JPH11266364A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image processor which can detect a dot low in number of lines with high resolution by performing peak detection of a dot area in an inputted image with a specific matrix and changing a separation condition of the dot area according to the resolution of the input image. SOLUTION: Peak detection 2 and 3 processing detects the peak of a dot by using two matrixes of different large and small sizes. There are peak detection 2 that uses a large matrix 5×5 and peak detection 3 that uses a small matrix. Respective peak detection 2 and 3 detects a mountain peak of dots and a valley peak. Because the matrix size is large in the peak detection 2, the peak of dots low in number of lines can be detected. Because a matrix is small in the peak detection 3, dots of a high line number can be detected. A dot deciding part 4 finally decides whether or not to be a dot area. Thus, it is possible to solve resolution dependency of dot area detection.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus such as a copying machine, a multifunction peripheral, a printer, and a facsimile, which includes a halftone dot area separating method for detecting a halftone dot area in an image.

[0002]

2. Description of the Related Art In a printed document or the like, when gradation is required, such as in a photograph, a method of expressing gradation in a pseudo manner by changing the size of a halftone dot is used. Newspaper photos are examples of this. The halftone dots have the number of lines with different intervals between the dots, and the higher the number of lines, the smoother the gradation. However, if the performance of the output device or the recording paper is low, a high number of lines may not be used, and a halftone dot having a different number of lines is used for each printed document.

A halftone dot separation method for detecting a halftone portion of a printed document includes, for example, a peak peak of a halftone dot using an N × N matrix as disclosed in Japanese Patent Application Laid-Open No. 2-123478. And the valley peak are detected, and from the relationship between the number of peaks included in the block of N pixels × N pixels whose size is set in advance and the number of peaks of the block of interest, whether or not the block of interest is a halftone dot region There was something to decide.

However, as the resolution of an input image increases, the actual size of one pixel decreases. The interval between the halftone dots is a fixed distance, but as the resolution increases, the number of pixels between them increases, and the reference range of the matrix for the document becomes smaller. When the reference range becomes small, it greatly affects the detection of a peak and the determination of a halftone dot area. This effect is greater as the halftone dots with a lower screen ruling have a greater distance between the points. In the prior art, since the matrix size is arbitrary but the size is constant, there is a disadvantage that the higher the resolution of the input image, the lower the number of halftone dots can be detected.

For example, in the case of an input image of 400 dpi, 3
It is apparent that a halftone dot up to 100 lines can be detected with a × 3 matrix, but if a matrix of the same size is used in an input image of 600 dpi, only a halftone dot of up to 120 lines can be detected. Further, the area separation is sometimes used to change the content of the image processing depending on the separation result and obtain a good output image. For example, there is a case where it is desired to perform a simple binarization process on a character portion and perform an error diffusion process on other portions. In such a case, an edge detection process is performed to detect a character portion. However, if the original includes a photograph and characters composed of halftone dots, the halftone portion of the photograph is detected as an edge, and the character is detected. In some cases, the same binarization processing as that performed by the unit is performed. When a halftone image is simply binarized, moire (a type of image distortion) is likely to occur. Normally, halftone dots are detected and removed from the character portion to reduce moiré, and the halftone dot region is set so as to be included in the non-character portion, and error diffusion processing is performed.

[0006]

However, if the original contains a halftone portion of the number of lines around the boundary of the detectable range, the non-character portion region and the character portion region are mixed. If the above processing is performed, the image becomes a mixed image of simple binarization and error diffusion, and the image quality is significantly reduced. Halftone dots on the low screen ruling are less likely to suffer from moiré even if simple binarization processing is performed. Rather, binarization processing may have better image quality than error diffusion processing. That is, it is considered that the image quality of the output is improved by preventing the halftone dots having the low screen ruling from being detected as the halftone dot region.

The prior art has a problem in that a document including the number of lines around the number of lines that can be detected in a halftone dot region is liable to cause erroneous separation, which may degrade image quality. Also, when the matrix size is increased, it is usually necessary to change the area detection parameters at the same time. However, even if the matrix size is simply increased and the most appropriate parameter is set, a small character may be detected as a halftone dot, and the halftone dot region cannot be detected with high accuracy. When the resolution is high, it is useful to use a plurality of detection matrices having different sizes, but when the resolution of the input image is low, two matrices are not necessarily required to detect a halftone dot portion. If several matrices are used unnecessarily, unnecessary portions may be detected as halftone dots. For example, if the resolution of the input image is 600 dpi
In the case of (2), it is necessary to use two of the 5 × 5 and 3 × 3 matrices, but in the case of 400 dpi, one 3 × 3 matrix is sufficient. If a matrix of 5 × 5 and 3 × 3 is used at 400 dpi, small characters may be detected as a peak depending on the conditions. At this time, when the resolution of the input image changes, a portion which is not a halftone dot may be detected. And the disadvantage that the detection accuracy is reduced remains.

Accordingly, a first object of the present invention is to provide an image processing apparatus capable of detecting a halftone dot having a low frequency even at a high resolution. A second object of the present invention is to provide an image processing apparatus in which moiré is less likely to occur in a portion where a photograph and a character are mixed. A third object of the present invention is to provide an image processing apparatus which detects a dot area with high accuracy without detecting a small character as a halftone dot.

[0009]

According to the present invention, a reading means for reading an original, an image processing means for performing predetermined image processing on an image read by the reading means, and the image processing means And an image output means for outputting the image data obtained in the step (a), wherein the image processing means performs peak detection with respect to a halftone dot area in the input image using an N × N matrix. The first object is achieved by providing a change unit capable of performing halftone dot region separation and changing a halftone region separation condition according to the resolution of an input image. In the invention described in claim 2, in the invention described in claim 1, the first object is achieved by changing the matrix size.

According to a third aspect of the present invention, in the first or the second aspect of the present invention, the image processing apparatus further comprises a ruling side changing means for changing a ruling side having a lower detectable halftone frequency. Achieve two objectives. According to a fourth aspect of the present invention, in the third aspect of the present invention, the number-of-lines changing means includes means for associating the number of lines with a matrix size and setting a matrix size in a peak detection process.
The second object is achieved.

According to a fifth aspect of the present invention, in the first aspect of the invention, the third object is achieved by detecting a halftone dot by using a plurality of detection matrices simultaneously. In the invention described in claim 6, in the invention described in claim 5, the third object is achieved by the plurality of peak detection matrices having different matrix sizes. According to a seventh aspect of the present invention, in the fifth or sixth aspect, the third object is achieved by changing the number of matrices to be used depending on the resolution of the input image.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to FIGS. FIG.
FIG. 3 is a diagram showing an example of a double configuration of the overall configuration of the facsimile terminal. The reading unit includes a reading scanning unit 11, a reading processing unit 12, a line memory 15, and a line memory control unit 16.
The recording unit includes a recording scanning unit 13 and a recording processing unit 1
4, a line memory 15 and a line memory control unit 16.

The image information read by the reading scanning unit 11 is temporarily stored in a line memory 15 via a reading processing unit 12. According to the mode of the information compression unit 17, the redundancy is removed while reading out the image information from the line memory 15, and the image information is stored again in the RAM (random access memory) 22 through the data bus 25. This RAM
22 is used as a transmission buffer, and the communication control unit 19 is used again.
Via the modem 18. The signal modulated by the modem 18 is sent out to the line through the network control unit 24.

Conversely, image information received from the line passes through the network control unit 24, is demodulated by the modem 18, is output to the data bus 25 via the communication control unit 19, and is stored in the RAM 22. The RAM 22 is used as a receiving buffer, and is sequentially input to the information restoring unit 17 via the data bus 25. It is called again from the line memory 15, passes through the recording processing unit 14, and is reproduced as image information by the recording scanning unit 13. For image information, the network control unit 24,
It simply passes through the communication control unit 19, but the communication control information is interpreted and understood in the communication control unit 19, performs execution processing such as transmission control and error control, and performs non-volatile processing when journal information is required. It is stored in the RAM 22 having a sexual function.

A paging signal for controlling the network is controlled by a network control unit 24. Further, image information that does not require information compression and decompression is taken out from the reading processing unit 12 and the line memory 15 via the data bus 25 directly to the RAM 22 or the like, and can perform image processing and image communication.
Further, the information obtained by image processing or the information obtained by image synthesis is directly input, or is superimposed on the image information.
5 for recording and reproduction. The line memory 15 can be used separately from the information compression / decompression unit 17, and at the same time, it is also conceivable that the compressed information is decompressed and output via the line memory 15 irrespective of the reading scan and the recording scan. . The MPU 23 performs overall system control, information flow management, communication control, and network control.

The main processing of the system control is the panel unit 2
0 is control related to the man-machine interface, and control related to the mechanical configuration of the reading unit and the recording unit by the mechanism control unit 21. The panel unit 20 greatly differs in the number of input / output terminals and the control method depending on the types and types of panel switches and panel displays. The mechanism control unit 21 changes the types and quantities of drivers and sensors depending on the method and mechanical configuration of the reading unit and the recording unit, and affects the control complexity.

FIG. 2 is an example of a block diagram of the halftone dot area detection processing, and FIG. 3 is an example of pixels of a matrix. The peaks (peaks and valleys) are detected using an N × N matrix using the density difference between the surrounding density and the target pixel. Here, 2
The resolution of the input image is 400
A matrix of 3 × 3 pixels at dpi and 600 dpi
A matrix of 5 × 5 pixels for i is prepared.
When the resolution of the input image is notified from the system to the matrix determination unit, the matrix determination unit sets a matrix size corresponding to the resolution in the peak detection unit. The peak detection unit detects a halftone dot peak under the following conditions.

FIG. 4 is a diagram showing an example of block scanning. Next, in the halftone dot area determination unit, when at least one peak of a peak or a valley exists in a block, one peak of the block is counted. Here, the block size is 4
× 4 pixels. ABCD is a set of one block, and EFGH and IJKL are similarly considered. If the peak count of each block set is n or more (for example, 3 or more), the target block is determined to be a halftone dot area.

The peak detection conditions for a 3 × 3 matrix are as follows. Resolution of input image is 400dp
When i, the density p of the pixel of interest, and the densities m1 to m8 of the peripheral pixels, the peak and valley peak detection conditions are p <m1 and p <m2 and p <m3 and p <m4 and p <m4 and p <m5 and p If <m6 and p <m7 and p <m8, a valley peak is set. If p> m1 and p> m2 and p> m3 and p> m4 and p> m5 and p> m6 and p> m7 and p> m8, the peak is taken. Others are non-peak.

The conditions for detecting a peak in a 5 × 5 matrix are as follows. Assuming that the resolution of the input image is 600 dpi, the density of the pixel of interest is p, and the densities of the surrounding pixels are n1 to n8, the peak and valley peak detection conditions are p <n1, p <n2, p <n3, and p <n4 and If p <n5, p <n6, p <n7, and p <n8, a valley peak is set. If p> n1 and p> n2 and p> n3 and p> n4 and p> n5 and p> n6 and p> n7 and p> n8, the peak is taken. Others are non-peak.

FIG. 5 is an example of a block diagram of a halftone dot area detection process when the number of lines is also measured, and FIG. 6 is an example of a matrix pixel. Here, the resolution of the input image is 600 dpi
And Peak detection is the same as in the above method. Input 6
The default value of the matrix size at the time of 00 dpi is set to 5 × 5 in advance. In this case, 6
A halftone dot area of about five lines is detected. If 65 lines are used in the original document and you want to increase the number of lines that can be detected to 120 lines or more due to misseparation of halftone dots or user preference for image quality,
The user instructs the system to increase the number of lines through the operation panel. It is assumed that the correspondence between the number of lines and the matrix is set in advance in the number of lines-matrix correspondence section as follows. The system instructs the line number-matrix corresponding unit to change the number of lines, and the line number-matrix corresponding unit instructs the peak detector to indicate the size of the matrix size. After that, it is the same as the above method.

When detecting a halftone dot area of 120 lines or more,
A matrix for peak detection is set to 3 × 3. When a halftone dot area of 85 lines or more is detected, the matrix of peak detection is set to 4 × 4. When a halftone dot area of 65 lines or more is detected, a matrix for peak detection is set to 5 × 5.

FIG. 7 is an example of a block diagram of halftone dot area detection when a plurality of matrices are used at the same time, and FIG. 8 is an example of pixels of the matrix. In the peak detection processes 2 and 3,
The halftone dot peak is detected using two matrices of different sizes. Here, a large matrix 5 ×
5 and a peak detector 3 using a small matrix. Each peak detector detects peaks and valleys of halftone dots. However, the input image 1 is a gray scale image. In the peak detection 2, since the matrix size is large, the peak of the halftone dot with a low line ruling can be detected. In the peak detection 3, since the matrix is small, a halftone dot with a high screen ruling can be detected. The third halftone dot determination unit 4 finally determines whether or not the area is a halftone area. Details of each block are described below.

(Peak Detector 2) When the density of the target pixel is p and the densities of the surrounding pixels are n1 to n24, the peak and valley peak detection conditions are n1 <n7 <p, n3 <n8 <p, and n5 < n9 <p and n14 <n13 <p and n24 <n18 <p and n22 <n17 <p and n20 <n16 <p and n11 <n12 <p and 2 × p− (n1 + n24)> const. And 2 × p− (n3 + n22)> const. And 2 × p− (n5 + n20)> const. And 2 × p− (n11 + n14)> const. Then, the pixel of interest is the peak of the mountain. n1>n7> p and n3>n8> p and n5>n9> p and n14>n13> p and n24>n18> p and n22>n17> p and n20>n16> p and n11>n12> p and (n1 + n8 ) -2 × p> const. And (n2 + n7) -2 × p> const. And (n3 + n6) -2 × p> const. And (n4 + n5) -2 × p> const. Then, the pixel of interest is the valley peak. Otherwise, the pixel of interest is not a peak.

(Peak Detector 3) When the density of the target pixel is p and the densities of the peripheral pixels are m1 to m8, the peak and valley peak detection conditions are p> m1 and p> m2 and p> m3 and p>. m4 and p> m5 and p> m6 and p> m7 and p> m8 and 2 × p− (n1 + n8)> const. And 2 × p− (n2 + n7)> const. And 2 × p− (n3 + n6)> const. And 2 × p− (n4 + n5)> const. Then, the pixel of interest is the peak of the mountain. p <m1 and p <m2 and p <m3 and p <m4 and p <m5 and p <m6 and p <m7 and p <m8 and (n1 + n8) -2 × p> const. And (n2 + n7) -2 × p> const. And (n3 + n6) -2 × p> const. And (n4 + n5) -2 × p> const. Then, the pixel of interest is the valley peak. Otherwise, the pixel of interest is not a peak.

FIG. 4 is a diagram showing an example of scanning a block. In the halftone dot determining unit 4 (see FIG. 7), when at least one peak of a mountain or a valley exists in a block of a fixed size,
The block is a block including a peak (hereinafter, a peak block). Here, the block size is 4 × 4 pixels. ABCD is a set of one block, and EFG
H and IJKL are similarly considered. The number of peak blocks in each block set is n or more (for example, 3 or more)
If so, the block of interest is determined to be a dot area. For the input image, the peak detector 1 and the peak detector 2 simultaneously detect the halftone peak. In the halftone dot determination unit, the halftone dot area is determined in the same manner as in the past, irrespective of which is detected,
The halftone area 5 is determined.

FIG. 9 is an example of a block diagram of the halftone dot area detection when the peak detection processing is selected. The peak detection processing selection unit 32 is a part that switches the processing of peak detection according to the input resolution. When the resolution of the input image is 600 dpi, the setting is made so that the peak of the halftone dot is detected using both the peak detection unit 33 and the peak detection unit 36. When the input image is 400 dpi, the peak of the halftone dot is detected using only the peak detection unit 33. Others are the same as the above-mentioned method. The resolution of the input image is determined by the CPU when reading the image.
(Central processing unit) or notify the user through the operation panel.

[0028]

According to the first and second aspects of the present invention, when the resolution of an input image is high, it is difficult to separate a halftone dot area having a low screen ruling. Can be eliminated.

According to the third and fourth aspects of the present invention,
Even in a document including the number of lines around the number of lines that can be detected in the halftone dot region, erroneous separation hardly occurs, and the image quality can be prevented from lowering.

According to the fifth and sixth aspects of the present invention,
Even if the resolution is increased, the halftone dot area can be detected with high accuracy, and the processing accuracy can be improved by simply changing the matrix size using small characters as halftone dots. Claim 7
According to the present invention, even if the resolution of the input image is changed, erroneous separation of halftone dots does not occur, and the detection accuracy of halftone dot regions is improved.

[Brief description of the drawings]

FIG. 1 is an example of a double configuration of the overall configuration of a facsimile terminal.

FIG. 2 is an example of a block diagram of a dot area detection process.

FIG. 3 is an example of pixels in a matrix.

FIG. 4 is an example of block scanning.

FIG. 5 is an example of a block diagram of a halftone dot region detection process when the number of lines is also measured.

FIG. 6 is an example of a matrix pixel

FIG. 7 is an example of a block diagram of halftone dot region detection when a plurality of matrices are used at the same time.

FIG. 8 is an example of pixels in a matrix.

FIG. 9 is an example of a block diagram of halftone dot area detection in the case of peak detection processing selection.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input image data 2, 3 Matrix peak detection 4 Halftone dot determination part 5 Halftone area 11 Reading scanning part 12 Reading processing part 13 Recording scanning part 14 Recording processing part 15 Line memory 16 Line memory control part 17 Information compression part / decompression part Reference Signs List 18 modem 19 communication control unit 20 panel unit 21 mechanism control unit 22 RAM 23 MPU 24 network control unit 25 data bus 31 input image data 32 peak detection processing selection unit 33, 36 matrix peak detection 34 halftone dot determination unit 35 halftone area 37 Input image resolution

Claims (7)

[Claims] A reading means for reading a document; an image processing means for performing predetermined image processing on an image read by the reading means; an image output means for outputting image data obtained by the image processing means In the image processing device, the image processing means can perform halftone dot separation by performing peak detection with respect to a halftone dot region in an input image using an N × N matrix, An image processing apparatus, comprising: changing means for changing a dot area separation condition according to a resolution of an input image. 2. The image processing apparatus according to claim 1, wherein said changing means changes a matrix size. 3. The image processing apparatus according to claim 1, further comprising a ruling side changing means for changing a ruling side having a lower detectable halftone frequency. 4. The image processing apparatus according to claim 3, wherein the number-of-lines changing unit includes means for associating the number of lines with a matrix size and setting a matrix size for peak detection processing. 5. The image processing apparatus according to claim 1, wherein halftone dots are detected by simultaneously using a plurality of detection matrices. 6. A plurality of peak detection matrices having different matrix sizes.
The image processing apparatus according to any one of the preceding claims. 7. The image processing apparatus according to claim 5, wherein the number of matrices to be used is changed according to the resolution of the input image.