JP2022026577A - Image processing device, image processing method, and program - Google Patents

Abstract

To minimize the smoothing of an edge portion such as characters and further improve the processing efficiency by attenuating a frequency component peculiar to halftone dots in a pinpoint manner.SOLUTION: An image processing device 1 includes: an image acquisition unit 15 that acquires an input image D1 including halftone dots; a first image generation unit 20 that generates a first image D5 representing a spatial frequency component of the input image D1 on the basis of the input image D1; a second image generation unit 30 that generates a second image D6 obtained by rotating the first image D1 by a predetermined angle; a mask data generation unit 40 that compares the first image D5 and the second image D6, and generates mask data D7 corresponding to the spatial frequency component of the input image D1; and an output image generation unit 50 that performs mask processing on the input image D1 on the basis of the mask data D7 and generates an output image D9.SELECTED DRAWING: Figure 4

Description

The present invention relates to an image processing apparatus, an image processing method and a program, and more particularly to a technique for generating an output image in which the influence of halftone dots is reduced from an input image including halftone dots.

An image processing device such as an MFP (Multifunction Peripherals) is provided with an image reading unit on the upper part of the device body. The image reading unit optically reads the image of the original set by the user to generate image data. When a printed document is read by this image reading unit, the frequency due to the number of halftone dots used in the print halftoning and the frequency in the digital processing of the image reading unit interfere with each other, and moire occurs in the scanned image. Various methods have been conventionally proposed to reduce this moiré.

For example, Patent Document 1 proposes a method of accurately detecting only a character edge portion and performing optimum processing on a halftone dot portion and a character edge portion based on the detection result. However, in this conventional technique, since the halftone dots are not discriminated, it is not possible to perform appropriate processing for the halftone dot region including characters and the gradation to which the halftone dots are combined. Further, when smoothing the halftone dots, it is necessary to perform smoothing processing according to the number of screen lines and the screen angle, but since the halftone dots are not discriminated, it is suitable for the number of screen lines and the screen angle. The smoothing process cannot be performed.

On the other hand, the following Patent Documents 2 to 4 propose a method of appropriately processing both the halftone dot area and the character area.

For example, in Patent Document 2, each of a plurality of frequency bands obtained by wavelet transforming an image is divided into coefficients of a plurality of directional components to discriminate between a character region and a halftone dot region, and according to the discrimination result. A method of correcting each coefficient has been proposed.

Further, in Patent Document 3, the halftone dot region is detected based on the density change of the central pixel and its peripheral pixels, and the spatial frequency is determined by performing orthogonal transformation on the image, and the optimum smoothing filter for the halftone dot region is obtained. A method of selecting is proposed.

Further, in Patent Document 4, the halftone dot region and the character region are determined based on the brightness distribution of the input image, and the halftone dot region and the character region are determined based on the frequency distribution obtained by performing orthogonal transformation on the input image. It is proposed to make a determination, select a filter based on the two determination results, and perform filtering on the frequency data after orthogonal transformation.

Japanese Patent Application Laid-Open No. 2003-101774 Japanese Unexamined Patent Publication No. 2002-247368 Japanese Unexamined Patent Publication No. 2001-11136 Japanese Unexamined Patent Publication No. 2000-354162

However, in Patent Document 2, since it is determined from the edge amount in a predetermined frequency band and its continuity that it is a character region, the portion where the halftone dots overlap is recognized as a continuous edge, and it is erroneously determined as a character region. This may cause deterioration of image quality. Further, in order to accurately determine from a low screen line number to a high screen line number, it is necessary to perform an operation for each of a plurality of frequency bands and a plurality of directional components, which increases the amount of operation. There is a problem that the processing speed is low.

Further, in Patent Document 3, there is a problem that the characters in the halftone dot region are also smoothed. Further, in Patent Document 3, it is necessary to perform a process of detecting a halftone dot region and a process of determining a spatial frequency in parallel, and there is a problem that the circuit scale increases when it is realized by hardware.

Further, in Patent Document 4, it is necessary to perform the area determination process based on the input image and the area determination based on the frequency distribution in parallel, and there is a problem that the circuit scale increases when it is realized by hardware. Further, since the region is separated based on the frequency distribution, there is a problem that the amount of calculation increases and the processing speed is low. Further, in Patent Document 4, since filters corresponding to the respective frequency components of the halftone dot region and the character region are selected and filtered, appropriate processing can be performed on the region where the halftone dots and the characters coexist. There is also the problem of not being able to do it.

In the method conventionally proposed as described above, it is necessary to individually implement a means for discriminating between the halftone dot area and the character area, a means for detecting the frequency of the halftone dot area, and a means for performing smoothing processing. For example, when it is realized by a hardware circuit, the circuit configuration becomes complicated, which leads to an increase in the circuit scale. Further, with the improvement of the processing capacity of the hardware processor in recent years, even when the above three means are realized by software, the processing speed is lowered due to the parallel processing, and the processing efficiency is improved. There is a problem that it is difficult.

Therefore, the present invention has been made to solve the above problem by pinpointly attenuating the frequency component peculiar to the halftone dot without discriminating between the character area and the halftone dot area as in the conventional case. It is an object of the present invention to provide an image processing device, an image processing method, and a program capable of minimizing the smoothing of edge portions such as characters and further improving the processing efficiency. ..

In order to achieve the above object, the invention according to claim 1 is an image processing apparatus, which is an image acquisition means for acquiring an input image including mesh dots, and a spatial frequency component of the input image based on the input image. A comparison between the first image generation means for generating the represented first image, the second image generation means for generating the second image obtained by rotating the first image by a predetermined angle, and the first image and the second image. A mask data generation means that generates mask data corresponding to the spatial frequency component of the input image, and an output image generation means that performs mask processing on the input image based on the mask data and generates an output image. It is a configuration characterized by being provided.

The invention according to claim 2 is characterized in that, in the image processing apparatus of claim 1, the second image generation means generates the second image by rotating the first image 90 degrees clockwise. It is a configuration.

According to a third aspect of the present invention, in the image processing apparatus of the first aspect, the second image generation means generates the second image by rotating the first image counterclockwise by 90 degrees. It is a characteristic configuration.

The invention according to claim 4 is the image processing apparatus according to any one of claims 1 to 3, wherein when the first image generation means generates the first image, the first image is based on a predetermined threshold value. The configuration is characterized by performing binarization processing.

The invention according to claim 5 is the image processing apparatus according to claim 4, wherein the first image generation means has a predetermined range centered on the origin of the frequency coordinate axis in the first image when performing the binarization process. The configuration is characterized in that the region is replaced with a predetermined value.

The invention according to claim 6 is the image processing apparatus according to any one of claims 1 to 5, wherein when the mask data generation means generates the mask data, the frequency coordinate axes in the first image or the second image are used. The configuration is characterized in that a predetermined value is set in a predetermined range region centered on the origin.

The invention according to claim 7 is the image processing apparatus according to any one of claims 1 to 6, wherein the mask data generation means obtains the mask data by performing a logical operation between the first image and the second image. It is a configuration characterized by being generated.

The invention according to claim 8 is the image processing apparatus according to any one of claims 1 to 7, wherein the mask data generation means opens the data obtained by comparing the first image and the second image. The configuration is characterized in that the mask data is generated by performing a process or a closing process.

The invention according to claim 9 is the image processing apparatus according to any one of claims 1 to 8, wherein the mask data generation means expands with respect to the data obtained by comparing the first image and the second image. The configuration is characterized in that the mask data is generated by performing the processing.

The invention according to claim 10 is the image processing apparatus according to any one of claims 1 to 9, wherein the output image generation means generates the output image by performing a logical operation between the mask data and the input image. It is a configuration characterized by that.

The invention according to claim 11 is the image processing apparatus according to any one of claims 1 to 10, wherein the first image generation means performs a Fourier transform on the input image and converts the input image into a power spectrum image. It is a configuration characterized by generating the first image by the above.

The invention according to claim 12 is based on the data obtained by performing a logical calculation between the mask data and the data after Fourier transform of the input image in the image processing apparatus according to claim 11. The configuration is characterized in that the output image is generated by performing an inverse Fourier transform.

The invention according to claim 13 further includes an image reading means for reading an image of a document in the image processing apparatus according to any one of claims 1 to 12, wherein the image reading means reads the document. The configuration is characterized in that the image generated by the above is acquired as the input image.

The invention according to claim 14 is the image processing apparatus according to claim 13, wherein the first image generation means, the second image generation means, and the like, depending on the reading mode when the image reading means reads a document. The configuration is further provided with a mask data generation means and a control means for switching whether or not to operate the output image generation means.

The invention according to claim 15 is the image processing apparatus according to claim 14, wherein when the reading mode is a print document reading mode or a copy document reading mode, the first image generation means and the second image. The configuration is characterized in that the generation means, the mask data generation means, and the output image generation means are operated.

The invention according to claim 16 is the image processing apparatus according to any one of claims 1 to 15, wherein the image acquisition means divides the input image into a plurality of block images having the same vertical and horizontal sizes. The first image generation means generates the first image based on each of the plurality of block images, and the output image generation means produces a plurality of output block images generated from each of the plurality of block images. The configuration is characterized in that the output image is generated by synthesizing.

The invention according to claim 17 is the image processing apparatus according to any one of claims 1 to 15, wherein when the vertical and horizontal sizes of the input image are different, the image acquisition means is vertical with respect to the input image. By performing the magnification conversion process for either the direction or the horizontal direction, a processing target image having the same vertical and horizontal sizes is generated, and the first image generation means is the first image generation means based on the processing target image. 1 Image is generated, and the output image generation means generates the output image by performing a process that is an inverse conversion of the magnification conversion process on the output target image generated from the process target image. It is a characteristic configuration.

The invention according to claim 18 is an image processing method, in which a first step of acquiring an input image including mesh dots and a first image representing a spatial frequency component of the input image based on the input image are generated. The second step is to generate a second image obtained by rotating the first image by a predetermined angle, and the first image and the second image are compared with each other to correspond to the spatial frequency component of the input image. The configuration is characterized by having a fourth step of generating mask data to be generated, and a fifth step of performing mask processing on the first image based on the mask data and generating an output image.

The invention according to claim 19 is a program, in which a first step of acquiring an input image including net dots and a first image representing a spatial frequency component of the input image based on the input image are displayed on a computer. The second step to generate, the third step to generate the second image obtained by rotating the first image by a predetermined angle, the first image and the second image are compared, and the spatial frequency component of the input image is used. The configuration is characterized in that a fourth step of generating corresponding mask data and a fifth step of performing mask processing on the first image based on the mask data and generating an output image are executed.

According to the present invention, it is possible to pinpoint the frequency component peculiar to the halftone dot without discriminating between the character area and the halftone dot area as in the conventional case, and the edge portion of the character or the like is smoothed. It can be minimized. Further, according to the present invention, the processing efficiency can be improved as compared with the conventional case.

It is a figure which shows one configuration example of an image processing apparatus. It is a block diagram which shows one configuration example of the control mechanism of an image processing apparatus. It is a figure which shows an example of the hardware composition of an image processing part. It is a block diagram which illustrates the functional structure in an image processing part. It is a figure which shows an example of an input image. It is a figure which shows the concept of a power spectrum image. It is a figure which shows the power spectrum image generated from the input image of FIG. It is a figure which shows the 1st image generated from a power spectrum image. It is a figure which shows the 2nd image generated from the 1st image. It is a figure which shows an example of the mask data. It is a figure which shows the example which performed the mask processing on the power spectrum image of FIG. It is a figure which shows an example of an output image. It is a flowchart which shows an example of the image processing sequence performed in the image processing unit. It is a figure which shows the example of the detailed processing procedure of the binarization process. It is a figure which shows the example of the detailed processing procedure of the mask data generation processing. It is a figure which shows the example of the detailed processing procedure of a mask processing. It is a flowchart which shows an example of the processing procedure performed by a control unit. It is a block diagram which illustrates the functional structure of the image processing part in 2nd Embodiment. It is a figure which shows the example of the division method by a block division part. It is a block diagram which illustrates the functional structure of the image processing part in 3rd Embodiment. It is a figure which shows the example of the magnification conversion by the magnification conversion unit.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the embodiments described below, the elements common to each other are designated by the same reference numerals, and duplicate description thereof will be omitted.

(First Embodiment)
FIG. 1 is a diagram showing a configuration example of an image processing device 1 according to an embodiment of the present invention. The image processing device 1 is configured as an MFP having a scan function, a copy function, and a print function, and executes a scan job, a copy job, a print job, and the like.

The image processing apparatus 1 includes a scanner unit 2 on the upper part of the apparatus main body. The scanner unit 2 optically reads the image of the original set by the user to generate image data. The scanner unit 2 includes an image reading unit 4 that reads an image of a document, and a document transporting unit 3 that automatically transports the document to a reading position by the image scanning unit 4. For example, when a scan job or a copy job is executed with a document set in the document transport unit 3, the document transfer unit 3 automatically transports the documents set by the user one by one. The image reading unit 4 interlocks with the document transporting operation by the document transporting unit 3, and reads the image of the document when the automatically transported document passes through a predetermined scanning position.

Further, the image processing apparatus 1 includes a printer unit 5 at the lower part of the apparatus main body. The printer unit 5 is provided with a plurality of paper feed cassettes 5a at the lower portion thereof, and feeds sheets such as printing paper contained in the paper feed cassette 5a one by one to form an image on the sheets, and forms an image on the upper portion. The sheet on which the image is printed is ejected to the output tray 5b of the above.

Further, the image processing apparatus 1 is provided with an operation panel 6 that can be operated by the user on the front side of the apparatus main body. The operation panel 6 serves as a user interface when the user uses the image processing device 1. The operation panel 6 displays various operation screens that can be operated by the user, and accepts operations by the user. For example, the operation panel 6 receives a job setting operation and a job execution start instruction by the user. When the operation panel 6 receives the job execution start instruction by the user, the image processing device 1 starts the execution of the job specified by the user.

FIG. 2 is a block diagram showing a configuration example of the control mechanism of the image processing device 1. The image processing device 1 includes an image memory 7, a control unit 8, a communication interface 9, and an image processing unit 10 in addition to the scanner unit 2, the printer unit 5, and the operation panel 6 described above.

The image memory 7 is a memory for storing image data obtained by executing a job. For example, in the case of a scan job or a copy job, the image data generated by the scanner unit 2 is stored in the image memory 7.

The control unit 8 comprehensively controls the execution of jobs in the image processing device 1. For example, the control unit 8 controls the operation screen displayed on the operation panel 6 and sets the job based on the operation by the user. Further, the control unit 8 controls the execution of the job designated by the user by controlling the operations of the scanner unit 2 and the printer unit 5.

For example, when the user is instructed to execute the scan job, the control unit 8 operates the scanner unit 2 to control the reading operation of the document set by the user. Then, the control unit 8 stores the image data output from the scanner unit 2 in the image memory 7. Further, the control unit 8 operates the image processing unit 10 as necessary, and causes the scanner unit 2 to perform image processing on the image data stored in the image memory 7.

The communication interface 9 is an interface for connecting the image processing device 1 to a network such as a LAN (Local Area Network) and communicating with an external device. The image processing device 1 can receive a print job transmitted from an external device via the communication interface 9. Upon receiving the print job via the communication interface 9, the control unit 8 operates the printer unit 5. As a result, the printer unit 5 forms an image on the sheet based on the print job, and ejects the sheet printed on the output tray 5b.

When the image generated by the scanner unit 2 contains halftone dots, the image processing unit 10 performs image processing to reduce the influence of the halftone dots from the image, and is a characteristic image processing function in the present invention. Is to realize. For example, the image processing unit 10 functions when the control unit 8 determines that it is necessary to smooth the net dots when a scan job or a copy job is performed in the image processing device 1, and the image processing unit 2 displays an image. The image data stored in the memory 7 is acquired as an input image, and image processing is performed on the input image to reduce the influence of the mesh dots. In the following embodiment, an example in which the image processing unit 10 performs image processing by software processing will be described.

FIG. 3 is a diagram showing an example of the hardware configuration of the image processing unit 10. The image processing unit 10 includes, for example, a CPU 11, a ROM 12, and a RAM 13. The CPU 11 is a hardware processor that reads out and executes a program 14 stored in advance in the ROM 12. The ROM 12 is a non-volatile storage device that stores the program 14. The program 14 stored in the ROM 12 is a program that causes the CPU 11 to perform image processing for generating an output image in which the influence of the halftone dots is reduced from the input image including the halftone dots. The RAM 13 is a volatile storage device for storing temporary data, images, and the like generated by the CPU 11 executing the program 14.

The image processing unit 10 having the hardware configuration as described above acquires an input image including halftone dots by the CPU 11 executing the program 14, and pinpoints the frequency components peculiar to the halftone dots included in the input image. It is possible to minimize the smoothing of edge parts such as characters by performing the damping process. Hereinafter, the image processing performed by the image processing unit 10 will be described in detail.

FIG. 4 is a block diagram illustrating a functional configuration in the image processing unit 10. By executing the program 14, the CPU 11 of the image processing unit 10 functions as an image acquisition unit 15, a first image generation unit 20, a second image generation unit 30, a mask data generation unit 40, and an output image generation unit 50. do. By making these functions in order, the image processing unit 10 generates an output image in which the influence of halftone dots is reduced from the input image.

The image acquisition unit 15 is a processing unit that acquires an input image D1 including halftone dots. That is, the image acquisition unit 15 acquires the image data generated by the scanner unit 2 and stored in the image memory 7 as the input image D1. FIG. 5 is a diagram showing an example of the input image D1. The input image D1 shown in FIG. 5 is an image including halftone dots. Hereinafter, an example of processing for the input image D1 shown in FIG. 5 will be described.

When the image acquisition unit 15 acquires the input image D1 from the image memory 7, the image acquisition unit 15 outputs the processing target image D2 based on the input image D1 to the first image generation unit 20. At this time, the image acquisition unit 15 may output the input image D1 as it is to the first image generation unit 20 as the processing target image D2. However, when the input image D1 is image data represented by the three primary colors of R (red), G (green), and B (blue) for each pixel, the input image D1 is used as it is as the processing target image D2 in the subsequent stage. If the data is output to the first image generation unit 20, the same processing must be performed on the three images of each color component of R, B, and G in the first image generation unit 20 and subsequent units, and the processing efficiency is lowered. ..

Therefore, when the input image D1 is image data represented by the three primary colors of R, G, and B, the image acquisition unit 15 converts the RGB color space of the input image D1 into the YCrCb color space, and Y (brightness) component. It is preferable to generate one shading image represented only by the image as the processing target image D2, and output the one processing target image D2 to the first image generation unit 20. As a result, in the first image generation unit 20 and subsequent units, it is sufficient to perform processing on one image, so that the processing efficiency can be improved.

The first image generation unit 20 generates a first image representing the spatial frequency component of the input image D1 based on the processing target image D2 output from the image acquisition unit 15. The first image generation unit 20 includes a Fourier transform unit 21, an image generation unit 22, and a binarization processing unit 23, which sequentially perform processing on the image D2 to be processed to produce a first image. Generate.

The Fourier transform unit 21 performs a discrete Fourier transform (DFT) on the image D2 to be processed. As a result, the image D2 to be processed is converted into frequency data. The Fourier transform unit 21 temporarily stores the frequency data in the RAM 13 or the like as the data D3 after the Fourier transform. Further, the Fourier transform unit 21 outputs the data D3 after the Fourier transform to the image generation unit 22.

The image generation unit 22 generates a power spectrum image D4 in which the spatial frequency component of the input image D1 is represented by a grayscale image based on the Fourier transform data D3. FIG. 6 is a diagram showing the concept of a power spectrum image. The power spectrum image is an image in which the horizontal axis is the horizontal frequency and the vertical axis is the vertical frequency, and the magnitude of the frequency component in each direction in the input image D1 is represented by the brightness. In this power spectrum image, the central origin represents the DC component of the input image D1, and the farther the distance R from the origin is, the higher the frequency component is represented.

FIG. 7 is a diagram showing a power spectrum image D4 generated from the input image D1 of FIG. In general, effective image elements (characters, figures, paintings, etc.) included in the input image D1 have a lower frequency than the halftone dots, so that the frequency is low near the origin as shown in the power spectrum image D4 shown in FIG. Appears as a bright spot in the area. On the other hand, since the halftone dots included in the input image D1 have higher frequencies than the effective image elements, they appear as bright spots in a high frequency region away from the origin as shown in the power spectrum image D4 shown in FIG.

When the image generation unit 22 generates the power spectrum image D4 based on the Fourier transform data D3, the image generation unit 22 outputs the power spectrum image D4 to the binarization processing unit 23.

The binarization processing unit 23 generates the first image D5 by binarizing the power spectrum image D4 with a predetermined threshold value. For example, the binarization processing unit 23 converts the central region P1 (see FIG. 6) within a predetermined distance R from the origin in the power spectrum image D4 into a value of “0”. That is, the binarization processing unit 23 converts the central region P1 to "0" in order to prevent the mask processing described later from being performed on the frequency domain of the portion where the valid image element exists. I will leave it. Then, the binarization processing unit 23 performs binarization processing based on a predetermined threshold value on the region outside the central region P1. As a result, the binarization processing unit 23 can generate the binary image shown in FIG. 8 as the first image D5 from the power spectrum image D4 shown in FIG. 7. In FIG. 8, the pixel having the gradation value “0” is shown in black, and the pixel having the gradation value “1” is shown in white.

When the binarization processing unit 23 binarizes the power spectrum image D4 to generate the first image D5, the binarization processing unit 23 outputs the first image D5 to each of the second image generation unit 30 and the mask data generation unit 40.

The second image generation unit 30 generates the second image D6 from the first image D5. As described above, when the input image D1 is converted to the power spectrum image D4, the halftone dots included in the input image D1 are bright spots in a region outside the central region P1 in which a valid image element is present as described above. Appears as. The bright spots formed by the halftone dots have the characteristic that they appear at the same position even when the power spectrum image D4 is rotated by 90 degrees. That is, when the original power spectrum image and the power spectrum image rotated by 90 degrees are superposed, if there is a frequency component in which the bright spots overlap in a region outside the central region P1, the overlapped frequency component is present. It means that it is the frequency component of the halftone dot. Therefore, the second image generation unit 30 includes an image rotation unit 31 that rotates the first image D5 to generate the second image D6.

The image rotation unit 31 generates the second image D6 by rotating the first image D5 by a predetermined angle. As described above, the bright spots due to the halftone dots appearing in the power spectrum image appear at the same position even when rotated by 90 degrees. Therefore, the image rotation unit 31 generates the second image D6 by rotating the first image D5 by 90 degrees. The direction in which the first image D5 is rotated may be clockwise or counterclockwise. That is, the image rotation unit 31 generates the second image D6 by rotating the first image D5 clockwise or counterclockwise by 90 degrees. However, the rotation angle at which the image rotation unit 31 rotates the first image D5 does not necessarily have to be 90 degrees. For example, the image rotation unit 31 can generate a similar second image D6 by rotating the first image D5 clockwise or counterclockwise by 270 degrees. Then, the second image generation unit 30 outputs the second image D6 generated by the image rotation unit 31 to the mask data generation unit 40.

FIG. 9 is a diagram showing a second image D6 generated by rotating the first image D5 shown in FIG. 8 counterclockwise by 90 degrees. In this way, the second image D6 is generated by rotating the first image D5. Then, two images, the first image D5 and the second image D6, are output to the mask data generation unit 40.

The mask data generation unit 40 generates mask data corresponding to the frequency component of the halftone dots included in the input image D1 by comparing the first image D5 and the second image D6. The mask data generation unit 40 includes an image comparison unit 41 and a mask adjustment unit 42.

The image comparison unit 41 compares the first image D5 and the second image D6, and generates intermediate data that is the source of the mask data. For example, the image comparison unit 41 generates intermediate data that is the source of mask data by performing a logical registration calculation for each pixel between the first image D5 and the second image D6. In both the first image D5 and the second image D6, the gradation value for each pixel is binarized to "0" or "1". Therefore, by performing the logical product calculation for each pixel, the image comparison unit 41 uses the bright spot (point of gradation value “1”) in the first image D5 and the bright spot (gradation value “1”) in the second image D6. "Point)) and the part that overlaps with are extracted to generate intermediate data. Then, the image comparison unit 41 outputs the intermediate data to the mask adjustment unit 42.

The mask adjusting unit 42 generates the mask data D7 from the intermediate data output from the image comparison unit 41. The mask adjusting unit 42 performs a process of removing noise included in the intermediate data, for example, by performing an opening process on the intermediate data. The opening process is, for example, a process in which an intermediate data is subjected to a shrinkage process (erosion) and then an expansion process (dilation).

Further, the mask adjusting unit 42 performs a process of filling small black spots inside the bright spots corresponding to the halftone dots by, for example, performing a closing process on the intermediate data. The closing process is, for example, a process of performing an expansion process (dilation) and then a contraction process (erosion) on the intermediate data.

The mask adjusting unit 42 may perform both the opening process and the closing process on the intermediate data, or may perform only one of them. Further, the mask adjusting unit 42 may omit the opening process and the closing process when there is little noise in the intermediate data and there are no small black dots inside the bright spots corresponding to the halftone dots.

Then, the mask adjusting unit 42 performs expansion processing (dilation) on the intermediate data, and converts the size of the bright spots corresponding to the halftone dots in the intermediate data into a size larger by a predetermined ratio to generate the mask data D7. ..

FIG. 10 is a diagram showing mask data D7 generated by the mask adjusting unit 42. As shown in FIG. 10, the mask data D7 is generated as data that can fill the frequency component of the halftone dots included in the input image D1. When the mask data generation unit 40 generates the mask data D7 as shown in FIG. 10, the mask data generation unit 40 outputs the mask data D7 to the output image generation unit 50.

The output image generation unit 50 performs mask processing on the input image D1 based on the mask data D7, and generates an output image D9 in which the influence of halftone dots is reduced. The output image generation unit 50 includes a mask processing unit 51 and an inverse Fourier transform unit 52.

The mask processing unit 51 performs mask processing on the input image D1 based on the mask data D7. For example, as shown in FIG. 11, the mask processing unit 51 performs a process of attenuating the frequency component of the halftone dots by painting the portion corresponding to the frequency component of the halftone dots in black based on the mask data D7. Note that FIG. 11 shows an example in which the power spectrum image D4 of FIG. 7 is masked.

For example, the mask processing unit 51 generates data in which the black and white of the mask data D7 is inverted (mask data in which "0" and "1" of the mask data D7 are exchanged). As a result, the mask data D7 shown in FIG. 10 becomes an image in which black and white are inverted. Then, the mask processing unit 51 generates post-mask processing data by performing a logical operation between the mask data in which the black and white are inverted and the input image D1 (more strictly, the frequency data of the input image D1). More specifically, the mask processing unit 51 reads the post-Fourier transform data D3 generated by the Fourier transform unit 21 from the RAM 13, and sets the values of each frequency component in the mask data in which black and white are inverted, and after the Fourier transform. The post-mask data D8 is generated by calculating the logical product with the value of each frequency component in the data. Since black is "0" and white is "1" in the mask data, the frequency component of the halftone dots becomes 0 by the logical product operation, whereas the frequency components other than the halftone dots can be left as they are. .. Therefore, the frequency component of halftone dots can be attenuated pinpointly by this mask processing.

In addition to this, for example, the mask processing unit 51 sets the values of the real part and the imaginary part of the Fourier transform data D3 corresponding to the coordinates where the value is "0" in the mask data in which black and white are inverted to "0". Alternatively, the post-masking data D8 in which the frequency component of the net point is attenuated may be generated by calculating the logical product of the real part and the imaginary part using the coefficient M. good. From the viewpoint of attenuation, the value of the coefficient M is a value satisfying 0 <M <1. Then, the mask processing unit 51 outputs the masked data D8 to the inverse Fourier transform unit 52.

The inverse Fourier transform unit 52 generates an output image D9 by performing an inverse discrete Fourier transform (IDFT) on the masked data D8. FIG. 12 is a diagram showing an example of the output image D9 generated by the inverse Fourier transform unit 52. In the output image D9 shown in FIG. 12, since the frequency component of the halftone dots is attenuated, the image is an image in which the influence of the halftone dots is reduced.

When the color conversion from the RGB color space to the YCrCb color space is performed in the image acquisition unit 15, the output image D9 generated by the inverse Fourier transform unit 52 becomes an image of the Y (brightness) component. Therefore, when it is necessary to output image data represented by the three primary colors of R, G, and B, the output image generation unit 50 is generated by the Y component generated by the inverse Fourier conversion unit 52 and the image acquisition unit 15. Image data expressed in RGB may be generated by reconverting the YCrCb color space into the RGB color space based on the Cr component and the Cb component obtained.

As described above, the image processing apparatus 1 of the present embodiment can pinpoint the frequency component peculiar to the halftone dots without discriminating between the character area and the halftone dot area as in the conventional case. Therefore, the image processing apparatus 1 of the present embodiment can reduce the influence of halftone dots included in the input image D1 while minimizing the smoothing of edge portions such as characters. Further, as shown in FIG. 4, the image processing apparatus 1 of the present embodiment has processing by the first image generation unit 20, processing by the second image generation unit 30, processing by the mask data generation unit 40, and output. Since the output image D9 can be generated by sequentially performing the processing by the image generation unit 50, the parallel processing as in the conventional case is not involved. Therefore, when the image processing unit 10 performs the above-mentioned image processing by software processing, it is possible to suppress a decrease in the processing speed and improve the processing efficiency. Further, even when the above-mentioned image processing in the image processing unit 10 is realized by hardware, it is possible to realize it with a simple circuit configuration because it is not necessary to perform parallel processing, and the circuit scale is increased. There is also an advantage that it can be prevented.

Next, the processing procedure performed by the image processing unit 10 will be described. FIG. 13 is a flowchart showing an example of an image processing sequence performed by the image processing unit 10. When the image processing unit 10 starts this processing, it acquires an input image D1 including halftone dots (step S10). At this time, the image processing unit 10 performs a process of converting the RGB color space of the input image D1 into the YCrCb color space, if necessary.

Next, the image processing unit 10 performs a Fourier transform process on the input image D1 (step S11), and saves the data D3 after the Fourier transform in the RAM 13 (step S12). Further, the image processing unit 10 generates a power spectrum image D4 of the input image D1 based on the Fourier transform data D3 (step S13), and performs binarization processing on the power spectrum image D4 (step S14).

FIG. 14 is a diagram showing two examples of the detailed processing procedure of the binarization process (step S14). First, the binarization process shown in FIG. 14A will be described. When the binarization process of FIG. 14A is started, the image processing unit 10 converts the region P1 within a predetermined distance R from the origin of the power spectrum image D4 to “0” (step S20). This prevents the low frequency components constituting the effective image element in the input image D1 from being attenuated. Subsequently, the image processing unit 10 acquires a threshold value as a reference when performing binarization (step S21). Then, the image processing unit 10 binarizes the region outside the range of the predetermined distance R from the origin in the power spectrum image D4 with the threshold value acquired in step S21 (step S22). As a result, the first image D5 obtained by binarizing the power spectrum image D4 is generated.

Next, the binarization process shown in FIG. 14B will be described. When the binarization process of FIG. 14B is started, the image processing unit 10 acquires the threshold value T1 to be applied within a predetermined distance R from the origin of the power spectrum image D4 (step S24), and also obtains the power spectrum image. Acquire the threshold value T2 to be applied outside the range of the predetermined distance R from the origin of D4 (step S25). That is, in this example, the threshold value T1 applied to the low frequency component constituting the effective image element in the input image D1 and the threshold value T2 applied to the high frequency component including halftone dots are set as different values. To be acquired. Here, the threshold value T1 is a higher value than the threshold value T2. Then, the image processing unit 10 binarizes the region P1 within a predetermined distance R from the origin of the power spectrum image D4 based on the threshold value T1 (step S26). Further, the image processing unit 10 binarizes a region outside the range of a predetermined distance R from the origin of the power spectrum image D4 based on the threshold value T2 (step S27). As a result, the first image D5 obtained by binarizing the power spectrum image D4 is generated.

In step S14, one of the two binarization processes described above is adopted, and the first image D5 in which the power spectrum image D4 is binarized is generated. Returning to the flowchart of FIG. 13, when the image processing unit 10 generates the first image D5, the image processing unit 10 saves the first image D5 in the RAM 13 (step S15).

Next, the image processing unit 10 rotates the first image D5 90 degrees clockwise or counterclockwise to generate a second image D6 (step S16), and stores the second image D6 in the RAM 13 (step S16). Step S17).

Subsequently, the image processing unit 10 executes a mask data generation process (step S18). FIG. 15 is a diagram showing two examples of detailed processing procedures of the mask data generation processing (step S18).

First, the mask data generation process shown in FIG. 15A will be described. When the image processing unit 10 starts the mask data generation process of FIG. 15A, the first image D5 and the second image D6 are read out (step S30), and the first image D5 and the second image D6 are displayed for each pixel. (Step S31). As a result, the frequency components of the halftone dots forming the bright spots at the same positions in the first image D5 and the second image D6 are extracted. Then, the image processing unit 10 executes an expansion process for expanding the bright spot on the image obtained by the logical product operation (step S32). That is, the image processing unit 10 generates mask data D7 having a size slightly larger than the size of the bright spot corresponding to the frequency component of the halftone dot. As a result, the frequency component of the halftone dots can be attenuated without any shortage in the mask processing in the subsequent stage. Then, the image processing unit 10 stores the mask data D7 generated by performing the expansion processing in the RAM 13 (step S33).

Next, the mask data generation process shown in FIG. 15B will be described. When the image processing unit 10 starts the mask data generation process of FIG. 15B, the first image D5 and the second image D6 are read out (step S35), and the first image D5 and the second image D6 are displayed for each pixel. (Step S36), the intermediate data is generated by performing the logical product operation of. Then, the image processing unit 10 executes an opening process or a closing process on the intermediate data (step S37). The image processing unit 10 may perform both the opening process and the closing process on the intermediate data. After that, the image processing unit 10 executes expansion processing on the intermediate data to which the opening processing or closing processing has been performed, and generates mask data D7 (step S38). Then, the image processing unit 10 stores the mask data D7 generated by performing the expansion processing in the RAM 13 (step S39).

In step S18, one of the two mask data generation processes described above is adopted, and the mask data D7 is generated. Returning to the flowchart of FIG. 13, when the image processing unit 10 generates the mask data D7, the image processing unit 10 executes the mask processing (step S19). FIG. 16 is a diagram showing two examples of the detailed processing procedure of the mask processing (step S19).

First, the mask processing shown in FIG. 16A will be described. When the mask processing of FIG. 16A is started, the image processing unit 10 reads out the Fourier transform data D3 and the mask data D7 stored in the RAM 13 (step S40). At this time, the image processing unit 10 generates mask data in which the black and white of the mask data D7 is inverted. Then, the image processing unit 10 performs a logical product calculation of the real number part of the data D3 after the Fourier transform and the mask data in which black and white are inverted for each pixel (every frequency) (step S41), and the data D3 after the Fourier transform. The logical product calculation of the imaginary part and the mask data D7 in which black and white is inverted is also performed for each pixel (for each frequency) (step S42). Then, the image processing unit 10 stores the real number part and the imaginary number part obtained by the logical operation in the RAM 13 as data to be subjected to the inverse Fourier transform (step S43). The mask process shown in FIG. 16A requires a logical operation of the mask data for each pixel (every frequency), which requires a relatively long processing time.

Next, the mask processing shown in FIG. 16B will be described. When the mask processing of FIG. 16B is started, the image processing unit 10 reads out the Fourier transform data D3 and the mask data D7 stored in the RAM 13 (step S45). Then, the image processing unit 10 converts the value of the real part of the Fourier transform data D3 corresponding to the coordinates of 0 in the black and white inverted mask data to 0 (step S46), and in the black and white inverted mask data. The value of the imaginary part of the Fourier transform data D3 corresponding to the coordinates of 0 is converted to 0 (step S47). That is, in the mask processing of FIG. 16B, the coordinates having a value of 0 in the mask data in which black and white are inverted are specified, and the values of the real part and the imaginary part of the Fourier transform data D3 corresponding to the coordinates are specified. Is simply replaced with 0, so that the processing can be performed in a shorter time than the mask processing of FIG. 16A, and the processing efficiency can be improved. Then, the image processing unit 10 stores the real number part and the imaginary number part obtained in steps S46 and S47 in the RAM 13 as data to be subjected to the inverse Fourier transform (step S48).

In the mask processing of FIG. 16B, an example in which the real part and the imaginary part of the Fourier transform data D3 are replaced with 0 is given in steps S46 and S47, but the present invention is not limited to this. For example, in step S46, the logical product operation of the value of the real part of the Fourier transform data D3 corresponding to the coordinate of 0 and the coefficient M may be performed in the mask data in which black and white are inverted, or in step S47. Then, in the mask data in which black and white are inverted, the logical product operation of the value of the real part of the Fourier transform data D3 corresponding to the coordinate of 0 and the coefficient M may be performed.

Returning to the flowchart of FIG. 13 again, the image processing unit 10 then reads the data to be inverted Fourier transformed from the RAM 13 and executes the inverse Fourier transform process on the data (step S20). As a result, the output image D9 in which the influence of halftone dots is reduced is generated. Then, the image processing unit 10 saves the output image D9 in the image memory 7 (step S21). As a result, the control unit 8 can acquire the output image D9 from the image memory 7. When the output image D9 is generated, the image processing unit 10 performs an inverse Fourier transform process and then, if necessary, converts the color space of the output image D9 from the YCrCb color space to the RGB color space.

After that, the image processing unit 10 determines whether or not to end the image processing (step S22). For example, when the scanner unit 2 continuously scans a plurality of original images, it is necessary to perform the same processing on the second and subsequent images. In this case, the image processing unit 10 determines that the image processing is to be continued (NO in step S22), returns to step S10, and repeats the processing of steps S10 to S21 described above for the second and subsequent images. On the other hand, when the image processing for the images of all the originals scanned by the scanner unit 2 is completed, the image processing unit 10 determines that the image processing is completed (YES in step S22), and all the processing is completed. As described above, in the image processing of the present embodiment, all the processing can be performed sequentially.

Next, the processing procedure of the control unit 8 when the scan job is executed in the image processing device 1 will be described. FIG. 17 is a flowchart showing an example of a processing procedure performed by the control unit 8. When the control unit 8 starts this process, the control unit 8 sets the document scanning mode based on the job setting operation by the user (step S50). Then, the control unit 8 determines whether or not the set reading mode is the print document reading mode (step S51). The print document reading mode is a mode for scanning an image of a document printed by the image processing apparatus 1. As a result of the above determination, if the print document reading mode is not set (NO in step S51), the control unit 8 next determines whether or not the copy document reading mode is set (step S52). The copy document reading mode is a mode for scanning the image of the document copied and output by the image processing apparatus 1.

When the scanning mode is the print document scanning mode or the copy document scanning mode (YES in step S51 or YES in step S52), the control unit 8 executes the scan job based on the user's instruction to start executing the scan job. When started (step S53), the image processing unit 10 is made to function, and the halftone dot smoothing process by the image processing unit 10 is executed (step S54). Therefore, when the execution of the scan job is started, the halftone dot smoothing process (process of FIGS. 13 to 16) described above is performed in the image processing unit 10. The control unit 8 continues the image processing by the image processing unit 10 until the scan job is completed (NO in step S55). On the other hand, when the scan job is completed (YES in step S55), the control unit 8 ends the image processing by the image processing unit 10 and ends the job execution.

On the other hand, when the scanning mode is neither the print document scanning mode nor the copy document scanning mode (NO in step S51 and NO in step S52), the control unit 8 is based on the user's instruction to start executing the scan job. And start the execution of the scan job (step S56). In this case, the control unit 8 does not function the image processing unit 10. That is, when the image of the original document read by the scanner unit 2 is an image that does not include halftone dots, the control unit 8 does not operate the image processing unit 10 during the execution of the scan job, and the above-mentioned image processing is not performed. To control. When the execution of the scan job is completed (YES in step S57), the processing by the control unit 8 is completed.

As described above, in the image processing device 1 of the present embodiment, the control unit 8 determines whether or not it is necessary to reduce the influence of the halftone dots according to the reading mode specified by the user, and the influence of the halftone dots is determined. When it is determined that the reduction is necessary, the image processing unit 10 is made to function during the execution of the scan job, and the image data output from the scanner unit 2 is used as the input image D1 to perform the halftone dot smoothing process described above. To control.

As described above, when the image processing apparatus 1 of the present embodiment acquires the input image D1 including halftone dots, it generates a first image D5 representing the spatial frequency component of the input image D1 based on the input image D1. Then, the image processing device 1 generates a second image D6 obtained by rotating the first image D5 by a predetermined angle, and compares the first image D5 with the second image D6 to correspond to the spatial frequency component of the input image D1. The mask data D7 to be processed is generated. Further, the image processing apparatus 1 is configured to perform mask processing on the input image D1 based on the mask data D7 and generate an output image D9. The image processing device 1 having such a configuration can perform pinpoint mask processing on the frequency components of halftone dots included in the input image D1, and the edge portions such as characters are smoothed. It is possible to obtain an output image D9 in which the influence of halftone dots is reduced more efficiently than in the past while minimizing this.

(Second Embodiment)
Next, a second embodiment of the present invention will be described. In the first embodiment, the first image D5 is rotated 90 degrees to generate the second image D6, and the mask data is generated by calculating the logical product of the first image D5 and the second image D6 for each pixel. The process to be performed was explained. When such processing is appropriately performed, it is premised that the vertical and horizontal sizes of the first image D5 and the vertical and horizontal sizes of the second image D6 are equal to each other. That is, the process described in the first embodiment can be suitably used when the input image D1 is a square image. However, the actual situation is that the document read by the scanner unit 2 is not a square but a rectangle in many cases. In view of such circumstances, the present embodiment describes a mode in which the scanner unit 2 can be applied to an image generated by scanning a rectangular original. The overall configuration of the image processing apparatus 1 in this embodiment is the same as that described in the first embodiment.

FIG. 18 is a block diagram illustrating a functional configuration of the image processing unit 10 in the present embodiment. By executing the program 14, the CPU 11 of the image processing unit 10 executes the image acquisition unit 15, the first image generation unit 20, the second image generation unit 30, the mask data generation unit 40, and the mask data generation unit 40, as in the first embodiment. , Functions as an output image generation unit 50. Of these, the image acquisition unit 15 and the output image generation unit 50 are different from the first embodiment. That is, the image acquisition unit 15 includes the block division unit 16, and the output image generation unit 50 includes the composition processing unit 53, which is different from the first embodiment.

When the block division unit 16 of the image acquisition unit 15 acquires the input image D1 from the image memory 7, the block division unit 16 divides the input image D1 into a plurality of block images. For example, when the input image D1 is rectangular and the vertical and horizontal sizes are different, the block division unit 16 divides the input image D1 into a plurality of block images having the same vertical and horizontal sizes. Therefore, the plurality of block images generated by the block dividing unit 16 are all square. When a plurality of block images are generated by the block division unit 16, the image acquisition unit 15 sequentially outputs each of the plurality of block images as a processing target image D2 to the first image generation unit 20.

FIG. 19 is a diagram showing an example of a division method by the block division unit 16. First, in the example shown in FIG. 19A, a plurality of block images are generated by dividing each of the horizontal direction (X direction) and the vertical direction (Y direction) of the input image D1 by a predetermined reference length. do. In this case, the lengths of the input image D1 in the horizontal direction and the vertical direction may not be an integral multiple of the reference length. In such a case, by making the last block image in the horizontal direction and the vertical direction a block image in a state of overlapping with the previous block image, all the block images can be generated as a square. In the case of the division method shown in FIG. 19A, a large number of block images are generated from the input image D1 by the block division unit 16.

Next, the example shown in FIG. 19B is a division method for generating two block images from the input image D1. The size of the input image D1 shown in FIG. 19B is larger in the vertical direction than in the horizontal direction. In this case, the block dividing unit 16 divides the input image D1 in the vertical direction by the same length as the horizontal length. Also at this time, the vertical length of the input image D1 may not be an integral multiple of the horizontal length. In such a case, the block dividing unit 16 refers to the first block image divided downward at a position having the same length as the lateral length with respect to the upper end of the input image D1 and the lower end of the input image D1. It is divided into a second block image divided upward at a position having the same length as the lateral length. That is, the first block image and the second block image are generated as images in a state of being overlapped with each other. If the size in the vertical direction is smaller than that in the horizontal direction, the input image D1 may be divided in the horizontal direction.

The block division unit 16 may adopt either of the above two division methods. However, in the case of the division method shown in FIG. 19A, the number of block images generated by the block division increases, and the area of each block image becomes small. As described in the first embodiment, when the individual block images are subjected to Fourier transform as the processing target image D2 to generate a power spectrum image, when the size of the block image becomes small, the frequency component of the halftone dots is appropriately adjusted. It may not be possible to detect. Further, as the number of block images increases, the number of times of processing from the Fourier transform to the inverse Fourier transform on the block images increases, so that the processing efficiency decreases. Therefore, it is preferable to adopt the division method shown in FIG. 19 (b) rather than the division method shown in FIG. 19 (a). In the case of the division method shown in FIG. 19B, the number of block images generated by block division can be minimized, and the area of each block image can be maximized. There is an advantage that the possibility that the frequency component cannot be detected properly can be reduced, and the decrease in processing efficiency can be suppressed.

When a plurality of block images generated by the block division unit 16 are sequentially output to the first image generation unit 20 as the image to be processed D2, the image processing described in the first embodiment is performed for each of the plurality of block images. Is done. The output image D9 generated by the inverse Fourier transform unit 52 of the output image generation unit 50 becomes an output block image corresponding to each block image. Therefore, in the present embodiment, the output image generation unit 50 is provided with a composition processing unit 53, and a plurality of output block images sequentially output from the inverse Fourier transform unit 52 are combined. That is, the composition processing unit 53 generates the output image D10 corresponding to the input image D1 by synthesizing the output block images corresponding to the individual block images based on the division method applied in the block division unit 16. ..

As described above, when the input image D1 is not a square, the image processing apparatus 1 of the present embodiment divides the input image D1 to generate a square block image, and the block image is used as a processing target from the Fourier transform unit 21. Image processing up to the inverse Fourier transform unit 52 is performed. Therefore, it is possible to appropriately perform processing such as generating mask data by calculating the logical product of the first image D5 and the second image D6 for each pixel, and the frequency of the halftone dots included in the input image D1. The components can be appropriately attenuated.

The configurations and operations other than those described above in the present embodiment are the same as those described in the first embodiment.

(Third Embodiment)
Next, a third embodiment of the present invention will be described. In the second embodiment, when the input image D1 is not a square, an example is described in which the input image D1 is divided into square block images so that appropriate image processing can be performed. On the other hand, in the present embodiment, when the input image D1 is not a square, a mode for generating a square image by enlarging or reducing either the vertical direction or the horizontal direction of the input image D1 will be described. The overall configuration of the image processing apparatus 1 in this embodiment is the same as that described in the first embodiment.

FIG. 20 is a block diagram illustrating a functional configuration of the image processing unit 10 in the present embodiment. By executing the program 14, the CPU 11 of the image processing unit 10 executes the image acquisition unit 15, the first image generation unit 20, the second image generation unit 30, the mask data generation unit 40, and the mask data generation unit 40, as in the first embodiment. , Functions as an output image generation unit 50. Of these, the image acquisition unit 15 and the output image generation unit 50 are different from the first embodiment. That is, the image acquisition unit 15 is provided with the magnification conversion unit 17, and the output image generation unit 50 is also provided with the magnification conversion unit 54, which is different from the first embodiment.

When the magnification conversion unit 17 of the image acquisition unit 15 acquires the input image D1 from the image memory 7, it detects the vertical and horizontal sizes of the input image D1 and inputs one size so as to match the other size. A square image is generated by performing enlargement processing or reduction processing in the uniaxial direction with respect to the image D1. That is, in the present embodiment, one square image is generated from one input image D1. Then, when the magnification conversion unit 17 generates a square image, the image acquisition unit 15 outputs the image to the first image generation unit 20 as the image to be processed D2.

FIG. 21 is a diagram showing an example of magnification conversion by the magnification conversion unit 17. The input image D1 shown in FIG. 21 is a rectangular image in which the size in the vertical direction (Y direction) is larger than the size in the horizontal direction (X direction). In this case, the magnification conversion unit 17 performs reduction processing in the Y direction so that the size in the vertical direction matches the size in the horizontal direction, for example. As a result, as shown in FIG. 21, the rectangular input image D1 is converted into a square image, so that the square image can be used as the processing target image D2. Note that FIG. 21 illustrates a case where the reduction process in the Y direction is performed so that the size in the vertical direction matches the size in the horizontal direction, but the method for obtaining a square image is not limited to this. For example, the magnification conversion unit 17 may perform enlargement processing in the X direction so that the size in the horizontal direction matches the size in the vertical direction.

When the square image generated by the magnification conversion unit 17 is output to the first image generation unit 20 as the image to be processed D2, the image processing described in the first embodiment is performed on the square image. The output image D9 generated by the inverse Fourier transform unit 52 of the output image generation unit 50 is an output target image corresponding to a square image. Therefore, in the present embodiment, the output image generation unit 50 is provided with a magnification conversion unit 54 to convert the magnification of the output target image output from the inverse Fourier transform unit 52. That is, the magnification conversion unit 54 performs the magnification conversion processing (reduction or enlargement) opposite to the magnification conversion processing (enlargement or reduction) performed by the magnification conversion unit 17 on the output target image, thereby performing the original input image. The output image D10 corresponding to D1 is generated.

As described above, when the input image D1 is not a square, the image processing device 1 of the present embodiment performs a magnification conversion process on the input image D1 in either the vertical direction or the horizontal direction to obtain a square image. The generated square image is used as the image to be processed D2, and the image processing from the Fourier transform unit 21 to the inverse Fourier transform unit 52 is performed. Therefore, it is possible to appropriately perform processing such as generating mask data by calculating the logical product of the first image D5 and the second image D6 for each pixel, and the frequency of the halftone dots included in the input image D1. The components can be appropriately attenuated. Since the information regarding the frequency component of the halftone dots is not lost even if the magnification conversion process for the input image D1 is performed as in the present embodiment, the frequency component of the halftone dots can be attenuated normally.

Further, since the image processing device 1 of the present embodiment generates one square image from one input image D1, the number of times of image processing from the Fourier transform unit 21 to the inverse Fourier transform unit 52 is one input image. It is excellent in processing efficiency because it only needs to be performed once per D1. That is, the method according to the present embodiment can perform image processing more efficiently than the method described in the second embodiment.

The configurations and operations other than those described above in the present embodiment are the same as those described in the first embodiment.

(Modification example)
Although the embodiments relating to the present invention have been described above, the present invention is not limited to the contents described in the above embodiments, and various modifications can be applied.

For example, in the first embodiment, when the power spectrum image D4 is binarized in the first image generation unit 20 to generate the first image D5, the power spectrum image D4 is within a predetermined distance R from the origin of the frequency coordinate axis. An example of converting the region P1 of the above into a predetermined value (for example, a value of “0”) has been described. However, the process of converting the region P1 within the range of the predetermined distance R from the origin of the frequency coordinate axis to a predetermined value (for example, the value of "0") is not necessarily limited to the process performed at the time of the binarization process. do not have. For example, when the mask data generation unit 40 generates the mask data D7, the region P1 within a predetermined distance R from the origin of the frequency coordinate axis in the first image D5 or the second image D6 is set to a predetermined value (for example, "" It may be set to the value of "0"). Even in this case, there is no change in the action and effect described in the first embodiment.

Further, in the above embodiment, the case where the image processing unit 10 mainly performs image processing by software processing is illustrated, but the present invention is not limited to this. That is, the image processing unit 10 realizes each of the image acquisition unit 15, the first image generation unit 20, the second image generation unit 30, the mask data generation unit 40, and the output image generation unit 50 in a hardware configuration. It doesn't matter.

Further, in the above embodiment, a case where the function of the image processing unit 10 for reducing the influence of halftone dots is mounted on the image processing device 1 configured as an MFP is illustrated. However, the above-mentioned image processing unit 10 is not necessarily limited to that mounted on an image processing device 1 such as an MFP. For example, the function of the image processing unit 10 described above may be realized by a computer such as a general personal computer.

Further, in the above embodiment, the case where the program 14 executed by the CPU 11 of the image processing unit 10 is stored in the ROM 12 in advance is illustrated. However, the program 14 may be installed in the image processing unit 10 after the fact. In this case, the program 14 may be provided in a manner that can be installed in the image processing unit 10 via a network or the like, or is recorded on a computer-readable recording medium such as a CD-ROM or a USB memory. It may be provided in the state of being provided.

1 Image processing device 2 Scanner unit 4 Image reading unit (image reading means)
8 Control unit (control means)
10 Image processing unit 15 Image acquisition unit (image acquisition means)
20 First image generation unit (first image generation means)
30 Second image generation unit (second image generation means)
40 Mask data generation unit (mask data generation means)
50 Output image generation unit (output image generation means)

Claims (19)

Image acquisition means for acquiring input images including halftone dots, and
A first image generation means for generating a first image representing the spatial frequency component of the input image based on the input image, and
A second image generation means for generating a second image obtained by rotating the first image by a predetermined angle,
A mask data generation means that compares the first image with the second image and generates mask data corresponding to the spatial frequency component of the input image.
An output image generation means that performs mask processing on the input image based on the mask data and generates an output image.
An image processing device characterized by comprising. The image processing apparatus according to claim 1, wherein the second image generation means generates the second image by rotating the first image 90 degrees clockwise. The image processing apparatus according to claim 1, wherein the second image generation means generates the second image by rotating the first image 90 degrees counterclockwise. The first image generation means according to any one of claims 1 to 3, wherein when the first image is generated, the first image is binarized based on a predetermined threshold value. Image processing device. The first image generation means is characterized in that, when performing the binarization process, a region in a predetermined range centered on the origin of the frequency coordinate axis in the first image is replaced with a predetermined value. The image processing apparatus according to 4. The mask data generation means is characterized in that when the mask data is generated, a predetermined value is set in a predetermined range region centered on the origin of the frequency coordinate axis in the first image or the second image. The image processing apparatus according to any one of claims 1 to 5. The image processing apparatus according to any one of claims 1 to 6, wherein the mask data generation means generates the mask data by performing a logical operation between the first image and the second image. Claim 1 is characterized in that the mask data generation means generates the mask data by performing an opening process or a closing process on the data obtained by comparing the first image and the second image. 7. The image processing apparatus according to any one of 7. The mask data generation means according to claim 1 to 8, wherein the mask data generation means generates the mask data by performing an expansion process on the data obtained by comparing the first image and the second image. The image processing apparatus according to any one. The image processing apparatus according to any one of claims 1 to 9, wherein the output image generation means generates the output image by performing a logical operation between the mask data and the input image. The first image generation means according to any one of claims 1 to 10, wherein the first image generation means performs a Fourier transform on the input image and generates the first image by converting the input image into a power spectrum image. The image processing device described. The output image generation means is characterized in that the output image is generated by performing an inverse Fourier transform based on the data obtained by performing a logical calculation between the mask data and the data after Fourier transform of the input image. The image processing apparatus according to claim 11. Image reading means for reading the image of the original,
Further prepare
The image processing apparatus according to any one of claims 1 to 12, wherein the image acquisition means acquires an image generated by the image reading means reading a document as the input image. Whether or not the first image generation means, the second image generation means, the mask data generation means, and the output image generation means are operated according to the reading mode when the image reading means reads the original. Control means to switch between
The image processing apparatus according to claim 13, further comprising. When the reading mode is a print document reading mode or a copy document reading mode, the control means obtains the first image generation means, the second image generation means, the mask data generation means, and the output image generation means. The image processing apparatus according to claim 14, wherein the image processing apparatus is operated. The image acquisition means divides the input image into a plurality of block images having the same vertical and horizontal sizes.
The first image generation means generates the first image based on each of the plurality of block images.
The output image generation means according to any one of claims 1 to 15, wherein the output image generation means generates the output image by synthesizing a plurality of output block images generated from each of the plurality of block images. Image processing device. When the vertical and horizontal sizes of the input image are different, the image acquisition means performs magnification conversion processing for either the vertical direction or the horizontal direction on the input image in the vertical direction and the horizontal direction. Generates processed images of equal size and
The first image generation means generates the first image based on the image to be processed.
The output image generation means is characterized in that the output image is generated by performing a process that is an inverse conversion of the magnification conversion process on the output target image generated from the process target image. The image processing apparatus according to any one of 15. The first step to acquire an input image including halftone dots,
A second step of generating a first image representing the spatial frequency component of the input image based on the input image, and
A third step of generating a second image obtained by rotating the first image by a predetermined angle,
A fourth step of comparing the first image with the second image and generating mask data corresponding to the spatial frequency component of the input image.
In the fifth step of performing mask processing on the first image based on the mask data and generating an output image,
An image processing method characterized by having. On the computer
The first step to acquire an input image including halftone dots,
A second step of generating a first image representing the spatial frequency component of the input image based on the input image, and
A third step of generating a second image obtained by rotating the first image by a predetermined angle,
A fourth step of comparing the first image with the second image and generating mask data corresponding to the spatial frequency component of the input image.
In the fifth step of performing mask processing on the first image based on the mask data and generating an output image,
A program characterized by executing.

Images