JP7225858B2 - Inspection device, inspection method and inspection program - Google Patents

Description

The present invention relates to an inspection device, an inspection method, and an inspection program, and more particularly, to an inspection device for detecting noise in an image formed on a recording medium, an inspection method executed by the inspection device, and an inspection that causes a computer to execute the inspection method. Regarding the program.

2. Description of the Related Art An apparatus for inspecting an image formed on a recording medium such as paper in an image forming apparatus such as an MFP (Multi Function Peripheral) is known. For example, Japanese Patent Application Laid-Open No. 2009-43111 describes an image processing method for detecting, from image data, pixel defects caused by dust or scratches adhering to a transparent original. This image processing method includes the steps of generating grayscale image data from color image data, dividing the grayscale image data into a plurality of blocks, and for each of the blocks, a block within each block. calculating a standard deviation of the grayscale image data of pixels in each of said blocks; calculating an average value of said grayscale image data of pixels within each block; calculating a threshold value using the standard deviation, a coefficient determined according to the standard deviation, and the average value; and detecting.

However, in the image processing method described in Japanese Patent Application Laid-Open No. 2009-43111, the threshold value is determined from the average value and standard deviation of the grayscale image data of the pixels in the block. There is a problem that false detection increases. Specifically, if density unevenness caused by an engine that forms an image in an image forming apparatus exists even in a halftone uniform density portion, the density unevenness may be erroneously detected as a pixel defect. .

Japanese Patent Application Laid-Open No. 2009-43111

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and one of the objects of the present invention is to provide an inspection apparatus capable of improving detection accuracy.

Another object of the present invention is to provide an inspection method capable of improving detection accuracy.

Still another object of the present invention is to provide an inspection program capable of improving detection accuracy.

In order to achieve the above object, according to one aspect of the present invention, an inspection apparatus processes each of a plurality of pixels constituting processing target data in which a plurality of pixels are arranged two-dimensionally as a pixel of interest. and noise detection means for detecting noise from the data to be processed, wherein the noise detection means indicates the characteristics of the surrounding area based on the pixel values of a plurality of pixels included in the surrounding area determined by a predetermined relative position to the pixel of interest. feature amount determining means for determining a quantity; threshold value determining means for determining a threshold value for a pixel of interest based on the feature amount; and comparing means for comparing the determined threshold value with the pixel value of the pixel of interest. candidate determination means for determining the target pixel as a candidate pixel when the pixel value of the target pixel is out of the range determined by the threshold value determined for the target pixel;
a noise region determining means for determining, as a noise region in which noise appears, a region consisting of a set of candidate pixels in which a predetermined number or more are consecutive .

According to this aspect, the feature amount of the target pixel and the surrounding area determined by the predetermined relative position is determined, the threshold value for the target pixel is determined based on the feature amount, and the threshold value and the pixel value of the target pixel are determined. be compared. Therefore, the threshold value is determined without using the pixel value of the pixel of interest, so that the threshold value is determined to be an appropriate value. As a result, it is possible to provide a detection device capable of improving detection accuracy.

Preferably, sub-feature determining means for determining a sub-feature representing a block feature for each of a plurality of blocks included in the peripheral region; candidate block determining means for determining one or more candidate blocks from among them, wherein the feature amount determining means determines the feature amount of the surrounding area based on pixel values of a plurality of pixels included in each of the one or more candidate blocks. to decide.

According to this aspect, one or more candidate blocks are determined from among the plurality of blocks based on sub-feature amounts determined for each of the plurality of blocks included in the peripheral region, and are included in each of the one or more candidate blocks. A feature amount of the peripheral area is determined based on the pixel values of the plurality of pixels. Therefore, the feature amount is determined by excluding pixels whose pixel values are different from those of other pixels among the plurality of pixels included in the peripheral region, so that the accuracy of the threshold value can be improved.

Preferably, for each of a plurality of sampling regions each determined by a predetermined relative position to the pixel of interest, a sampling feature quantity for determining a sampling feature quantity representing a feature of the sampling region based on pixel values of a plurality of pixels included in the sampling region. The apparatus further comprises determining means, and peripheral area determining means for determining one or more of the plurality of sampling areas as the peripheral area based on the sampling feature amount determined for each of the plurality of sampling areas.

According to this aspect, one or more of the plurality of sampling regions are determined as the peripheral region based on the sampling feature amounts of the plurality of sampling regions determined by the pixel of interest and the predetermined relative position. For this reason, the sampling areas containing characters or the like can be excluded from among the plurality of sampling areas, so that the precision of the threshold value can be improved.

Preferably, the sampling feature amount is the standard deviation or the standard deviation and average value of pixel values of a plurality of pixels.

According to this aspect, it is possible to determine whether or not the pixel of interest has a unique value in comparison with a plurality of pixels included in the surrounding area.

Preferably, the image forming apparatus further comprises reading means for reading a recording medium on which an image is formed and outputting image data, and processing target data generating means for generating processing target data from the image data.

According to this aspect, a defect in image formation by the image forming apparatus can be detected.

Preferably, an original data obtaining means for obtaining original data that is the basis of an image formed on a recording medium by the image forming apparatus; a characteristic area determining means for determining a characteristic area in which a predetermined characteristic is expressed in the original data; wherein the feature quantity determination means determines the feature quantity of the peripheral region based on a plurality of pixels included in a region other than the feature region in the peripheral region.

According to this aspect, the feature amount of the peripheral region is determined based on the plurality of pixels included in the peripheral region other than the feature region in which the predetermined feature is expressed in the original data. Therefore, since the feature area included in the original data is excluded from the peripheral area, it is possible to improve the accuracy of the threshold value.

Preferably, the processing target data generating means generates the processing target data by changing the resolution of the image data output by the reading means.

According to this aspect, the processing target data is generated by changing the resolution of the image data, so high-frequency noise contained in the image data can be removed.

According to another aspect of the present invention, the inspection apparatus detects noise from the data to be processed by processing each of a plurality of pixels constituting the data to be processed in which a plurality of pixels are arranged two-dimensionally as a pixel of interest. a noise detection means for reading a recording medium on which an image is formed by an image forming apparatus, and a reading means for outputting image data; and a processing target data generating means for generating processing target data from the image data; The generating means generates data to be processed by changing the resolution of the image data output by the reading means, and the noise detecting means detects pixel values of a plurality of pixels included in a peripheral area determined by a predetermined relative position to the target pixel. feature amount determination means for determining a feature amount indicating a feature of a peripheral region based on the threshold value determination means for determining a threshold value for a target pixel based on the feature amount; and the determined threshold value and the target pixel and a comparison means for comparing the pixel value of the target pixel with the pixel value of the target pixel, and the noise detection means determines that noise appears in the target pixel when the pixel value of the target pixel is outside the range determined by the threshold value determined for the target pixel. Decide on the noise region.

According to this aspect, a noise region having the same size as the pixel of interest can be detected.

Preferably, the processing target data generating means generates data obtained by setting an average value of pixel values of a plurality of pixels in a predetermined range including the pixel in place of the pixel values of the plurality of pixels included in the image data output by the reading means. Generate as data to be processed.

According to this aspect, data in which the pixel values of the plurality of pixels included in the image data are set to the average value of the pixel values of the plurality of pixels in the predetermined range including the pixel is generated as the processing target data. Therefore, it is possible to remove high-frequency fluctuation components included in the image data.

Preferably, the noise detection means includes candidate determination means for determining the pixel of interest as a candidate pixel when the pixel value of the pixel of interest is outside the range determined by the threshold value determined for the pixel of interest; A noise area determining means for determining an area consisting of a set of consecutive candidate pixels as a noise area in which noise appears.

According to this aspect, it is possible to detect a noise region that is larger in size than the pixel of interest.

Preferably, the peripheral area is an area around the pixel of interest excluding the pixel of interest.

According to this aspect, the pixel value of the pixel of interest does not serve as a parameter for determining the threshold value, so the accuracy of the threshold value can be improved.

Preferably, the peripheral area is an area that is at least the first distance away from the pixel of interest.

According to this aspect, it is possible to detect noise having a size larger than the size of the pixel of interest.

Preferably, the peripheral area is an area below the second distance from the pixel of interest.

Preferably, the peripheral area is an area within a second distance in a predetermined direction from the pixel of interest.

According to this aspect, it is possible to detect streak noise that is orthogonal to a predetermined direction.

Preferably, the peripheral area includes a first area at a second distance or less in a first direction from the pixel of interest, and a second area at a second distance or less in a second direction intersecting the first direction from the pixel of interest.

According to this aspect, noise of stripes parallel to the first direction and noise of stripes parallel to the second direction can be detected.

According to still another aspect of the present invention, an inspection method is an inspection method executed by an inspection apparatus, wherein each of a plurality of pixels constituting processing target data in which a plurality of pixels are two-dimensionally arranged is a pixel of interest. The noise detection step includes a noise detection step of detecting noise from the data to be processed by processing as, the noise detection step detects the surrounding area based on the pixel values of a plurality of pixels included in the surrounding area determined by the pixel of interest and a predetermined relative position a feature amount determination step of determining a feature amount indicating the feature of a pixel of interest; a threshold determination step of determining a threshold for a pixel of interest based on the feature amount; It includes a comparison step of comparing, and a step of detecting noise from the processing target data based on the comparison result in the comparison step .

According to this aspect, it is possible to provide an inspection method capable of improving detection accuracy.

According to still another aspect of the present invention, the inspection program removes noise from the data to be processed by processing each of the plurality of pixels constituting the data to be processed in which the plurality of pixels are arranged two-dimensionally as a pixel of interest. A noise detection step is executed by a computer, and the noise detection step determines a feature amount indicating the characteristics of the surrounding area based on the pixel values of a plurality of pixels included in the surrounding area determined by the pixel of interest and a predetermined relative position. a threshold value determination step of determining a threshold value for a pixel of interest based on the feature value; a comparison step of comparing the determined threshold value with the pixel value of the pixel of interest; a candidate determination step of determining the pixel of interest as a candidate pixel when the pixel value of is out of the range determined by the threshold value determined for the pixel of interest;
and a noise region determination step of determining a region formed by a set of candidate pixels in which a predetermined number or more of consecutive pixels is a noise region in which noise appears.

According to this aspect, it is possible to provide an inspection program capable of improving detection accuracy.

1 is a perspective view showing the appearance of an image forming system according to one embodiment of the present invention; FIG. It is a block diagram which shows an example of the outline|summary of the hardware configuration of an inspection apparatus. It is a block diagram which shows an example of the function which CPU with which an inspection apparatus has. It is a figure which shows an example of a surrounding area. FIG. 4 is a diagram showing an example of a plurality of blocks included in a peripheral area; FIG. It is a figure which shows an example of a pixel value. 6 is a flowchart showing an example of the flow of detection processing; 4 is a flowchart showing an example of the flow of noise detection processing; FIG. 4 is a diagram showing an example of a sampling area; FIG. 11 is a block diagram showing an example of functions of a CPU included in an inspection apparatus according to a second modified example; FIG. 11 is a flowchart showing an example of the flow of noise detection processing in the second modified example; FIG. FIG. 11 is a diagram showing an example of a peripheral area in a third modified example; FIG. 14 is a diagram showing an example of a peripheral area in a fourth modified example;

An image forming apparatus according to embodiments of the present invention will be described below with reference to the drawings. In the following description, the same parts are given the same reference numerals. Their names and functions are also the same. A detailed description thereof will therefore not be repeated.

FIG. 1 is a perspective view showing the appearance of an image forming system according to one embodiment of the present invention. Referring to FIG. 1, the image forming system includes an MFP 100 which is a multi-function machine, an inspection device 200 and a post-processing device 300 . MFP 100 functions as an image forming apparatus and forms an image on a recording medium based on image data. The MFP 100 can form an image on any of a plurality of types of recording medium as a recording medium on which an image is to be formed. The recording medium includes paper such as paper, OHP (overhead projector) sheet, cloth, and the like. Also, in the following description, unless otherwise specified, a case where paper is used as a recording medium will be described as an example.

The inspection apparatus 200 is supplied with a sheet having an image formed thereon from the MFP 100, and detects, as noise, portions of the image formed on the sheet that are caused by dust, dirt, and defects during image formation. The inspection apparatus 200 includes a first scanner 206 and a second scanner 207, which are photoelectric conversion elements. The first scanner 206 and the second scanner 207 are arranged across a transport path for paper supplied from the MFP 100 . Therefore, the inspection apparatus 200 can simultaneously read the images formed on the front and back sides of the paper supplied from the MFP 100 .

Post-processing device 300 is supplied with paper from inspection device 200 . The post-processing device 300 has a plurality of paper discharge trays, and has a sorting function of rearranging a plurality of sheets and discharging them to one of the plurality of paper discharge trays. Further, the post-processing device 300 discharges the paper on which noise is detected from the image formed on the paper by the inspection device 200 to a paper discharge tray different from the paper on which noise is not detected. Specifically, when an error signal is input from the inspection device 200, the paper corresponding to the error signal is ejected to a predetermined paper ejection tray. Post-processing device 300 may include a punching mechanism and a stapling mechanism. The punch mechanism is a mechanism that punches holes in a bundle of sheets of paper. The stapling mechanism is a mechanism for driving staples into a bundle of a plurality of sheets.

FIG. 2 is a block diagram showing an example of an overview of the hardware configuration of the inspection device. Referring to FIG. 2, inspection apparatus 200 includes a central processing unit (CPU) 201 for controlling the entire inspection apparatus 200, and a ROM (Read Only Memory) 202 for storing programs to be executed by CPU 201. , a RAM (Random Access Memory) 203 used as a work area for the CPU 201, an EPROM (Erasable Programmable ROM) 204 for storing data in a nonvolatile manner, a communication unit 205 for connecting the CPU 201 to the MFP 100 and the post-processing device 300, A first scanner 206 , a second scanner 207 and a serial interface 210 are included.

A serial interface 210 is an interface for serial communication with an external device. Here, the serial communication is the USB (Universal Serial Bus) standard. The serial interface 210 can be connected to an external device that can communicate according to the USB standard. The CPU 201 can access external devices via the serial interface 210 . The external device includes storage devices such as a USB memory 211 and a CD drive. Here, a case where the external device is the USB memory 211 will be described as an example. The USB memory 211 includes a semiconductor memory such as EPROM and a serial communication circuit. The CPU 201 loads a program recorded in the USB memory 211 attached to the serial interface 210 into the RAM 203 and executes it.

Each of the first scanner 206 and the second scanner 207 is a photoelectric conversion element that converts received light into image data as an electrical signal and outputs the image data. In this embodiment, the first scanner 206 and the second scanner 207 are CCDs (Charge Coupled Devices). Note that the photoelectric conversion element is not limited to a CCD, and other elements such as a CMOS sensor may be used. First scanner 206 optically reads an image formed on the surface of a sheet supplied from MFP 100 and outputs image data, which is electronic data, to CPU 201 . Second scanner 207 optically reads an image formed on the back surface of a sheet supplied from MFP 100 and outputs image data, which is electronic data, to CPU 201 . Based on the image data output by the first scanner 206, the CPU 201 detects noise from the image formed on the surface of the paper, and based on the image data output by the second scanner 207, the CPU 201 detects noise formed on the surface of the paper. detect noise from the image.

The medium for storing the program executed by the CPU 201 is not limited to the USB memory 211, and may be an optical disk (CD-ROM (Compact Disk ROM), MO (Magnetic Optical Disc)/MD (Mini Disc)/DVD (Digital Versatile Disc). )), optical card, mask ROM. The programs here include not only programs that can be directly executed by the CPU 201, but also source programs, compressed programs, encrypted programs, and the like.

Also, the inspection device 200 may be connected to the Internet via the communication unit 205 or the serial interface 210 . In this case, CPU 201 can download a program from a computer connected to the Internet and store it in EPROM 204 . Also, a program may be written to the EPROM 204 by a computer connected to the Internet. The CPU 201 loads a program stored in the EPROM 204 into the RAM 203 and executes it.

FIG. 3 is a block diagram showing an example of functions of a CPU provided in the inspection apparatus. The functions shown in FIG. 3 are realized by the CPU 201 of the inspection apparatus 200 by executing the inspection program stored in the ROM 202 , the EPROM 204 or the USB memory 211 . Referring to FIG. 3 , CPU 201 includes a reading control portion 51 , a processing target data generation portion 53 , a noise detection portion 55 , an original data acquisition portion 57 and a characteristic region determination portion 59 .

The reading control unit 51 controls the first scanner 206 and the second scanner 207 . When MFP 100 forms an image on the surface of a sheet, reading control unit 51 controls first scanner 206 to read the image formed on the surface of the sheet, and processes the image data output by first scanner 206. Output to the target data generation unit 53 . In this case, the reading control unit 51 does not drive the second scanner 207 . When MFP 100 forms images on the front and back sides of a sheet, reading control unit 51 controls first scanner 206 to read the image formed on the front side of the sheet, and controls second scanner 207 to print the sheet. to read the image formed on the back surface of the The read control unit 51 outputs the image data output by the first scanner 206 and the image data output by the second scanner 207 to the processing target data generation unit 53 . Image data is data in which pixels are arranged two-dimensionally. A pixel contains a pixel value. The image data may be color or monochrome. The color image data is data indicating the lightness of each color of red (R), blue (G), and green (G). Monochrome image data is data indicating brightness.

The image data output by the first scanner 206 and the image data output by the second scanner 207 differ only in the image to be read, and the processing executed by the CPU 201 is the same. Therefore, in the following description, the case where the image data output by the first scanner 206 is output to the processing target data generation unit 53 will be described as an example, unless otherwise specified.

The processing target data generation unit 53 generates processing target data from the image data output from the reading control unit 51 . The data to be processed is data to be processed by the noise detection unit 55 . When color image data is input from the reading control unit 51, processing target data corresponding to each of the R, G, and B image data is generated. When monochrome image data is input from the read control unit 51, processing target data corresponding to each of the monochrome image data is generated. The R, G, and B image data and the monochrome image data differ only in color, and the same processing is performed on the image data. A case where is input will be described as an example.

When forming an image on a sheet of paper, the MFP 100 forms the image on the sheet of paper after performing tone reproduction processing on the original data that is the basis of the image. Tone reproduction processing includes, for example, error diffusion processing and screen processing. Therefore, the image formed on the paper by the MFP 100 contains high-frequency fluctuation components. The processing target data generation unit 53 generates processing target data by removing high-frequency fluctuation components from the image data. The processing target data generation unit 53 includes a resolution conversion unit 61 and an averaging unit 63 . The resolution converter 61 converts the resolution of image data. Specifically, the resolution conversion unit 61 reduces the image data to reduce the resolution. Reduction processing can remove high-frequency fluctuation components, and the number of pixels can be reduced, so the load of subsequent processing can be reduced.

The averaging unit 63 averages the image data. The averaging process is a process of setting the pixel value of each of a plurality of pixels included in image data to the average value of the pixel values of that pixel and a plurality of pixels surrounding that pixel. Pixels around a pixel refer to pixels adjacent to the pixel. Here, using a 3×3 rectangular filter, the pixel value of the central pixel of the nine pixels is set to the average value of the pixel values of the nine pixels. By executing the averaging process, high frequency components that could not be removed by the resolution conversion section 61 can be removed. Processing target data generation unit 53 outputs data obtained by performing resolution conversion processing and averaging processing on image data to noise detection unit 55 as processing target data. In this embodiment, the resolution conversion process and the averaging process are executed, but either one may be executed as long as the high-frequency fluctuation component can be removed from the image data.

Original data obtaining portion 57 controls communication portion 205 to obtain original data from MFP 100 . The original data is the original data of the image formed on the paper by the MFP 100 . Original data acquiring portion 57 outputs the original data acquired from MFP 100 to characteristic region determining portion 59 .

Characteristic region determination unit 59 determines a characteristic region included in the original data. A characteristic region is a region having a predetermined characteristic included in the image represented by the original data. Here, the predetermined features are edges and background. Characteristic region determining portion 59 outputs position information indicating the position of the characteristic region in the original data to noise detecting portion 55 . Characteristic region determination portion 59 includes an edge detection portion 65 and a background detection portion 67 .

The edge detection unit 65 detects pixels forming an edge from the original data, and determines a region including a plurality of pixels forming the edge as a feature region. A method of detecting a plurality of pixels forming an edge, for example, determines a pixel having a low brightness among a plurality of pixels having a brightness difference equal to or greater than a predetermined value as an edge pixel. A filter for detecting pixels forming an edge may be used.

The background detection unit 67 detects pixels forming the background from the original data, and determines an area including a plurality of pixels forming the background as a feature area. The background here is the area of the paper where no image was formed. A method of detecting a plurality of pixels forming the background determines, for example, pixels whose lightness is equal to or higher than a predetermined threshold as pixels forming the background. It should be noted that a filter may be used to detect the pixels forming the background.

The noise detection unit 55 detects noise in the image data by processing the processing target data generated by the processing target data generation unit 53 and detecting pixels forming a noise region. Noise detection portion 55 includes a target pixel selection portion 71 , a feature quantity determination portion 73 , a threshold value determination portion 75 , a comparison portion 77 , a candidate pixel determination portion 79 , and a noise region determination portion 81 .

The pixel-of-interest selection unit 71 sequentially selects one of a plurality of pixels included in the data to be processed as a pixel of interest. When the comparison by the comparison unit 77 is completed for the target pixel, the target pixel selection unit 71 newly selects one of the one or more pixels that are not selected as the target pixel from among the plurality of pixels included in the processing target data. the target pixel. The target pixel selection unit 71 outputs pixel identification information for identifying the target pixel to the feature amount determination unit 73 and the comparison unit 77 each time it selects the target pixel.

The feature quantity determination unit 73 determines the feature quantity of the peripheral region whose relative position to the pixel of interest is determined. The peripheral area is an area around the pixel of interest excluding the pixel of interest. Preferably, the peripheral area is an area that is at least the first distance away from the pixel of interest. Also, the peripheral area is an area below the second distance from the pixel of interest. The second distance is greater than the first distance.

FIG. 4 is a diagram showing an example of the peripheral area. Referring to FIG. 4, the peripheral area 511 is an area surrounded by a rectangular outer frame with one side of M pixels and a rectangular inner frame with one side of N pixels. The peripheral region 511 is positioned relative to the pixel of interest 501 . Specifically, the relative positions of the peripheral region 511 and the target pixel 501 are determined such that the target pixel is positioned at the center of the outer frame and the center of the inner frame of the peripheral region 511 .

The first distance of the peripheral region 511 is M/√2 and the second distance is N/2. Note that the unit is the size of a pixel. N is determined by the size of the noise region in which the noise present in the image data appears. It is preferable to determine the value of N so that the noise region is contained within the inner frame of the surrounding region when the pixel of interest is positioned at the edge of the noise region. The noise region is a detection target of the noise detection section 55 . For example, when the size of the noise region is determined in advance by experiment, N is determined from the size twice that size. Also, M determines the number of pixels included in the peripheral region 511 . If the number of pixels is small, the accuracy of the feature amount described later is lowered, and if the number of pixels is large, the load for calculating the feature amount increases. For this reason, it is preferable to determine it by considering the balance between accuracy and load.

Note that the size of a plurality of pixels included in the processing target data is determined by the reduction ratio in the resolution conversion when the processing target data is generated by resolution-converting the image data by the processing target data generation unit 53 . Further, in the present embodiment, one type of peripheral region is determined for the pixel of interest, but a plurality of types of peripheral regions having different inner frame sizes may be determined. In this case, it is possible to detect a plurality of types of noise regions with different sizes.

FIG. 5 is a diagram showing an example of a plurality of blocks included in the peripheral area. Referring to FIG. 5, the peripheral area 511 includes 16 blocks AP. The size and shape of the 16 blocks are the same. Note that although the plurality of blocks have the same size and shape here, the plurality of blocks may not have the same size and shape.

Returning to FIG. 3 , the feature amount determination section 73 includes a sub-feature amount determination section 85 , a candidate block determination section 87 and an exclusion section 89 .

The excluding unit 89 excludes a block including at least part of the characteristic region from among the plurality of blocks. The characteristic regions are regions determined by the characteristic region determination unit 59, and include a region forming the background and a region forming the edge. The area that constitutes the background in the original data is originally an area where no toner adheres to the paper, and is an area where no streaks or spots occur on the paper. Spots are generated in areas where toner or the like adheres to the paper, and are often circular in shape. A region forming an edge has a large change in brightness, and is a region in which a character or the like is represented.

The sub-feature amount determination unit 85 determines a sub-feature amount for each of the plurality of remaining blocks that have not been excluded by the exclusion unit 89 among the plurality of blocks. A sub-feature amount is the average and standard deviation of the pixel values of a plurality of pixels included in the block. Since the sub-feature amounts of the plurality of blocks excluded by the exclusion unit 89 are not calculated, the sub-feature amounts can be prevented from being affected by the background or characters, and the processing load can be reduced.

The candidate block determination unit 87 determines one or more candidate blocks from among the plurality of remaining blocks that have not been excluded by the exclusion unit 89 among the plurality of blocks. The candidate block determination unit 87 determines a block whose sub-feature amount satisfies a predetermined condition as a candidate block. A predetermined condition is a predetermined condition. In this example, the predetermined condition is that the standard deviation is equal to or less than the first threshold. The predetermined condition may be that the standard deviation is equal to or less than the first threshold and the average value is equal to or less than the second threshold, or that the average value is equal to or less than the second threshold.

The feature quantity determination unit 73 determines the feature quantity of the peripheral region using pixel values of a plurality of pixels in one or more candidate blocks as parameters. The feature amount is the average and standard deviation of pixel values. Feature amount determining portion 73 outputs the determined feature amount to threshold value determining portion 75 .

The threshold determination unit 75 determines a threshold for the pixel of interest based on the feature amount of the surrounding area corresponding to the pixel of interest. The threshold includes an upper limit value and a lower limit value. The threshold determination unit 75 determines the value obtained by adding the value obtained by multiplying the standard deviation by the coefficient to the average value as the upper limit value, and determines the value obtained by subtracting the value obtained by multiplying the standard deviation by the coefficient from the average value as the lower limit value. do. Threshold determining portion 75 outputs the threshold to comparing portion 77 . In this example, the coefficient is 3. In order to improve the detection accuracy, the coefficient should be set to a smaller value.

A comparison unit 77 compares the pixel value of the pixel of interest with a threshold value. The comparison unit 77 outputs a selection instruction to the pixel-of-interest selection unit 71 upon completion of the comparison. The candidate pixel determination unit 79 determines the target pixel as a candidate pixel when the pixel value of the target pixel is out of the range determined by the threshold as a result of the comparison by the comparison unit 77 . The candidate pixel determination unit 79 does not determine the target pixel as a candidate pixel when the pixel value of the target pixel is within the range determined by the threshold as a result of the comparison by the comparison unit 77 . When determining the pixel of interest as a candidate pixel, candidate pixel determination section 79 outputs candidate pixel identification information for identifying the candidate pixel to noise region determination section 81 .

The pixel-of-interest selection unit 71 selects a new pixel of interest in response to a selection instruction being input from the comparison unit 77 . Therefore, when the target pixel selection unit 71 selects all of the plurality of pixels included in the processing target data, the comparison unit 77 compares the pixel values of the plurality of pixels included in the processing target data with the threshold value.

The noise region determination unit 81 determines the noise region after the candidate pixel determination unit 79 determines whether or not all the plurality of pixels included in the processing target data are candidate pixels. A noise region is a region containing a predetermined number or more of a plurality of continuous candidate pixels. The predetermined number is predetermined based on the size of the noise region. For example, assuming that the noise region is circular, the number of pixels included in the circular region having a diameter greater than or equal to that diameter may be used. If even one noise region is detected, noise region determination section 81 outputs an error signal to post-processing device 300 .

FIG. 6 is a diagram showing an example of pixel values. Referring to FIG. 6, the pixel values of the pixels forming the spots and their surrounding pixels are shown. The pixels that form the spots form a noise area that is included in the image data due to the adhesion of toner or the like to the paper that is the object of reading. A solid line indicates the threshold value calculated when the peripheral region 511 shown in FIG. 5 is determined for the pixels forming the spot. The threshold value is determined from the average value AV1 and the standard deviation of the pixel values of a plurality of pixels included in the peripheral region 511, the upper limit value TH1 obtained by adding three times the standard deviation to the average value AV1, and the standard deviation from the average value AV1. It is shown as the lower limit TL1 minus 3 times. Since the threshold value is determined for each pixel of interest, the threshold values of the pixels other than the pixels forming the spot are different from the threshold values determined for the pixels forming the spot, but are approximate values. Therefore, in the drawing, the threshold values of the pixels other than the pixels forming the spots are shown as the same values as the threshold values of the pixels forming the spots.

A dotted line indicates the threshold determined by the conventional method. Specifically, when a rectangular area of 15 pixels vertically and 15 pixels horizontally is determined centering on the pixel forming the spot, the calculated threshold value is indicated. The threshold value is obtained from the average value AV2 and the standard deviation of the pixel values of a plurality of pixels included in the rectangular area, the upper limit value TH2 obtained by adding three times the standard deviation to the average value AV2, and the standard deviation from the average value AV2. It is shown as the lower limit TL2 with three times subtracted.

In the threshold value of the conventional method, the pixel values of the pixels forming the spot are included in the parameters for calculating the threshold value. The pixel value is not included in the parameters for calculating the threshold. Specifically, in the present embodiment, the average value is calculated by excluding the pixel values of the pixels forming the spots. Therefore, the average value of the peripheral portion of the pixel of interest becomes closer to the true value than in the conventional case. Also, in this embodiment, the range of variation in pixel values is smaller than in the conventional case. In other words, the difference between the upper limit TH1 and the lower limit TL1 is smaller than the difference between the upper limit TH2 and the lower limit TL2. Therefore, the threshold in this embodiment is more accurate than the conventional threshold.

According to the conventional method, the pixel value of the pixel constituting the spot is larger than the upper limit value TH2, and the pixel The value is smaller than the lower limit value TL2. Therefore, a portion including pixels P1, P2, P3, P4, P5, and P6 is detected as noise.

On the other hand, according to the inspection apparatus 200 of the present embodiment, the pixel values of the pixels forming the spots are larger than the upper limit value TH1, and the pixels P1, P2, P3, P4, P5, and P6 are respectively selected. The pixel value is equal to or greater than the lower limit value TL1 in a portion of a predetermined size including the pixel value. Therefore, the peripheral portions including the pixel P1, pixel P2, pixel P3, pixel P4, pixel P5, and pixel P6 are not detected as noise.

Further, according to the inspection apparatus 200 of the present embodiment, since the upper limit TH1 is smaller than the upper limit TH2 in the conventional method, the size of the area detected as a spot is larger than the size detected as a spot in the conventional method. will also grow.

FIG. 7 is a flowchart showing an example of the flow of detection processing. The detection process is a process executed by CPU 201 of inspection apparatus 200 when CPU 201 executes an inspection program stored in ROM 202 , EPROM 204 or USB memory 211 . Referring to FIG. 7, CPU 201 determines whether an image has been read. The process waits until the image is read (NO in step S01), and if the image is read (YES in step S01), the process proceeds to step S02. A sheet on which an image is formed is supplied from the MFP 100, and it is determined whether or not the first scanner 206 has read the image formed on the sheet.

In step S02, the image data output by the first scanner 206 is subjected to resolution conversion processing. The resolution of the image data output by the first scanner 206 is converted, and the process proceeds to step S03. Here, reduction processing is performed to reduce the resolution in order to remove high frequency components contained in the image data.

In the next step S03, an averaging process is executed, and the process proceeds to step S04. The image data whose resolution has been converted in step S02 is to be processed. Averaging processing is processing for converting the pixel value of each of a plurality of pixels included in image data into the average value of the pixel values of that pixel and its surrounding pixels. As a result, high-frequency components included in the image data that could not be removed by the resolution conversion process are removed. Note that the averaging process can be omitted if the high frequency components are sufficiently removed by the resolution conversion process.

In step S04, original data is acquired from MFP 100, and the process proceeds to step S05. The original data is the data from which the image read in step S01 is based, and is the data from which the MFP 100 forms the image on paper. In step S05, edges are detected from the original data, and the process proceeds to step S06. A pixel whose difference in lightness between adjacent pixels is equal to or greater than a predetermined threshold value is detected as an edge. In step S06, the background is detected from the original data, and the process proceeds to step S07. A pixel whose pixel value is greater than or equal to a predetermined threshold value is detected as background.

In step S07, the region containing the pixels detected as the edge in step S05 and the region containing the pixels detected as the background in step S06 are determined as feature regions, and the process proceeds to step S08.

In step S08, noise detection processing is performed using the image data on which the averaging processing has been performed in step S03 as processing target data. Details of the noise detection process will be described later, but this is a process of detecting a noise area included in image data. In the next step S09, the process branches depending on the result of the noise detection process. If the noise area is detected, the process proceeds to step S10, and if the noise area is not detected, the process proceeds to step S11.

In step S10, post-processing device 300 is controlled to discharge the paper to the retraction tray, and the process returns to step S01. When the error signal is output to the post-processing device 300, the post-processing device 300 discharges the paper to the retraction tray. In step S11, post-processing device 300 is controlled to discharge the paper to the normal transport path, and the process returns to step S01. Since no error signal is output to the post-processing device 300, the post-processing device 300 discharges the paper to the normal transport path.

FIG. 8 is a flowchart showing an example of the flow of noise detection processing. The noise detection process is a process executed in step S08 of FIG. 7, and the characteristic region is determined before the noise detection process is executed. Also, in the noise detection process, the data to be processed is the target of the process. Referring to FIG. 8, one of a plurality of pixels included in the data to be processed is selected as a pixel of interest (step S21), and the process proceeds to step S22. In step S22, a plurality of blocks in the peripheral area are specified, and the process proceeds to step S23. A peripheral area corresponding to the pixel of interest is determined, and predetermined blocks A to P are specified in the peripheral area.

In the next step S23, blocks including characteristic regions are excluded from the processing targets from among the plurality of blocks specified in step S22, and the process proceeds to step S24. This determines a threshold that removes the effects of edges and background.

In step S24, sub-feature amounts are determined for each of a plurality of blocks, and the process proceeds to step S25. In this embodiment, the sub-feature amount is the standard deviation. The standard deviation of the pixel values of the pixels included in each block is calculated for each block. In step S25, a block whose sub-feature amount satisfies a predetermined condition is determined as a candidate block, and the process proceeds to step S26. The predetermined condition is that the standard deviation is equal to or less than the first threshold. The predetermined condition may be the condition that the standard deviation is equal to or less than the first threshold and the average value is equal to or less than the second threshold, or the condition that the average value is equal to or less than the second threshold.

In step S26, the feature amount of the peripheral area is determined, and the process proceeds to step S27. The average value AV1 and standard deviation of a plurality of pixels included in one or more blocks determined as the candidate area in step S25 are calculated as feature amounts of the peripheral area. In step S27, a threshold for the pixel of interest is determined based on the feature amount. Here, as the threshold values, an upper limit value TH1 obtained by adding three times the standard deviation to the average value AV1 and a lower limit value TL1 obtained by subtracting three times the standard deviation from the average value AV1 are determined.

In step S28, it is determined whether or not the pixel value of the pixel of interest is outside the threshold range. If the pixel value of the pixel of interest is larger than the upper limit TH1 or smaller than the lower limit TL1, it is determined to be out of the threshold range. If the pixel value of the pixel of interest is outside the threshold range, the process proceeds to step S29; otherwise, the process proceeds to step S30. In step S29, the pixel of interest is determined as a candidate pixel, and the process proceeds to step S30.

In step S30, it is determined whether or not there is an unselected pixel that has not been selected as the pixel of interest in the data to be processed. If there is an unselected pixel, the process returns to step S21, but if not, the process proceeds to step S31.

In step S31, a set of candidate pixels in which a predetermined number or more of consecutive pixels are extracted. The predetermined number is determined by the size of the noise region included in the processing target data. In other words, it is determined by the size of the area to be detected in the image data. Since the reduction ratio of the resolution conversion process is known, the size of the data to be processed is determined, and the size of the noise area included in the data to be processed is determined from the size of the area to be detected in the image data. The predetermined number is determined so that the length of a plurality of pixels continuous in one direction is the minimum value of the diameter of the noise region included in the processing target data.

In step S32, it is determined whether or not a set of candidate pixels in which a predetermined number or more of consecutive pixels have been extracted in step S31. If the set of candidate pixels has been extracted, the process proceeds to step S33; otherwise, the process returns to the detection process. In step S33, the area where the candidate pixels gather is determined as the noise area, and the process returns to the detection process.

<First modification>
The inspection apparatus 200 in the first modified example deletes pixel values instead of deleting blocks when calculating sub-feature amounts. Specifically, the sub-feature amount determination unit 85 in the first modified example selects the pixel values, which are the parameters for calculating the sub-feature amount, from the pixels other than the pixels included in the feature region among the plurality of pixels included in the block. Let it be the pixel value of the pixel. Further, the feature amount determination unit 73 sets the pixel values of the pixels other than the pixels included in the feature region among the plurality of pixels included in the candidate block as the pixel values used as the parameters for calculating the feature amount. As a result, even if a block includes a background or an edge, the pixel values of the pixels included in the background or edge do not serve as parameters for calculating sub-feature amounts and feature regions. Therefore, it is possible to prevent the sub-feature amount of the block and the feature amount of the surrounding area from being affected by the background or edge.

<Second modification>
The inspection apparatus 200 in the second modified example sets the peripheral area for the pixel of interest to one or more selected from among a plurality of sampling areas.

FIG. 9 is a diagram showing an example of sampling regions. FIG. 9A shows the first sampling area 521. FIG. The first sampling area 521 is positioned relative to the pixel of interest 501 . Specifically, the first sampling area 521 includes the peripheral area blocks A, B, C, F, and H shown in FIG. FIG. 9B shows the second sampling area 522. FIG. The second sampling area 522 is positioned relative to the pixel of interest 501 . Specifically, the second sampling area 522 includes the peripheral area blocks C, D, E, G, and I shown in FIG. FIG. 9C shows the third sampling area 523. FIG. A third sampling area 523 is positioned relative to the pixel of interest 501 . Specifically, the third sampling area 523 includes the peripheral area blocks H, J, L, M, and N shown in FIG. FIG. 9D shows the fourth sampling area 524. FIG. A fourth sampling area 524 is positioned relative to the pixel of interest 501 . Specifically, the fourth sampling area 524 includes the peripheral area blocks I, K, N, O, and P shown in FIG.

FIG. 10 is a block diagram showing an example of functions of a CPU included in an inspection apparatus according to the second modification. The function shown in FIG. 10 is different from the function shown in FIG. 3 in that a sampling feature quantity determination unit 91 and a peripheral region determination unit 92 are added. Since other functions are the same as those shown in FIG. 3, the description will not be repeated here.

The sampling feature amount determination unit 91 determines sampling feature amounts of each of the first sampling region 521 to the fourth sampling region 524 . A sampling feature amount is a standard deviation of pixel values of a plurality of pixels included in the sampling area.

The peripheral region determination unit 92 determines, as the peripheral region, a sampling region whose sampling feature amount satisfies a predetermined condition from among the first sampling region 521 to the fourth sampling region 524 . The predetermined condition is that the standard deviation is equal to or less than the first threshold. The predetermined condition may be the condition that the average value is equal to or less than the first threshold value and the average value is equal to or less than the second threshold value, or the condition that the average value is equal to or less than the second threshold value.

For example, if only the first sampling area 521 satisfies the predetermined condition and the other second sampling area 522 to fourth sampling area 524 do not satisfy the predetermined condition, the first sampling area 521 is determined as the peripheral area. Also, when all of the first to fourth sampling regions 521 to 524 satisfy a predetermined condition, all of the first to fourth sampling regions 521 to 524 are determined as the peripheral region. In this case, it becomes the same as the peripheral region 511 shown in FIG.

The detection process shown in FIG. 7 is executed in the CPU 201 included in the inspection apparatus 200 in the second modified example. However, the noise detection process in step S08 is different.

FIG. 11 is a flowchart showing an example of the flow of noise detection processing in the second modified example. 11, the difference from the noise detection process shown in FIG. 8 is that steps S41 to S43 are added between steps S21 and S22. Since other processing is the same as the processing shown in FIG. 8, the description will not be repeated here.

In step S41, a plurality of sampling regions are specified for the pixel of interest. Then, a sampling feature amount is determined for each of the plurality of sampling regions (step S42). Then, one or more sampling regions whose sampling feature amount satisfies the first condition are determined as peripheral regions (step S43).

Since the inspection apparatus 200 in the second modification determines the surrounding area from among the first sampling area 521 to the fourth sampling area 524, if there is a character, background, or the like around the pixel of interest, the character and the sampling area including the background is not determined as the surrounding area. Therefore, the influence of characters and background can be excluded from the threshold value.

<Third Modification>
FIG. 12 is a diagram showing an example of the peripheral area in the third modified example. Referring to FIG. 12, a peripheral area 531 in the third modification is an area that is at least the first distance from the pixel of interest 501 and is at least the second distance from the pixel of interest 501 . The first distance is the size of N/2 pixels and the second distance is the distance of M/2 pixels. In the inspection apparatus 200 according to the third modification, it is possible to improve the accuracy of detecting a noise region whose diameter is equal to or less than the first distance. A plurality of peripheral regions having different first distances may be used in the third modification. Also, the peripheral area shown in FIG. 4 and the peripheral area shown in FIG. 12 may be used.

<Fourth Modification>
FIG. 13 is a diagram showing an example of the peripheral area in the fourth modified example. Referring to FIG. 13 , peripheral region 541 in the fourth modification is a region that is at least a first distance away from target pixel 501 in the Y-axis direction and is at least a second distance away from target pixel 501 in the Y-axis direction. is the area of The first distance is the size of N/2 pixels and the second distance is the distance of M/2 pixels. As a result, it is possible to detect a noise area of streaks parallel to the X-axis.

Note that the peripheral area 541 in the fourth modification may be an area that is at least the first distance away from the pixel of interest 501 in the X-axis direction, and may be an area that is at least the second distance away from the pixel of interest 501 in the X-axis direction. . As a result, it is possible to detect a noise region of streaks parallel to the Y-axis.

A plurality of peripheral regions having different first distances may be used in the fourth modification. Further, noise may be detected by combining one or more of the peripheral areas shown in FIG. 4, the peripheral areas shown in FIG. 12, and the peripheral areas shown in FIG.

<Fifth Modification>
Inspection apparatus 200 in the present embodiment may be incorporated in MFP 100 functioning as an image forming apparatus. In this case, a transport path may be provided for transporting a sheet on which an image is formed by MFP 100 to a document reading unit provided in MFP 100 so that the document reading unit can read the image formed on the sheet.

Moreover, the inspection device 200 may be a scanner device as long as it has a function of reading a document.

As described above, the inspection apparatus 200 according to the present embodiment processes each of a plurality of pixels forming the processing target data as a pixel of interest. Specifically, the inspection apparatus 200 determines a feature amount based on the pixel values of a plurality of pixels included in a peripheral area defined by a predetermined relative position to the target pixel, and determines a threshold value for the target pixel based on the feature amount. is determined and the pixel value of the pixel of interest is compared with the threshold value. Therefore, it is possible to detect whether the pixel of interest has a value different from that of the pixels in the surrounding area, thereby improving the detection accuracy.

In addition, the inspection apparatus determines a sub-feature amount indicating a block feature for each of a plurality of blocks included in the peripheral area, and based on the sub-feature amount determined for each of the plurality of blocks, selects one of the plurality of blocks. The above candidate blocks are determined, and the feature amount of the surrounding area is determined based on the pixel values of a plurality of pixels included in each of the one or more candidate blocks. Therefore, the feature amount is determined by excluding pixels whose pixel values are different from those of other pixels among the plurality of pixels included in the peripheral region, so that the accuracy of the threshold value can be improved.

Also, the feature amount is the standard deviation or the standard deviation and average value of the pixel values of a plurality of pixels. Therefore, it is possible to determine whether or not the pixel of interest has a peculiar value in comparison with a plurality of pixels included in the surrounding area.

The inspection apparatus 200 also includes a first scanner 206 that reads a sheet on which an image is formed by the MFP 100 functioning as an image forming apparatus, outputs image data, and generates processing target data from the image data. Therefore, defects in image formation by the MFP 100 can be detected.

Also, inspection apparatus 200 acquires original data that is the basis of an image formed on paper by MFP 100, determines a characteristic region of the original data, and determines a characteristic region of the original data based on a plurality of pixels included in a region other than the characteristic region in the peripheral region. to determine the feature value of the surrounding area. Therefore, since the feature area included in the original data is excluded from the peripheral area, it is possible to improve the accuracy of the threshold value.

In addition, since the inspection apparatus 200 generates processing target data by changing the resolution of the image data output by the first scanner 206, high-frequency noise included in the image data can be removed.

Also, when the pixel of interest is outside the range determined by the threshold value, the inspection apparatus 200 determines the pixel of interest as a noise region where noise appears. Therefore, a noise area having the same size as the pixel of interest can be detected.

In addition, the inspection apparatus 200 determines the target pixel as a candidate pixel when the target pixel is out of the range determined by the threshold value, and sets an area formed by a set of candidate pixels in which a predetermined number or more are consecutive as a noise area where noise appears. decide. Therefore, it is possible to detect a noise area that is larger in size than the pixel of interest.

In addition, the inspection apparatus 200 sets the pixel value of a plurality of pixels included in the image data output by the first scanner 206 to the average value of the pixel values of a plurality of pixels in a predetermined range including the pixel as the processing target data. Generate as Therefore, high-frequency noise contained in the image data can be removed.

The inspection apparatus 200 in the second modification determines a sampling feature value based on the pixel values of the pixels included in each of the plurality of sampling regions each arranged at a predetermined relative position to the pixel of interest. Then, one or more of the plurality of sampling regions are determined as peripheral regions based on the sampling feature amount determined for each of the plurality of sampling regions. For this reason, the sampling areas containing characters or the like can be excluded from among the plurality of sampling areas, so that the precision of the threshold value can be improved.

It should be considered that the embodiments disclosed this time are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the above description, and is intended to include all modifications within the meaning and range of equivalents of the scope of the claims.

<Appendix>
(1) Preferably, the characteristic region determining means determines, as the characteristic region, a region including edge pixels forming an edge from the original data. According to this aspect, the threshold value can be made unaffected by edges present in the original data.
(2) Preferably, the characteristic region determining means determines, from the original data, a region including background pixels forming a background as the characteristic region. According to this aspect, the threshold can be made immune to the background present in the original data.
(3) Preferably, the first distance is determined based on the size of a detection target present in the processing target data. According to this aspect, the noise to be detected can be accurately detected.

100 MFP, 200 inspection device, 300 post-processing device, 201 CPU, 202 ROM, 203 RAM, 204 EPROM, 205 communication unit, 206 first scanner, 207 second scanner, 210 serial interface, 211 USB memory, 51 reading control unit , 53 processing target data generation unit, 55 noise detection unit, 57 original data acquisition unit, 59 characteristic region determination unit, 61 resolution conversion unit, 63 averaging unit, 65 edge detection unit, 67 background detection unit, 71 target pixel selection unit , 73 feature amount determination unit, 75 threshold value determination unit, 77 comparison unit, 79 candidate pixel determination unit, 81 noise region determination unit, 85 sub-feature amount determination unit, 87 candidate block determination unit, 89 exclusion unit, 91 sampling feature Quantity determining unit 92 Peripheral area determining unit 501 Pixel of interest 511 Peripheral area 521 to 524 First to fourth sampling areas 531 Peripheral area 541 Peripheral area.

Claims (17)

noise detection means for detecting noise from the data to be processed by processing each of the plurality of pixels constituting the data to be processed in which the plurality of pixels are arranged two-dimensionally as a pixel of interest;
The noise detection means includes a feature amount determination means for determining a feature amount indicating a feature of the surrounding area based on pixel values of a plurality of the pixels included in the surrounding area determined by a predetermined relative position to the target pixel;
threshold determination means for determining a threshold for the pixel of interest based on the feature quantity;
comparison means for comparing the determined threshold value with the pixel value of the pixel of interest;
candidate determination means for determining the target pixel as a candidate pixel when the pixel value of the target pixel is outside the range determined by the threshold value determined for the target pixel;
an inspection apparatus comprising : a noise region determination unit that determines a region formed by a set of said candidate pixels in which a predetermined number or more of consecutive pixels is a noise region in which noise appears. sub-feature determining means for determining a sub-feature representing a feature of each of a plurality of blocks included in the peripheral area;
further comprising candidate block determination means for determining one or more candidate blocks from among the plurality of blocks based on the sub-feature amounts determined for each of the plurality of blocks;
2. The inspection apparatus according to claim 1, wherein said feature quantity determination means determines the feature quantity of said peripheral region based on pixel values of said plurality of pixels included in each of said one or more candidate blocks. Sampling feature for determining a sampling feature value indicating a feature of the sampling region based on pixel values of the plurality of pixels included in the sampling region for each of a plurality of sampling regions each determined by a predetermined relative position to the target pixel. a quantity determining means;
3. The apparatus according to claim 1, further comprising peripheral area determining means for determining one or more of said plurality of sampling areas as said peripheral area based on said sampling feature amount determined for each of said plurality of sampling areas. Inspection equipment as described. 4. The inspection apparatus according to claim 3, wherein said sampling feature amount is a standard deviation or a standard deviation and an average value of pixel values of said plurality of pixels. reading means for reading a recording medium on which an image is formed by the image forming apparatus and outputting image data;
5. The inspection apparatus according to any one of claims 1 to 4, further comprising processing target data generating means for generating said processing target data from said image data. an original data acquisition unit for acquiring original data that is the basis of the image formed on the recording medium by the image forming apparatus;
further comprising feature area determining means for determining a feature area in which a predetermined feature is expressed in the original data,
6. The inspection apparatus according to claim 5, wherein said feature quantity determining means determines said feature quantity of said peripheral region based on said plurality of pixels included in a region other than said feature region in said peripheral region. 7. The inspection apparatus according to claim 5, wherein said processing target data generating means generates said processing target data by changing the resolution of said image data output by said reading means. noise detection means for detecting noise from the data to be processed by processing each of the plurality of pixels constituting the data to be processed in which a plurality of pixels are arranged two-dimensionally as a pixel of interest;
reading means for reading a recording medium on which an image is formed by the image forming apparatus and outputting image data;
a processing target data generation means for generating the processing target data from the image data;
The processing target data generating means generates the processing target data by changing the resolution of the image data output by the reading means,
The noise detection means includes a feature amount determination means for determining a feature amount indicating a feature of the surrounding area based on pixel values of a plurality of the pixels included in the surrounding area determined by a predetermined relative position to the target pixel;
threshold determination means for determining a threshold for the pixel of interest based on the feature quantity;
a comparing means for comparing the determined threshold value with the pixel value of the pixel of interest;
The noise detection means determines the pixel of interest to be a noise region in which noise appears when the pixel value of the pixel of interest is outside the range determined by the threshold value determined for the pixel of interest. The processing target data generation means sets an average value of the pixel values of the plurality of pixels in a predetermined range including the pixel in place of the pixel values of the plurality of pixels included in the image data output by the reading means. The inspection apparatus according to any one of claims 6 to 8, wherein data is generated as said processing target data. The noise detection means determines the target pixel as a candidate pixel when the pixel value of the target pixel is out of the range determined by the threshold value determined for the target pixel;
9. The inspection apparatus according to claim 8 , further comprising noise region determining means for determining, as a noise region in which noise appears, a region formed by a set of said candidate pixels in which a predetermined number or more are consecutive. 11. The inspection apparatus according to any one of claims 1 to 10, wherein said peripheral area is an area around said pixel of interest excluding said pixel of interest. 11. The inspection apparatus according to any one of claims 1 to 10, wherein said peripheral area is an area at least a first distance away from said pixel of interest. 13. The inspection apparatus according to claim 1, wherein said peripheral area is an area within a second distance from said pixel of interest. 13. The inspection apparatus according to claim 1, wherein said peripheral area is an area within a second distance in a predetermined direction from said pixel of interest. The peripheral region includes a first region within a second distance in a first direction from the pixel of interest, and a second region within the second distance in a second direction intersecting the first direction from the pixel of interest. The inspection device according to any one of claims 1 to 12. An inspection method performed by an inspection device,
A noise detection step of detecting noise from the data to be processed by processing each of the plurality of pixels constituting the data to be processed in which a plurality of pixels are arranged two-dimensionally as a pixel of interest,
The noise detection step includes a feature amount determination step of determining a feature amount indicating a feature of the surrounding area based on pixel values of a plurality of the pixels included in the surrounding area determined by a predetermined relative position to the target pixel;
a threshold determination step of determining a threshold for the pixel of interest based on the feature quantity;
a comparing step of comparing the determined threshold value with the pixel value of the pixel of interest;
a candidate determination step of determining the target pixel as a candidate pixel when the pixel value of the target pixel is outside the range determined by the threshold value determined for the target pixel;
and a noise region determination step of determining a region formed by a set of said candidate pixels in which a predetermined number or more of said pixels are continuous as a noise region in which noise appears. causing a computer to execute a noise detection step of detecting noise from the data to be processed by processing each of the plurality of pixels constituting the data to be processed in which the plurality of pixels are arranged two-dimensionally as a pixel of interest;
The noise detection step includes a feature amount determination step of determining a feature amount indicating a feature of the surrounding area based on pixel values of a plurality of the pixels included in the surrounding area determined by a predetermined relative position to the target pixel;
a threshold determination step of determining a threshold for the pixel of interest based on the feature quantity;
a comparing step of comparing the determined threshold value with the pixel value of the pixel of interest;
a candidate determination step of determining the target pixel as a candidate pixel when the pixel value of the target pixel is outside the range determined by the threshold value determined for the target pixel;
and a noise region determination step of determining a region formed by a set of said candidate pixels in which a predetermined number or more of consecutive pixels are a noise region in which noise appears.

Images