JP2023039140A - Image inspection device, image inspection method and program - Google Patents

Abstract

To provide an image inspection device which can accurately detect a stripe-shaped defect while suppressing erroneous detection.SOLUTION: A detection unit 433 includes a calculation part 435, a setting part 436 and a determination part 437. The calculation part 435 calculates a first feature amount indicating a feature of a stripe-shaped defect for each pixel of a read image. The setting part 436 calculates the first feature amount and the second feature amount for each pixel of an output object image and sets an allowable range that includes the calculated first feature amount and has the width according to the calculated second feature amount. The determination part 437 determines whether or not the stripe-shaped defect exists in the read image by comparing the first feature amount of each pixel of the read image with the allowable range set for each pixel of the output object image. The absolute value of the second feature amount is larger as a change amount of the concentration along the direction orthogonal to the stripe-shaped defect is larger.SELECTED DRAWING: Figure 3

Description

The present disclosure relates to an image inspection apparatus, an image inspection method, and a program.

2. Description of the Related Art Conventionally, there have been cases where a streak-like defect occurs on a sheet of paper on which an image has been formed by an image forming apparatus. Therefore, an image inspection apparatus has been developed for inspecting a sheet output from an image forming apparatus. Japanese Patent Application Laid-Open No. 2017-173000 (Patent Document 1) discloses that a streak-like defect on a read image is detected by using a reference image serving as an inspection standard and a read image generated by reading a sheet using a reading device. An inspection device for inspection is disclosed. This inspection apparatus calculates a difference value between each pixel of the first difference image indicating the difference between the reference image and the read image and pixels separated by a predetermined number in a direction perpendicular to the possible streak, and calculates the calculated difference value. is used to determine whether or not the read image has a streak-like defect. As the reference image, an image to be output, which is the original of the image formed on the paper by the image forming apparatus, is used.

Japanese Patent Application Laid-Open No. 2017-173000

The read image is an RGB image in which the value of each pixel indicates the density of RGB. On the other hand, since the output target image is data for printing, it is a CMYK image in which the value of each pixel indicates the density of CMYK. Therefore, in order to compare the read image and the output target image, it is necessary to color-convert the output target image into an RGB image. This color conversion is generally performed by processing using a lookup table or the like that takes into consideration the characteristics of the image forming apparatus and reading device. However, errors can occur in the image formation by the image forming apparatus and the reading by the reading device when generating the read image. Due to this error, erroneous detection may occur in which it is determined that a streak-like defect exists even though there is no streak-like defect.

The present disclosure has been made to solve the problems described above, and the object thereof is to provide an image inspection apparatus, an image inspection method, and a program capable of suppressing false detection and accurately detecting streak-like defects. to provide.

According to one aspect, the image inspection device inspects a sheet on which an image has been formed by the image forming device. The image inspection apparatus includes: a first acquisition unit that acquires a read image generated by reading a sheet; a second acquisition unit that acquires an output target image that is the source of the image formed on the sheet by the image forming apparatus; a detection unit that detects a streak-like defect on the read image using the image and the output target image. The detection unit includes a calculation unit, a setting unit, and a determination unit. The calculation unit calculates a first feature amount indicating the feature of the streak-like defect for each pixel of the read image. The setting unit calculates a first feature amount and a second feature amount for each pixel of the output target image, and sets a width including the calculated first feature amount and according to the calculated second feature amount. Set the tolerance that you have. The determination unit determines whether or not a streak-like defect exists in the read image by comparing the first feature amount of each pixel of the read image with the allowable range set for each pixel of the output target image. The absolute value of the second feature amount increases as the amount of change in density along the direction orthogonal to the streak defect increases.

According to another aspect, an image inspection method inspects a sheet output from an image forming apparatus. The image inspection method includes the steps of acquiring a read image by reading a sheet, acquiring an output target image that is the original of the image formed on the sheet by the image forming apparatus, and using the read image and the output target image. and detecting streak defects on the paper. The step of detecting includes a step of calculating a first feature quantity indicating the characteristics of the streak defect for each pixel of the read image, and a step of calculating the first feature quantity for each pixel of the output target image. Further, the step of detecting includes: calculating a second feature amount for each pixel of the output target image; It includes setting a tolerance range with a width depending on the amount. Furthermore, in the detecting step, whether or not a streak-like defect exists in the read image is determined by comparing the first feature value of each pixel of the read image with an allowable range set for each pixel of the output target image. including the step of determining. The absolute value of the second feature amount increases as the amount of change in density along the direction orthogonal to the streak defect increases.

According to the present disclosure, erroneous detection is suppressed, and streak-like defects are accurately detected.

1 is a diagram showing a schematic configuration of an image inspection system including an image inspection apparatus according to an embodiment of the present disclosure; FIG. 2 is a block diagram showing a configuration example of a control system of the image forming apparatus; FIG. 1 is a block diagram showing a configuration example of a control system of an image inspection apparatus according to an embodiment; FIG. It is a figure which shows an example of a secondary differential filter. It is a figure which shows an example of a 1st-order differential filter. FIG. 10 is a diagram showing the relationship between changes in pixel values in a gradation region of an output target image, first feature amounts, and second feature amounts; 4 is a flow chart showing an example of the flow of an image inspection method according to the embodiment; FIG. 11 is a block diagram showing a configuration example of a control system of an image inspection apparatus according to modification 2;

An image inspection apparatus according to an embodiment according to the present disclosure will be described below with reference to the drawings. In the following description, identical parts and components are given identical reference numerals. Their names and functions are also the same. Therefore, detailed description of these will not be repeated. Note that the embodiments and modifications described below may be selectively combined as appropriate.

<Image forming system>
FIG. 1 is a diagram showing a schematic configuration of an image inspection system including an image inspection apparatus according to an embodiment of the present disclosure. As shown in FIG. 1 , the image inspection system 1 includes an image forming device 2 and an image inspection device 3 .

The image forming apparatus 2 is an example of an image forming apparatus that forms an image on a sheet by an electrophotographic method that forms an image using static electricity. The image forming apparatus 2 forms a color image on a sheet of paper in a tandem format in which toner images of four colors of yellow (Y), magenta (M), cyan (C), and black (K) are superimposed. A PC (Personal Computer) or the like operated by an operator is connected to the image forming apparatus 2 via a LAN (Local Area Network) (not shown). A job is input from the PC to the image forming apparatus 2 via the LAN. The image forming apparatus 2 performs various processing such as image forming processing according to the input job.

The image forming apparatus 2 includes an operation display section 23 , an image input section 24 , a paper feeding section 28 , a conveying section 25 and an image forming section 26 .

The operation display unit 23 includes a display unit such as a liquid crystal panel and an operation unit such as a touch sensor. The display unit and the operation unit are integrally formed as a touch panel, for example. The operation display unit 23 generates an operation signal representing the details of the operation input by the operator to the operation unit, and supplies the operation signal to the control unit 20 (see FIG. 2 described later). Further, the operation display unit 23 displays the details of operations performed by the operator, setting information, etc. on the display unit based on display signals supplied from the control unit 20 . It should be noted that it is also possible to configure the operation unit with a mouse, tablet, or the like, and to configure it separately from the display unit.

The image input unit 24 optically reads an image from a document on a document platen and A/D converts the read image to generate image data. The image input unit 24 can also read an image from a document on the platen glass.

The paper feed unit 28 includes a paper feed mechanism including a plurality of paper feed trays, a paper feed roller provided for each paper feed tray, a separation roller, a paper feed/separation rubber, a delivery roller, and the like. Each paper feed tray stores paper that is pre-identified for each paper type (paper type, basis weight, paper size, etc.). transported towards

The transport unit 25 includes a plurality of transport rollers for transporting paper, and transports the paper fed from the paper feed unit 28 to the image inspection device 3 via the image forming unit 26 .

The image forming unit 26 performs an electrophotographic image forming process to form an image of four colors of C, M, Y and K on a sheet. The image forming section 26 includes four image forming units 263 for forming Y, M, C and K toner images, respectively, an intermediate transfer belt 264 , a transfer section 265 and a fixing device 266 . In this embodiment, the image forming unit 26 to which an electrophotographic method is applied will be described, but the image forming unit 26 is not limited to the example to which the electrophotographic method is applied. An image forming section may be used.

The four image forming units 263 are arranged in series (tandem) along the belt surface of the intermediate transfer belt 264 and form C, M, Y and K images respectively. The four image forming units 263 differ only in the colors of the images they form, and have the same configuration. As shown in FIG. 1, each of the four image forming units 263 has an exposure section 263a, a photosensitive drum 263b, a developing section 263c, a charging section 263d, a cleaning section 263e, and a primary transfer roller 263f.

Each image forming unit 263 charges the photoreceptor drum 263b by the charging unit 263d, and then irradiates the photoreceptor drum 263b with light from the exposure unit 263a based on the input job. to form an electrostatic latent image. Next, toner is supplied onto the photosensitive drum 263b by the developing section 263c, and a toner image is formed on the photosensitive drum 263b. The toner image formed on the photosensitive drum 263b is (primarily transferred) onto the intermediate transfer belt 264 by the primary transfer roller 263f. As a result, an image of each color is formed on the intermediate transfer belt 264 . After the primary transfer, the toner remaining on the photosensitive drum 263b is removed by the cleaning section 263e.

The transfer unit 265 transfers (secondary transfer) the toner image on the intermediate transfer belt 264 onto the paper.

The fixing device 266 heats and presses the paper onto which the toner image has been transferred, thereby performing a fixing process. The sheet on which the toner image is fixed by the fixing device 266 is conveyed to the image inspection device 3 by the conveying section 25 .

The image inspection device 3 includes a reading unit 30 and a detection device 4 . The reading unit 30 includes, for example, a linear image sensor (such as a CCD line sensor), an optical system, a light source, and the like. The reading unit 30 reads the paper on which the image is formed to generate analog image data, and outputs the generated analog image data to the detection device 4 . The detection device 4 detects streak-like defects on the paper based on the analog image data input from the reading unit 30 .

<Configuration of Image Forming Apparatus>
FIG. 2 is a block diagram showing a configuration example of the control system of the image forming apparatus. As shown in FIG. 2, the image forming apparatus 2 includes an operation display unit 23, an image input unit 24, a paper feeding unit 28, a conveying unit 25, and an image forming unit 26, as well as a control unit 20 and a communication I. A /F unit 21 and a storage unit 22 are provided.

The communication I/F unit 21 is an interface that transmits and receives data to and from the PC 6, which is a terminal operated by an operator, via a network or a dedicated line. A NIC (Network Interface Card), for example, is used as the communication I/F unit 21 .

The control unit 20 includes a CPU (Central Processing Unit) 201 , a ROM (Read Only Memory) 202 , a RAM (Random Access Memory) 203 and an input image processing unit 204 .

The input image processing unit 204 performs predetermined image processing (for example, rasterization processing) on the image included in the job input from the PC 6 via the communication I/F unit 21, and the value of each pixel indicates the CMYK density. Generate image data for The input image processing unit 204 also performs image processing on the image data obtained from the document read by the image input unit 24 to create image data for printing. The image data for printing is sent to the image forming section 26 and the image inspection device 3 .

The ROM 202 stores programs executed by the CPU 201, data used when the programs are executed, and the like. CPU 201 controls each unit constituting image forming apparatus 2 by reading a program stored in ROM 202 . The RAM 203 is temporarily written with variables, parameters, and the like generated during the arithmetic processing of the CPU 201 .

For example, the CPU 201 controls the transport unit 25 to drive the transport rollers and transport the paper. The CPU 201 outputs the output target image created by the input image processing unit 204 to the image forming unit 26 and controls the image forming unit 26 to form an image on paper.

Further, the CPU 201 outputs a display signal to the operation display section 23 to display various setting screens, various processing results, and the like on the operation display section 23 . The information displayed on the operation display unit 23 also includes streak detection results output from the image inspection apparatus 3 .

The storage unit 22 stores parameters used when the CPU 201 executes programs, data obtained by executing the programs, and the like. For example, the storage unit 22 stores information such as image forming conditions for each density level. A program executed by the CPU 201 may be stored in the storage unit 22 .

<Configuration of image inspection apparatus>
FIG. 3 is a block diagram showing a configuration example of the control system of the image inspection apparatus according to the embodiment. As shown in FIG. 3 , the image inspection device 3 includes a communication I/F section 32 and a paper transport section 33 in addition to the reading section 30 and the detection device 4 described above. The detection device 4 includes a control section 41 and a storage section 42 . The storage unit 42 is configured by, for example, a large-capacity HDD (Hard Disk Drive) or the like.

The communication I/F unit 32 is an interface that transmits and receives data to and from the image forming apparatus 2 via a network. A NIC, for example, is used as the communication I/F unit 32 .

The paper transport unit 33 drives a transport roller (not shown) under the control of the control unit 41 to transport the paper output from the image forming apparatus 2 .

The control unit 41 of the detection device 4 includes a CPU 411 , a ROM 412 and a RAM 413 .

The ROM 412 stores a program 440 executed by the CPU 411 and data used when the program 440 is executed. The ROM 412 is used as an example of a computer-readable non-transitory recording medium storing the program 440 executed by the CPU 411 .

The CPU 411 controls each part of the image inspection apparatus 3 by reading out the program 440 stored in the ROM 412 . An image processing unit 431 , a color conversion unit 432 , and a detection unit 433 are implemented by the CPU 411 executing the program 440 read from the ROM 412 .

The RAM 413 is temporarily written with variables, parameters, etc. generated during the arithmetic processing of the CPU 411 .

The image processing unit 431 performs various kinds of processing such as analog processing, A/D conversion processing, and shading processing on the analog image data received from the reading unit 30, and then converts the value of each pixel into a digital image data indicating the density of RGB. Image data (hereinafter referred to as "read image 420") is generated. That is, the image processing unit 431 operates as a first acquisition unit that acquires the read image 420 generated by reading the paper output from the image forming apparatus 2 . A read image 420 generated by the image processing unit 431 is temporarily stored in the RAM 413 .

The color conversion unit 432 converts printing image data (CMYK image) received from the image forming apparatus 2 via the communication I/F unit 32 into digital image data (hereinafter referred to as “output (referred to as "target image 421"). That is, the color conversion unit 432 operates as a second acquisition unit that acquires the output target image 421 that is the source of the image that the image forming apparatus 2 forms on paper. The output target image 421 generated by the color conversion unit 432 is temporarily stored in the RAM 413 .

The detection unit 433 detects a streak-like defect on the read image 420 using the read image 420 and the output target image 421 . The detection unit 433 stores the streak detection result 450 in the storage unit 42 . Note that the streak detection result 450 may be written not only in the storage unit 42 but also in the external storage device 5 connected to the image inspection apparatus 3 . The storage device 5 is, for example, a USB (Universal Serial Bus) memory, SSD (Solid State Drive), HDD, or the like connected to the image inspection apparatus 3 . By sending the streak detection result 450 to the storage device 5, the operator can display the streak detection result 450 saved in the storage device 5 and check the contents. The streak detection result 450 may be transferred to a cloud server (not shown) or the PC 6 (see FIG. 2) connected via the communication I/F unit 32 and stored.

<Structure of detector>
As shown in FIG. 3 , the detection unit 433 has a coordinate conversion unit 434 , a calculation unit 435 , a setting unit 436 and a determination unit 437 . The operation of each unit will be described below with reference to FIGS. 3 to 6. FIG.

Coordinate conversion unit 434 aligns read image 420 and output target image 421 . The image forming unit 26 of the image forming apparatus 2 misaligns the paper when forming an image on the paper, and the misalignment of the paper when the reading unit 30 of the image inspection device 3 reads the image on the paper. Therefore, positional deviation may occur between the read image 420 and the output target image 421 . The coordinate conversion unit 434 aligns the output target image 421 with the position of the read image 420 in order to correct such positional deviation. Specifically, the coordinate transformation unit 434 generates a coordinate transformation matrix for correcting positional deviation, and applies the generated coordinate transformation matrix to the output target image 421 .

The calculation unit 435 calculates a first feature amount indicating the feature of the streak defect for each pixel of the read image 420 .

A streak-like defect is caused by a scratch or dirt on a photosensitive drum, a roller, or the like provided in the image forming apparatus 2 . Therefore, the streak defect is parallel to the sheet conveying direction in the image forming apparatus 2 . Pixels with streak defects have density values (luminance values) that are not intended by the operator. Therefore, a streak-like defect appears as a gradation difference or density unevenness. For example, streak defects include white streaks that are lighter than the original image, black streaks that are darker than the original image, and the like. Therefore, the calculation unit 435 indicates the change in the density in the direction perpendicular to the sheet conveying direction in the image forming apparatus 2, and the first Calculate the feature amount. Specifically, the calculator 435 calculates the first feature amount using a secondary differential filter typified by a Laplacian filter.

FIG. 4 is a diagram showing an example of a secondary differential filter. The second-order differential filter shown in FIG. 4 generates a second-order differential value of a pixel value with respect to a pixel position along a direction D perpendicular to the sheet conveying direction in the image forming apparatus 2 . Specifically, the secondary differential filter converts the value A of the pixel of interest 7a and the values B and C of pixels 7b and 7c apart from the pixel of interest 7a by a predetermined number of pixels along the direction D into the following equations: Generate the value A' by substituting in (1).
A' = A x 2 + B x (-1) + C x (-1) Equation (1)
The predetermined number of pixels is predetermined according to the width of the detected streak-like defect, and is 3 in the example shown in FIG. The value A' generated using the secondary differential filter indicates the amount of change in the values of a plurality of pixels along the direction D centered on the pixel of interest 7a. By applying the secondary differential filter shown in FIG. 4 to the read image 420, an image is generated in which the value of each pixel is A'.

In a region with a uniform density, the difference between the value of the target pixel 7a and the values of pixels 7b and 7c separated by a predetermined number of pixels along the direction D is zero. Therefore, when the secondary differential filter is applied to a plurality of pixels centering on the target pixel 7a, the value of A' becomes 0.

On the other hand, when a streak-like defect exists in a part of a region having a uniform density, the value of the target pixel 7a included in the streak-like defect is the value of the pixel 7b, It differs greatly from the value of 7c. Therefore, when the secondary differential filter is applied to a plurality of pixels centering on the pixel of interest 7a, the absolute value of A' increases. For example, in a pixel included in a black streak, A' takes a positive value. A' takes a negative value in pixels included in the white streak.

In this way, the first feature amount, which is the difference between the values of the target pixel 7a and the pixels 7b and 7c that are separated from the target pixel 7a by a predetermined number of pixels along the direction D, represents the feature of the streak-like defect. However, the first feature value also responds to a portion of the read image where streak-like content exists. Striped content is not a defect and should not be detected as a defect. Therefore, it is desirable to distinguish between streak content and streak defects.

The setting unit 436 dynamically sets a threshold value for distinguishing a streak-like content from a streak-like defect for each pixel according to the output target image 421 . Specifically, the setting unit 436 sets the upper limit value and the lower limit value of the allowable range, which is the range of the first feature quantity that can be taken according to the content.

The determination unit 437 determines whether or not a streak-like defect exists in the read image 420 by comparing the first feature amount of each pixel of the read image 420 with the allowable range set for each pixel of the output target image. judge. For example, the determination unit 437 determines that a streak defect exists in the read image 420 when a predetermined number of pixels whose first feature values are outside the allowable range continue along the paper transport direction.

The details of how to set the allowable range will be described below. The setting unit 436 calculates the first feature amount for each pixel of the output target image 421 . The output target image 421 does not include a streak-like defect. Therefore, the first feature amount calculated from the output target image 421 indicates a value corresponding to the content. Therefore, the setting unit 436 sets an allowable range including the first feature amount calculated from the output target image 421 for each pixel. For example, the setting unit 436 sets an allowable range in which the first feature amount calculated from the output target image 421 is the median value. The two first feature amounts calculated from the read image 420 and the output target image 421 for the pixels included in the streaky content are the same or similar. Therefore, the first feature amount of the pixels included in the striped content in the read image 420 falls within the allowable range. As a result, by comparing the first feature amount calculated from the read image 420 and the permissible range, detection of streak-like content as a streak-like defect can be suppressed.

The output target image 421 is generated by color conversion by the color conversion unit 432 . When the image forming apparatus 2 normally performs image formation, it is preferable that the read image 420 and the output target image 421 have the same density of each pixel. Therefore, the color conversion unit 432 converts the print image data (CMYK image) into the output target image 421 (RGB image) using a lookup table or the like that is predetermined in consideration of the characteristics of the image forming apparatus 2 and the reading unit 30. ). However, errors may occur in image formation by the image forming apparatus 2 and reading by the reading unit 30 . In particular, errors in image formation and reading tend to increase in gradation areas where the density changes continuously. As the error increases, the possibility of erroneous detection that a streak-like defect exists even though there is no streak-like defect increases. Therefore, the setting unit 436 increases the width of the allowable range for pixels that tend to have large errors in image formation and reading.

The setting unit 436 calculates the second feature amount in addition to the first feature amount for each pixel of the output target image 421 . The absolute value of the second feature amount increases as the amount of change in density along the direction D perpendicular to the streak defect increases. In other words, the absolute value of the second feature amount is greater in the gradation area where the density continuously changes than in the non-gradation area where the density does not change. The setting unit 436 calculates the second feature amount using, for example, a primary differential filter.

FIG. 5 is a diagram showing an example of a primary differential filter. The first derivative filter shown in FIG. 5 produces the first derivative of pixel values for pixel locations along direction D. FIG. Specifically, the first-order differential filter substitutes values B and C of pixels 7b and 7c, which are separated from the pixel of interest 7a by a predetermined number of pixels along the direction D, into the following equation (2). A value A″ is generated for the pixel of interest 7a.
A″=B×(+1)+C×(-1) Equation (2)
The predetermined number of pixels is predetermined according to the width of the detected streak-like defect, and is 3 in the example shown in FIG.

The setting unit 436 sets the width of the allowable range of each pixel according to the second feature amount calculated from the output target image 421 .

In this way, the setting unit 436 calculates the first feature amount and the second feature amount for each pixel of the output target image, and includes the calculated first feature amount and the calculated second feature amount. Set a tolerance with a width that depends on the . For example, for each pixel of the output target image, the setting unit 436 sets the lower limit of the allowable range to (A'-k× |A"|), and the upper limit of the allowable range is set to (A'+k×|A"|), where k is a predetermined coefficient. A threshold image 422 is generated that indicates the upper and lower limits of the threshold image 422 is temporarily stored in the RAM 413 .

FIG. 6 is a diagram showing the relationship between changes in pixel values in the gradation area of the output target image, the first feature amount, and the second feature amount. In FIG. 6, a solid line 60 indicates changes in pixel values. A dashed line 61 indicates a change in the second feature amount. A dashed-dotted line 62 indicates a change in the first feature amount. In the example shown in FIG. 6, the values of the pixels in the gradation area of the output target image 421 change (increase) continuously along the direction D monotonically. Therefore, the first feature amount (A′) generated using the second order differential filter is 0 at each pixel, but the second feature amount (A″) generated using the first order differential filter is Since each pixel takes a positive constant value, an allowable range including the first feature amount and having a width corresponding to the second feature amount is set for each pixel, thereby reducing the error in image formation and reading. Even if is larger, the possibility of false detection is reduced.

<Image inspection method>
FIG. 7 is a flow chart showing an example of the flow of the image inspection method according to the embodiment. As shown in FIG. 7, the CPU 411 of the image inspection apparatus 3 executes step S1 of acquiring the read image 420 and step S2 of acquiring the output target image 421 in parallel. Specifically, in step S<b>1 , the CPU 411 generates the read image 420 by executing various processes on the analog image data received from the reading section 30 . In step S<b>2 , the CPU 411 generates an output target image 421 by color-converting the print image data (CMYK image) received from the image forming apparatus 2 .

After steps S1 and S2, CPU 411 aligns read image 420 and output target image 421 in step S3. Specifically, the CPU 411 aligns the output target image 421 with the position of the read image 420 in order to correct the positional deviation between the read image 420 and the output target image 421 .

After step S3, steps S4 and S5 are performed in parallel. Note that steps S4 and S5 may be executed in order. In step S<b>4 , the CPU 411 calculates a first feature amount for each pixel of the output target image 421 using a secondary differential filter. In step S5, the CPU 411 calculates a second feature amount for each pixel of the output target image 421 using a primary differential filter.

Next, in step S6, the CPU 411 sets, for each pixel, an allowable range including the first feature amount calculated in step S4 and having a width corresponding to the second feature amount calculated in step S5. In step S7, the CPU 411 generates a threshold image 422 indicating the upper and lower limits of the allowable range for each pixel.

On the other hand, in step S8 after step S2, the CPU 411 calculates the first feature amount for each pixel of the read image 420 using the secondary differential filter. In step S9, the CPU 411 compares the first feature amount calculated in step S8 with the allowable range indicated by the threshold image 422 for each pixel. In step S10, the CPU 411 determines whether or not a streak-like defect exists in the read image 420 based on the comparison result in step S9. After step S10, the process ends.

Note that steps S4 to S9 are executed for each of RGB. Then, in step S10, the CPU 411 causes the read image 420 to be streak-like in response to a predetermined number of consecutive pixels in which the first feature value of at least one of RGB is out of the allowable range along the paper transport direction. It is determined that a defect exists. Alternatively, the CPU 411 determines that a streak-like defect exists in the read image 420 in response to a predetermined number of consecutive pixels whose first feature values are outside the allowable range for all of RGB along the paper transport direction. may

<Advantages>
As described above, the image inspection device 3 inspects a sheet on which an image has been formed by the image forming device 2 . The image inspection apparatus 3 includes an image processing unit 431 that acquires a read image 420 generated by reading a sheet, and a color conversion unit that acquires an output target image 421 that is the original of the image formed on the sheet by the image forming apparatus 2. 432 and a detection unit 433 . The detection unit 433 detects a streak-like defect on the read image 420 using the read image 420 and the output target image 421 . The detection unit 433 includes a calculation unit 435 , a setting unit 436 and a determination unit 437 . The calculation unit 435 calculates a first feature amount indicating the feature of the streak defect for each pixel of the read image 420 . The setting unit 436 calculates the first feature amount and the second feature amount for each pixel of the output target image 421, and includes the calculated first feature amount and according to the calculated second feature amount. Sets a tolerance that has a width. The determination unit 437 compares the first feature amount of each pixel of the read image 420 with the allowable range set for each pixel of the output target image 421 to determine whether or not a streak-like defect exists in the read image 420. judge. The absolute value of the second feature amount increases as the amount of change in density along the direction orthogonal to the streak defect increases.

As described above, errors in image formation and reading tend to increase in gradation areas where the density changes continuously. According to the above configuration, the width of the allowable range is set in accordance with the second feature value, which has a larger absolute value as the amount of change in density along the direction orthogonal to the streak defect increases. Therefore, even if errors in image formation and reading become large, the possibility of erroneous detection in which content that is not a streak defect is detected as a streak defect is reduced.

<Modification 1>
The susceptibility to imaging and reading errors in gradation regions is color dependent. For example, depending on the characteristics of the image forming apparatus 2 and the reading unit 30, there are cases where errors are particularly likely to occur with reddish colors. Therefore, the second feature amount for defining the width of the allowable range may be adjusted according to the color of each pixel of the output target image 421 .

For example, the setting unit 436 calculates the second feature amount as follows for the output target image 421 in which each pixel indicates an RGB density value. The setting unit 436 calculates a value obtained by multiplying a value obtained by performing a process using a first-order differential filter on each density value of RGB for each pixel of the output target image 421 by a coefficient as a second feature amount. do. The coefficients are different for each color. When errors are particularly likely to occur in red-based colors, the coefficient corresponding to R is set larger than the coefficients corresponding to G and B. For example, the coefficient corresponding to G and B is 1, and the coefficient corresponding to R is 1.5. This suppresses erroneous detection even for colors that are prone to errors.

The coefficient for each color may be predetermined or may be set according to user input. When set according to user input, the image inspection device 3 provides an input screen for entering coefficients. The user may enter the coefficients in the provided input screen.

<Modification 2>
In the gradation area, the steeper the gradient of the monotonous change of the pixel value with respect to the pixel position, the more the error tends to occur. Since the visual sensitivity of streaks is low at such places, there is no problem even if they are excluded from detection targets for streak-like defects.

FIG. 8 is a block diagram showing a configuration example of a control system of an image inspection apparatus according to Modification 2. As shown in FIG. As shown in FIG. 8, the image inspection apparatus 3A according to Modification 2 is different from the image inspection apparatus 3 shown in FIG. differ in that

The detection region determination unit 438 determines a region to be detected for streak-like defects. The detection region determination unit 438 identifies non-inspection target regions in the output target image 421 in which the second feature amount exceeds the reference value. The detection area determination unit 438 excludes the area corresponding to the target area in the read image 420 from the detection target of the streak-like defect. As a result, non-inspection target areas where detection of streak-like defects is unnecessary are excluded from detection targets of streak-like defects, so that the inspection load of streak-like defects on the CPU 411 is reduced.

The reference value may be predetermined or may be set according to user input. When set according to user input, the image inspection apparatus 3 provides an input screen for inputting the reference value. The user should just input the reference value in the provided input screen.

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 scope and meaning equivalent to the scope of the claims.

1 image inspection system 2 image forming device 3, 3A image inspection device 4 detection device 5 storage device 7a pixel of interest 7b, 7c pixel 20, 41 control unit 21, 32 communication I/F unit 22 , 42 storage unit, 23 operation display unit, 24 image input unit, 25 conveying unit, 26 image forming unit, 28 paper feeding unit, 30 reading unit, 33 paper conveying unit, 201, 411 CPU, 202, 412 ROM, 203, 413 RAM, 204 input image processing section, 263 image forming unit, 263a exposure section, 263b photosensitive drum, 263c developing section, 263d charging section, 263e cleaning section, 263f primary transfer roller, 264 intermediate transfer belt, 265 transfer section, 266 fixing device, 420 read image, 421 output target image, 422 threshold image, 431 image processing unit, 432 color conversion unit, 433 detection unit, 434 coordinate conversion unit, 435 calculation unit, 436 setting unit, 437 determination unit, 438 detection Region determination unit, 440 program.

Claims (6)

An image inspection device for inspecting a sheet on which an image is formed by an image forming device,
a first acquisition unit that acquires a read image generated by reading the paper;
a second acquisition unit that acquires an output target image that is the source of the image that the image forming apparatus forms on the sheet;
a detection unit that detects a streak-like defect on the read image using the read image and the output target image;
The detection unit is
a calculation unit that calculates a first feature quantity indicating a feature of the streak-like defect for each pixel of the read image;
For each pixel of the output target image, the first feature amount and the second feature amount are calculated, and a width including the calculated first feature amount and corresponding to the calculated second feature amount is determined. a setting unit for setting an allowable range of
determining whether or not the streak-like defect exists in the read image by comparing the first feature amount of each pixel of the read image with the allowable range set for each pixel of the output target image; and a determination unit for
The image inspection apparatus, wherein the absolute value of the second feature amount increases as the amount of change in density along the direction orthogonal to the streak-like defect increases. The first feature amount is calculated using a secondary partial filter,
2. The image inspection apparatus according to claim 1, wherein said second feature amount is calculated using a first-order differential filter. each pixel of the read image and each pixel of the output target image indicate density values of a plurality of colors;
The second feature amount is calculated by multiplying a value obtained by processing using the first-order differential filter by a coefficient for each of the plurality of colors,
3. The image inspection apparatus according to claim 2, wherein said coefficient differs for each of said plurality of colors. The detection unit identifies a target region in the output target image in which the second feature value exceeds a reference value, and excludes a region corresponding to the target region in the read image from a detection target for the streak-like defect. 3. The image inspection apparatus according to claim 2, further comprising a detection area determining section. An image inspection method for inspecting a sheet output from an image forming apparatus,
obtaining a read image by reading the paper;
a step of acquiring an output target image that is the basis of the image to be formed on the paper by the image forming apparatus;
detecting streak-like defects on the paper using the read image and the output target image;
The detecting step includes:
calculating, for each pixel of the read image, a first feature quantity indicating a feature of the streak-like defect;
calculating the first feature amount for each pixel of the output target image;
calculating a second feature amount for each pixel of the output target image;
setting an allowable range including the calculated first feature amount and having a width corresponding to the calculated second feature amount for each pixel of the output target image;
determining whether or not the streak-like defect exists in the read image by comparing the first feature amount of each pixel of the read image with the allowable range set for each pixel of the output target image; and
The image inspection method, wherein the absolute value of the second feature amount increases as the amount of change in density along the direction perpendicular to the streak-like defect increases. A program that causes a computer to execute the image inspection method according to claim 5.

Images