Now, as a conventional technique for distinguishing between a character image portion and a halftone image portion, the device disclosed in Japanese Patent Application Laid-Open No. 58-3374 is a typical example.

This device (method) divides the document surface (screen) into blocks consisting of multiple pixels, detects the pixel with the maximum and minimum density level within each block, and calculates the difference in density level between them. demand. If this difference is larger than a predetermined value, it is determined that the area is a binary image area or a character image area; otherwise, it is determined that it is a grayscale image area or a halftone image area.

By the way, some manuscripts use halftone dots to reproduce halftones or emphasize specific parts of text. If a document part using such halftone dots (halftone image) is read as is and subjected to image processing, moiré will occur due to the relationship between the halftone dot repetition period and the sub-scanning period, which will impair image reproducibility. This will cause extreme deterioration. Therefore, conventional techniques have been used to detect the halftone image portion existing on the document and to change image processing between this portion and other portions. A typical halftone dot detection device for this purpose is the technique described in Japanese Patent Laid-Open No. 58-218271. This device compares a binary image signal obtained by scanning a document with a detection signal consisting of a black and white code string corresponding to the positions and sizes of the white dots and black dots on the document. . By determining whether these signals match, the halftone image portion is distinguished from other non-halftone image portions.

In this way, conventional methods have been to distinguish between ■character image parts and halftone image parts, and ■distinguish between halftone image parts and non-halftone image parts. used for processing.

``Problem to be Solved by the Invention'' By the way, the density or gradation of a halftone image is determined by the area percentage of the halftone dots in a unit area, but the number of halftone dots per unit area (the number of halftone lines or screen number of lines)
It is usually set arbitrarily in the range of 50 to 200 lines/inch depending on the printed matter. Despite this, conventional image signal processing devices perform uniform smoothing processing on halftone image regions. For this reason, when strong smoothing is performed, rough halftone dots are removed (smoothed), but on the other hand, smoothing becomes excessive for fine halftone dots.
The drawback was that the sharpness of the image was lost. On the other hand, when weak smoothing is performed, rough halftone dots cannot be removed, resulting in the above-mentioned 7-area problem.

In view of these circumstances, it is an object of the present invention to provide an image signal processing device that can satisfactorily smooth a wide range of halftone images.

1-Means for Solving the Problems" In the present invention, as shown in principle in FIG. The image signal processing device is equipped with a signal processing means 12 that performs signal processing on the image signal read for the halftone image in accordance with the number of lines determined by the halftone line number determination means 11.

When the signal processing means 12 performs smoothing using a spatial frequency filter, this filter coefficient may be changed according to the number of halftone dots.

"Example" The present invention will be described in detail below with reference to Examples.

FIG. 2 shows the image signal processing device of this embodiment. An image sensor 21 of this device scans a document 22 as shown in FIG. 3, for example, and outputs an analog image signal 23.

Here, the original 22 shown in FIG. 3 has a first halftone image area 24 printed with a halftone dot number of 100 lines/inch, a character area 25 where text is printed, and a halftone dot image area 25 with a halftone dot number of 150 lines/inch. It is assumed that the second halftone dot image area 26 is composed of three different printing areas.

The A/D converter 27 inputs an analog image signal 23 in synchronization with a video clock (not shown) and converts it into 6 bits (64 bits).
The digital image signal 28 is converted into a digital image signal 28 (gradual density). The digital image signal 28 is transmitted to a halftone dot number/character area determination circuit 29;
The signal is simultaneously supplied to each of the smoothing circuit 31 and the MTF correction circuit 32.

The halftone line number/character area determining circuit 29 determines whether the pixel to be processed (hereinafter referred to as the pixel of interest) falls within the character area or belongs to an area other than the character area, that is, a halftone image area. Then, these discrimination results are outputted as a first discrimination result signal 33.

The first determination result signal 33 becomes a control signal for the multiplexer 34. The halftone dot number/character area determining circuit 29 also determines the number of halftone dots for the region determined to be a halftone image. This determination result is supplied to the coefficient table 36 as a second determination result signal 35.

FIG. 4 shows a circuit portion for generating the first determination result signal 33 in the halftone dot number/character area determination circuit 29. This circuit part is data extraction circuit 3
8, a maximum density difference detection circuit 39 and a comparator 41. The data extraction circuit 38 consists of nine pixels P in a 3×3 matrix structure as shown in FIG.
. Their concentration data d for ~Pe. -d8 is extracted. Here, pixel Pa is the pixel of interest, and
2 is scanned in a plane, and this pixel of interest Po is also shifted in order.

In the maximum concentration difference detection circuit 39, the Ω degree data d1 to d8
Take the difference from the concentration data do for each of
Comparator 41 uses the maximum value as maximum value data 42.
Output to. Pixel of interest P. belongs to the halftone image area, the pixel of interest Pa and the adjacent pixels P1 to P
Since there is little difference in image density between 8 and 8, this maximum value data 42 has a small sensitivity. On the other hand, the pixel of interest P.

When belongs to the character image area, this maximum value data 42 takes a large value at the edge of the character or line drawing.

The comparator 41 compares such maximum value data 42 with a predetermined threshold value βTM, and when the maximum value data 42 is larger than the threshold value nT+(, the pixel of interest P.

The determination result that the character belongs to the character image area is output as a first determination result signal 33. Further, when the maximum value data 42 is equal to or smaller than the threshold value 17H, the pixel of interest P. The first determination result is that the image belongs to the halftone image area.
The determination result signal 33 is output as □. Of course, the circuit portion for creating the first discrimination result signal 33 can be modified in many other ways.

On the other hand, FIG. 6 shows the halftone line number/character area determination circuit 29.
3 shows a circuit portion for creating the second discrimination result signal 35 in FIG. A digital image signal 28 from the A/D converter 27 is input to a shift register 44 in this circuit section.

The shift register 44 has a 33-stage configuration, and accumulates image signals of 16 pixels on each of the left and right sides, including the pixel to be subjected to image processing (pixel of interest).

The area of image signals stored in this shift register 44 will be referred to as a signal processing area.

The image signal of one word processing area is inputted in parallel to the Fourier transform circuit 45 and subjected to Fourier transform. The converted output is input to a 4-value conversion circuit 46 for each spatial frequency component and is 4-valued. 2 output from each quaternary circuit 46
The bit 41 serialized data is used as address information for judgment R.
The data is supplied to an OM (read only memory) 47.

The judgment ROM 47 has pattern data corresponding to the number of halftone dots input manually, and the number of halftone dots corresponding to the contents of the address information is read out as a 1'1] fixed result and output as a second judgment result signal 35. do.

Now, assume that the digital image signal 28 obtained by reading the first halftone image area 24 is input to the shift register 44. The Fourier transform circuit 45 performs Fourier transform on the image signal group of the signal processing area having a halftone dot number of 100 lines/inch. The Fourier transform output has a waveform with a peak approximately at the center of the spatial frequency corresponding to the repetition frequency of the halftone dots, as shown in FIG. On the other hand, shift register 4
If the digital image signal 28 obtained by reading the second halftone image area 26 in FIG. 4 is input manually, the Fourier transform output will be as shown in FIG. 8, and the peak of the waveform will shift to the high frequency side. The position where this peak occurs is 150
This corresponds to the number of halftone lines per inch. Note that in a photographic image where there are no halftone dots, the peak will move to an infinite position.

In this way, the waveform pattern after Fourier transform changes depending on the number of dots, so these patterns and the judgment RO
The number of halftone dots can be determined by matching the pattern data stored in M47. The determination ROM 47 of this embodiment divides the number of halftone dots into a plurality of groups, determines which of these groups it matches, and outputs this as the second determination result signal 35. .

When the second discrimination result signal 35 is input manually, the coefficient table 36 selects filter coefficients of the spatial frequency filter according to the number of halftone dots, and supplies the selected filter coefficients to the smoothing circuit 31 . In the smoothing circuit 31, for a halftone dot number of 100 lines/inch, the peak portion of the waveform caused by the halftone dots is removed using a spatial frequency filter having frequency characteristics as shown in FIG. 9, for example.
Smooth the image signal. For a halftone dot number of 150 lines/inch, a spatial frequency filter having characteristics as shown in FIG. 10, for example, is used. The image signal 51 output from the smoothing circuit 31 is supplied to the multiplexer 34 as an image signal after processing in the halftone image area.

On the other hand, in the MTF correction circuit 32, the digital image signal 2
8 to M T F (klodulation Tran)
(Sfer Function) Correction is performed on the character image portion for resolution deterioration caused by the optical system. The image signal 52 output from the MTF correction circuit 32 is also supplied to the multiplexer 34.

The multiplexer 34 selects one of the two types of image signals 51 and 52 supplied temporally in parallel according to the content of the first discrimination result 1 word 33, and selects this as the image signal 53. Output. That is, when the halftone line number/character area determination circuit 29 determines that the pixel of interest belongs to a character image portion, the image signal 62 output from the MTF correction circuit 32 is output as the image signal 53, and This ensures the sharpness of the image. Further, when the halftone line number/character area determination circuit 29 determines that the pixel of interest belongs to the halftone image part, the image signal 51 smoothed by the smoothing circuit 31 is converted into the image signal 53.
, and appropriate smoothing for the halftone dots will be performed. The image signal 53 output from the multiplexer 34 is supplied to a recording section and a display section (not shown), and the image is reproduced.

In the embodiment described above, the character image (elephant) and halftone image were distinguished by a process different from the determination of the number of halftone dots, but the waveform after Fourier transformation of the character image is the same as that of the halftone image. Since there is a large difference, by writing the pattern data of the character image in the judgment RO; Vl in advance,
It is also possible to distinguish between both images.

Further, in the embodiment, the number of halftone dots is determined by the apparatus itself, but the image signal processing apparatus may be provided with a switch for selecting the number of halftone dots, and the number of halftone dots may be selected by an operator's operation.

"Effects of the Invention" As described above, according to the present invention, appropriate smoothing is performed according to the number of halftone lines, so high-quality image processing can be performed without unnecessary deterioration of image sharpness. can.

[Brief explanation of drawings]

Figure 1 is a block diagram showing the principle of the present invention, Figures 2 to 1
Fig. 0 is for explaining one embodiment of the present invention, Fig. 2 is a block diagram showing a schematic configuration of an image signal processing device, Fig. 3 is an explanatory diagram showing an example of a document, and Fig. 4 is a block diagram of the circuit part that creates the first discrimination result signal in the halftone dot line number/character area discrimination circuit, FIG. A block diagram of the circuit part that creates the second discrimination result signal in the dotted line number/character area judgment circuit, Fig. 7 is a characteristic diagram showing the frequency characteristics of a relatively coarse halftone image, and Fig. 8 is a graph showing the frequency characteristics of a comparatively coarse halftone image. A characteristic diagram showing the frequency characteristics of a thin halftone image, FIG. 9 is a characteristic diagram of the smoothing filter for the halftone image shown in FIG. 7, and FIG. It is a characteristic diagram of a filter for smoothing. 11... Halftone line number determination means, 12... Signal processing means, 29... Halftone line number/character/area judgment circuit. , 31
... Smoothing circuit, 36 ... Coefficient table, 45 ... Fourier transform circuit, 47 ... ROiVI for judgment 0 applicant
Representative of Fuji Xerox Co., Ltd.
Patent Attorney Umeo Yamanouchi Figure 1 Figure 3 Figure 4 Figure 5 Figure 2 Monthly Monthly Monthly Edition Figure 7 Akira 9 Zuko 1 So-called Taikyo lmata Figure 8 Figure 10

Claims (1)

[Claims] 1. A reading device that reads an image information source that includes at least a part of a halftone image area formed by halftone images,
halftone dot number determining means for determining the number of lines constituting halftone dots in the halftone dot image area; An image signal processing device characterized by comprising: a signal processing means for performing signal processing according to the signal processing method. 2. Image signal processing according to claim 1, characterized by comprising a signal processing means for changing filter coefficients of a spatial frequency filter for smoothing a halftone image according to the number of halftone lines. Device.