JP2022080750A - Image inspection device, image forming apparatus, and image inspection method - Google Patents

Abstract

To provide a method for detecting a defect that an inspection object does not have the characteristics that a non-defective product should have.SOLUTION: An image inspection device creates a first division image by using at least either one of a first processing image obtained by averaging luminance values of pixels in each area of a plurality of areas obtained by dividing an inspection image, or a second processing image obtained by averaging the luminance values of pixels in each area of a plurality of areas obtained by dividing the inspection image, the plurality of areas different in at least either one of a phase, a direction, or a size from the first processing image, and compares the luminance value of the pixels in a certain area and the luminance value of the pixels in a surrounding area with each other to extract a first defect candidate area. The image inspection device creates a second division image by using at least either one of a third processing image obtained by averaging the luminance values of pixels in each area of a plurality of areas obtained by dividing a non-defective image, or a fourth processing image obtained by averaging the luminance values of pixels in each area of a plurality of areas obtained by dividing the non-defective image, the plurality of areas different in at least either one of a phase, a direction, or a size from the third processing image, and extracts a second defect candidate area. The image inspection device compares the first defect candidate area and the second defect candidate area with each other to detect a defective area.SELECTED DRAWING: Figure 12

Description

The present invention relates to an image inspection device, an image forming device, and an image inspection method.

As an alternative to the visual inspection of defects by an operator (for example, inspection of cast or forged parts, inspection of printed matter), a technique of performing image processing on an image of an image to be inspected to detect defects is known.

Further, for the purpose of detecting defects at high speed and with high sensitivity, the captured image is divided into a plurality of regions, and the divided images are generated by averaging the brightness values of the pixels included in each region to reduce the resolution, and the defect candidate region is generated. (For example, see Patent Document 1).

In addition, for the purpose of suppressing erroneous detection of non-defective product features such as markings for identifying objects and shadows derived from shapes as defects, division image generation processing and defect candidate regions are used for a plurality of non-defective product images. A configuration is disclosed in which a region including a feature of a non-defective product in an image obtained by performing an extraction process is not extracted as a defect candidate region (see, for example, Patent Document 2).

However, in the techniques of Patent Documents 1 and 2, it may not be possible to detect a defect that the inspection target does not have the characteristics of a good product as it should be.

An object of the present invention is to detect a defect that the characteristic of a non-defective product is not present in the inspection target.

The image inspection device according to one aspect of the present invention is an image inspection device that inspects a defect of an inspection object, and is divided into an image pickup unit that captures an inspection image of the inspection object and the inspection image into a plurality of regions. In each of the regions, the first processed image obtained by averaging the brightness values of the pixels, or in each region where the inspection image is divided into a plurality of regions whose phase, direction, or size is different from the first processed image. In a certain region among a first divided image generation unit that generates a first divided image using at least one of a second processed image obtained by averaging the brightness values of pixels and a plurality of regions included in the first divided image. A first defect candidate extraction unit that compares the brightness value of a pixel with the brightness value of a pixel in a region surrounding a certain region and extracts a first defect candidate region, and a plurality of non-defective images to be compared with the inspection image. In each region divided into the regions, the third processed image obtained by averaging the brightness values of the pixels, or the good product image was divided into a plurality of regions in which at least one of the phase, direction, or size is different from the third processed image. Of the second divided image generation unit that generates the second divided image using at least one of the fourth processed images obtained by averaging the brightness values of the pixels in each region, and the plurality of regions included in the second divided image. A second defect candidate extraction unit that compares the brightness value of a pixel in a certain region with the brightness value of a pixel in a region surrounding the region to extract a second defect candidate region, the first defect candidate region, and the above. It has a defect detection unit that compares a second defect candidate region and detects a defect region.

According to the present invention, it is possible to detect a defect that the inspection target does not have the characteristics of a good product as it should be.

It is a figure which shows the whole structure example of the image forming apparatus which concerns on embodiment. It is a figure which shows the structural example of the image inspection part which concerns on embodiment. It is a figure of the hardware configuration example of the image forming apparatus which concerns on embodiment. It is a figure of the function configuration example of the processing unit of the image inspection unit which concerns on 1st Embodiment. It is a flow chart of the operation example of the image inspection part which concerns on 1st Embodiment. It is a figure of the processing example by the 1st division image generation part, (a) is the figure of the inspection image example, (b) is the figure of the 1st processing image example, (c) is the figure of the 2nd processing image example. It is a figure of the processing example by the 1st defect candidate extraction unit, (a) is the figure of the 1st defect candidate region example of FIG. 6 (b), (b) is the figure of the 1st defect candidate region example of FIG. 6 (c). It is a figure. It is a figure of the processing example by the 2nd division image generation part, (a) is the figure of the good product image example, (b) is the figure of the 3rd processing image example, (c) is the figure of the 4th processing image example. It is a figure of the processing example by the 2nd defect candidate extraction part, (a) is the figure of the 2nd defect candidate region example of FIG. 8 (b), (b) is the figure of the 2nd defect candidate region example of FIG. 8 (c). It is a figure. It is a figure of the function configuration example of the processing part of the image inspection part which concerns on 2nd Embodiment. It is a flow chart of the operation example of the image inspection part which concerns on 2nd Embodiment. It is a figure which shows the processing example by the processing part which concerns on 2nd Embodiment. It is a figure of the defect detection result example of the image inspection part which concerns on 2nd Embodiment, (a) is the figure of the 1st vote defect area, (b) is the figure of the 2nd vote defect area.

Hereinafter, embodiments for carrying out the invention will be described with reference to the drawings. In each drawing, the same components are designated by the same reference numerals, and duplicate description will be omitted as appropriate.

Further, the embodiments shown below exemplify an image inspection apparatus and an image forming apparatus for embodying the technical idea of the present invention, and the present invention is not limited to the embodiments shown below. Unless otherwise specified, the shapes of the components, their relative arrangements, the values of the parameters, etc. described below are not intended to limit the scope of the present invention to that alone, but are intended to be exemplified. Is. In addition, the size and positional relationship of the members shown in the drawings may be exaggerated in order to clarify the explanation.

The image inspection apparatus according to the embodiment is an apparatus for detecting defects in an inspection object by comparing a non-defective image corresponding to a sample with an inspection image obtained by capturing an image of the inspection object.

In the embodiment, the inspection image of the inspection object is imaged, and in each region where the inspection image is divided into a plurality of regions, the first processed image obtained by averaging the luminance values of the pixels, or at least one of the phase, direction, or size is the first. A first divided image is generated by using at least one of a second processed image in which the brightness values of the pixels are averaged in each region in which the inspection image is divided into a plurality of regions different from the one processed image. Then, among the plurality of regions included in the first divided image, the luminance value of the pixel in a certain region is compared with the luminance value of the pixel in the region around the region, and the first defect candidate region is extracted. As a result, the first defect candidate region in the inspection image is extracted at high speed and with high sensitivity by utilizing the mechanism in which the human unconsciously recognizes the difference from the normal state.

Further, in the embodiment, in each region where the non-defective product image to be compared with the inspection image is divided into a plurality of regions, the third processed image obtained by averaging the luminance values of the pixels, or at least one of the phase, direction or size is the third processed. A second divided image is generated by using at least one of a fourth processed image in which the brightness values of the pixels are averaged in each region in which the non-defective image is divided into a plurality of regions different from the image. Then, among the plurality of regions included in the second divided image, the luminance value of the pixel in a certain region is compared with the luminance value in the region around the region, and the second defect candidate region is extracted. As a result, the second defect candidate region in the non-defective image is extracted at high speed and with high sensitivity by utilizing the mechanism in which the human unconsciously recognizes the difference from the normal state.

Further, in the embodiment, the first defect candidate region and the second defect candidate region are compared, and the defect region is detected. For example, at least one of the first defect region in the first defect candidate region and not in the second defect candidate region or the second defect region in the second defect candidate region and not in the first defect candidate region is detected.

As a result, "positive defects" in the inspection image that should not be in the good product image are detected as the first defect region, and "negative defects" that are missing in the good product image are used as the second defect region. It also makes it possible to detect defects that the inspection target does not have the characteristics of a good product that should be detected.

In the following, the image inspection device according to the embodiment is applied to an image forming apparatus such as a commercial printing machine (production printing machine) that continuously prints a large number of sheets in a short time, and is printed on a recording medium such as paper. A case of detecting a defect in a printed image will be described as an example.

In addition, it is assumed that image formation and printing in the terms of the embodiment are synonymous. Further, in the embodiment, the printed image formed on the recording medium by the image forming apparatus will be described as an example of the inspection target object, and the non-defective product image to be compared with the printed image will be described as an example of the non-defective product.

In terms of embodiments, a printed image means a toner image formed (printed) on a recording medium. On the other hand, the inspection image means an image of electronic data captured by the imaging unit, and the non-defective image also means an image of electronic data.

[Embodiment]
<Overall configuration of image forming apparatus 1>
FIG. 1 is a diagram showing an example of the overall configuration of the image forming apparatus 1 according to the embodiment. As shown in FIG. 1, the image forming apparatus 1 has an image forming unit 100, an image inspection unit 200, and a stacker 300.

The image forming unit 100 includes an operation panel 101, a tandem electrophotographic image forming unit 103Y, 103M, 103C, 103K, a transfer belt 105, a secondary transfer roller 107, a paper feeding unit 109, and a transfer roller. It has a pair 102, a fixing roller 104, and an inversion path 106.

The operation panel 101 is an operation display unit that inputs various operations to the image forming unit 100 and the image inspection unit 200 and displays various screens.

In each of the image forming portions 103Y, 103M, 103C, and 103K, a toner image is formed by performing an image forming process (charging step, exposure step, developing step, transfer step, and cleaning step), and the formed toner image is formed. Is transferred to the transfer belt 105. In the present embodiment, a yellow toner image is formed on the image forming portion 103Y, a magenta toner image is formed on the image forming section 103M, a cyan toner image is formed on the image forming section 103C, and the cyan toner image is formed on the image forming section 103K. A black toner image is formed. However, the present invention is not limited to this, and the order of arrangement of the image forming units 103Y, 103M, 103C, and 103K may be changed as appropriate. Further, the image forming unit 100 may have an image forming unit that forms a toner image of a color other than yellow, magenta, cyan, and black. Colors other than yellow, magenta, cyan and black are white and the like.

The transfer belt 105 conveys a full-color toner image superimposed and transferred by the image forming units 103Y, 103M, 103C, and 103K to the secondary transfer position of the secondary transfer roller 107. In the present embodiment, the yellow toner image is first transferred (primary transfer) to the transfer belt 105, and then the magenda toner image, the cyan toner image, and the black toner image are sequentially superimposed and transferred. However, the present invention is not limited to this, and the order in which the toner images of each color are transferred to the transfer belt 105 may be appropriately changed. Further, in the following, when the colors are not particularly distinguished, the image-creating units 103Y, 103M, 103C, and 103K are referred to as image-creating units 103.

A plurality of recording media are superposed and accommodated in the paper feeding unit 109, and the recording media are fed. Examples of the recording medium include, but are not limited to, recording paper (transfer paper), and any medium capable of forming (recording) an image can be coated paper, thick paper, OHP (Overhead Projector) sheet. , Plastic film, prepreg, copper foil and the like.

The transport roller pair 102 transports the recording medium fed from the paper feed unit 109 on the transport path a in the direction of the arrow s.

The secondary transfer roller 107 collectively transfers (secondary transfer) the full-color toner image conveyed by the transfer belt 105 onto the recording medium conveyed by the transfer roller pair 102 at the secondary transfer position.

The fixing roller 104 fixes the full-color toner image on the recording medium by heating and pressurizing the recording medium on which the full-color toner image is transferred.

In the case of single-sided printing, the image forming unit 100 sends a recording medium on which a full-color toner image is fixed to the image inspection unit 200. On the other hand, in the case of double-sided printing, the image forming unit 100 sends a recording medium on which a full-color toner image is fixed to the inversion path 106.

The inversion path 106 reverses the front and back surfaces of the recording medium by switching back the transmitted recording medium, and conveys the recording medium in the direction of arrow t. The recording medium conveyed by the inversion path 106 is retransmitted by the transfer roller pair 102, the full-color toner image is transferred to the surface opposite to the previous one by the secondary transfer roller 107, and the toner image is fixed by the fixing roller 104. After that, it is sent to the image inspection unit 200 and the stacker 300.

The image inspection unit 200 arranged downstream of the image forming unit 100 includes a reading device 210, a background unit 220, and the like, and an image inspection for inspecting defects in the printed image formed on the recording medium sent from the image forming unit 100. This is an example of a device. The image inspection unit 200 compares the inspection image with the non-defective image and detects defects in the printed image. A non-defective image is an image that serves as a sample of a printed image.

In the embodiment, a non-defective image is created by processing image data which is the original data for forming a printed image. For example, information on print characteristics by the image forming unit 100 is acquired in advance by an experiment or simulation, and the image data is processed by converting the image data using the print characteristic information. Since the created non-defective product image is a non-defective product image created by digital processing, it can also be referred to as a digital non-defective product image or a digital master image.

In addition to the method of processing the image data, it is also possible to read a sample image with the reading device 210 to create a good product image. However, in this method, when the printed image is changed frequently, it is necessary to read the sample image by the reading device 210 every time the printed image is changed, so that the inspection efficiency may decrease. On the other hand, the method of processing the image data is preferable in that the inspection efficiency is further improved because the labor of reading the sample image by the reading device 210 can be omitted.

Defects in the printed image include streak images, spot images, stains, scratches, etc. that are visually recognized in the printed image. The streak is an image region that extends linearly in a printed image and has a different density than the surrounding region. The spot image is a small dot-shaped image area having a different density than the surrounding area in the printed image. For example, a Pochi image is a small dot-like image attached to a white area of a recording medium. Such streaks, spot images, etc. correspond to positive defects that should not be in the printed image. The image region corresponding to the positive defect is an example of the defect region and also an example of the first defect region.

In addition, defects in the printed image include white spots in the printed image, printing errors, and the like. The white spot is an image region in which the toner image is not formed even though the toner image should be formed on the recording medium. The area where the toner image should be formed corresponds to the characteristic of the good product as it should be, and the white spot corresponds to the negative defect that the characteristic of the good product should be not in the inspection target. The image region corresponding to the negative defect is an example of the defect region and also an example of the second defect region.

The reading device 210 is an example of an imaging unit that captures an inspection image of an inspection object. The inspection image is an image taken by the reading device 210 reading a printed image which is an inspection target. The details of the configuration of the image inspection unit 200 will be described below with reference to FIG.

The image inspection unit 200 discharges the recording medium for which the inspection of the printed image has been completed to the stacker 300. The stacker 300 has a tray 301. The stacker 300 stacks the recording medium discharged from the image inspection unit 200 on the tray 301.

<Structure of image inspection unit 200>
Next, the configuration of the image inspection unit 200 will be described with reference to FIG. 2. FIG. 2 is a diagram showing an example of the configuration of the image inspection unit 200. As shown in FIG. 2, the image inspection unit 200 includes a reading device 210, a background unit 220, a contact glass 230, a lighting unit 240, and a transport roller pair 250 and 251.

The transport roller pairs 250 and 251 transport the recording medium P sent from the image forming unit 100 in the Y direction of FIG. Either one of the transport roller pair 250 and 251 can be a drive roller pair that is rotationally driven by a drive unit such as a motor, and the other can be a driven roller pair that rotates according to the recording medium P to be transported.

The contact glass 230 is made of transparent glass and has a function of coming into contact with the conveyed recording medium P and suppressing fluttering of the recording medium P during reading (during imaging) by the reading device 210. As shown in FIG. 2, the recording medium P is conveyed between the contact glass 230 and the background portion 220 in the Y direction.

The lighting unit 240 is composed of an LED array or the like in which a plurality of LEDs (Light Emitting Diodes) are arranged in an axial direction (in the X direction in FIG. 2, hereinafter referred to as a width direction) such as a transfer roller pair 250, and is conveyed. The recording medium P is illuminated with a line of light.

However, the lighting unit 240 is not limited to this configuration, and LEDs of each color of red, green, and blue are turned on at the same time, light of each color is mixed, and light having a wide wavelength band close to white light is irradiated. May be. Further, it may have a configuration having one element that irradiates a long line-shaped light in the width direction, such as a fluorescent tube. According to the fluorescent tube, it is possible to irradiate white light having a uniform brightness in the width direction.

Furthermore, using a light guide member whose width direction is the longitudinal direction, white, red, green, and blue LEDs arranged at both ends of the light guide member are turned on and passed through the light guide member to make a line. You may irradiate the light of the shape. The light guide member can irradiate light having uniform brightness in the width direction. Further, a light guide lens for efficiently guiding the light from the LED array to the region where the edge in the width direction of the conveyed recording medium P passes may be provided.

The reading device 210 is provided on one surface side (positive Z direction side in FIG. 2) of the recording medium, and is realized by a CIS (Contact Image Sensor) or the like. More specifically, the reading device 210 includes mirrors 211 to 213, a lens 214, and a pixel array 215. The reflected light from the recording medium P of the light emitted from the illumination unit 240 is reflected by the mirrors 211 to 213, respectively, as shown by the broken line in FIG. 2, and is formed into an image on the light receiving surface of the pixel array 215 by the lens 214. Be done.

The pixel array 215 is an element in which PDs (Photo Diodes), which are photoelectric conversion elements that convert optical signals into electrical signals, are arranged in an array in the width direction. One photoelectric conversion element corresponds to one pixel, and outputs an electric signal according to the amount of light received. The pixel array 215 outputs an electric signal (image signal) of one line of pixels in the width direction. Further, at this time, the pixel array 215 receives the reflected light from the recording medium P conveyed in the Y direction by the transfer roller pair 250 and 251 at predetermined timings, and outputs an image signal for one line. Two-dimensional image data is acquired by connecting the image signals for one line output in this way in a direction orthogonal to the arrangement direction of the pixels in the pixel array 215.

Further, the pixel array 215 includes a pixel array 215R that receives red light, a pixel array 215G that receives green light, and a pixel array 215B that receives blue light, respectively, in the width direction and of pixels. They are arranged so that the arrangement directions are substantially parallel.

The pixel array 215R that receives red light is provided with a red color filter in front of the light receiving surface, and receives red light that has passed through the color filter. The red color filter passes light in the red wavelength band and absorbs or reflects light in other wavelength bands. Similarly, the pixel array 215G includes a green color filter and receives light in the green wavelength band, and the pixel array 215B includes a blue color filter and receives light in the blue wavelength band.

A CCD (Charge Coupled Device), CMOS (Complementary Metal-Oxide-Semiconductor), or the like may be used for the pixel array. Further, the pixel array 215 may be configured by using a CCD or CMOS area sensor having a two-dimensional pixel array. Further, in order to increase the light collection efficiency of the pixel array 215, a rod lens array or the like for guiding the light reflected by the recording medium P to the pixel array 215 may be provided.

The reading device 210 can receive the reflected light from the recording medium P to be read and output an image signal including an inspection image formed on the recording medium P.

The background portion 220 comes into contact with the other surface side of the recording medium P (the negative Z direction side in FIG. 2), and when the reading device 210 reads the edge of the recording medium P, the background member becomes the background member of the recording medium P. Have. The "other surface" is a surface opposite to the surface of the recording medium P on the side where the reading device 210 is arranged.

More specifically, the background portion 220 includes a revolver 221, a white small diameter roller 222, a black small diameter roller 223, a white large diameter roller 224, and a black large diameter roller 225. The white small diameter roller 222, the white large diameter roller 224, the black small diameter roller 223, and the black large diameter roller 225 are arranged around the columnar shaft of the columnar member included in the revolver 221 as shown in FIG. , Attached to the revolver 221.

The columnar member included in the revolver 221 is formed with a plurality of circular through holes penetrating in the direction of the columnar axis. Further, each through hole is formed so as to be arranged around the cylindrical axis of the revolver 221. By passing each roller through this through hole, each roller can be attached to the revolver 221.

The circular through hole does not necessarily have to have a circular cross-sectional shape, and may be a part of the circular cross-sectional shape. Further, when a prism member or the like is mounted instead of the roller, a rectangular through hole may be formed.

Further, the revolver 221 is rotatable around its cylindrical axis (direction of arrow u in FIG. 2). As an example, a motor mounted on the revolver 221 is rotationally driven by a control signal to rotate the revolver 221 in the direction of arrow u, and a predetermined roller among a plurality of rollers mounted on the revolver 221 is recorded. It can be brought into contact with the other surface of the medium P. However, the present invention is not limited to rotating the revolver 221 in response to a control signal, and an operator or the like of the image forming apparatus 1 manually rotates the revolver 221 to determine a predetermined roller among a plurality of mounted rollers. The roller may be brought into contact with the other surface of the recording medium P.

As described above, the white small diameter roller 222, the white large diameter roller 224, the black small diameter roller 223, and the black large diameter roller 225 are provided so as to be in contact with the other surface of the recording medium P by the rotation of the revolver 221.

The white small diameter roller 222 and the black small diameter roller 223 have the same roller diameter, but different roller colors. As an example, when the background color of the recording medium P is white, the color contrast between the recording medium P and the black small diameter roller 223 as a background member is increased by bringing the black small diameter roller 223 into contact with the other surface of the recording medium P. Since the height is high, it becomes easier to detect the edge position of the recording medium P. In the above, the white small diameter roller 222 and the black small diameter roller 223 have been described as examples, but the same applies to the white large diameter roller 224 and the black large diameter roller 225. Further, the white and black rollers are taken as an example, but the present invention is not limited to this, and rollers of other colors may be used depending on the color of the recording medium P.

On the other hand, the black small diameter roller 223 and the black large diameter roller 225 have the same roller color but different roller diameters. Therefore, when the black large-diameter roller 225 is brought into contact with the other surface of the recording medium P, the black large-diameter roller 225 pushes the recording medium P in the positive Z direction of FIG. 2, so that the black small-diameter roller 223 is pressed against the recording medium P. The height in the direction intersecting the surface of the recording medium P (Z direction in FIG. 2) can be made different from the case of contacting the other surface of the recording medium P. That is, when the black small-diameter roller 223 is brought into contact with the other surface of the recording medium P, when the black large-diameter roller 225 is brought into contact with the black large-diameter roller 225, one surface of the recording medium P is brought closer to the reading device 210. be able to.

The thickness of the recording medium P may differ depending on the type, and the distance (height) from one surface of the recording medium P to the reading device 210 differs between the thin recording medium P and the thick recording medium P. Due to such a difference in height, the image read by the reading device 210 may include a defocus.

In particular, since the reading device 210 is configured to be thin, the depth of field is shallow. Therefore, the read image is likely to be out of focus due to a slight difference in height from one surface of the recording medium P to the reading device 210 due to the difference in the thickness of the recording medium P. When an out-of-focus image is used, it becomes difficult to accurately detect the edge position of the recording medium P and the edge position of the image formed on the recording medium P.

On the other hand, in the embodiment, for example, in the case of a thin recording medium P, one surface of the recording medium P is brought closer to the reading device 210 by bringing the black large-diameter roller 225 into contact with the other surface of the recording medium P. be able to. On the contrary, in the case of the thick recording medium P, by bringing the black small diameter roller 223 into contact with the other surface of the recording medium P, one surface of the recording medium P can be kept away from the reading device 210.

In this way, the height from one surface of the recording medium P to the reading device 210 can be made constant regardless of the thickness of the recording medium P, and defocusing in the read image can be prevented. In the above, the black small diameter roller 223 and the black large diameter roller 225 have been described as examples, but the same applies to the white small diameter roller 222 and the white large diameter roller 224.

In the above-mentioned example, a roller is used as a background member, and the diameter of the roller is changed to make the "height" different, but the present invention is not limited to this. For example, the "height" can be made different by using a prism as a background member and changing the height (thickness) dimension in the cross-sectional shape of the prism.

<Hardware configuration of the image forming apparatus according to the embodiment>
Next, the hardware configuration of the image forming apparatus 1 will be described with reference to FIG. FIG. 3 is a block diagram showing an example of the hardware configuration of the image forming apparatus 1 according to the embodiment.

As shown in FIG. 3, the image forming apparatus 1 includes a controller 910, a short-range communication circuit 920, an engine control unit 930, an operation panel 940, and a network I / F 950.

Of these, the controller 910 is the CPU 901, which is the main part of the computer, the system memory (MEM-P) 902, the north bridge (NB) 903, the south bridge (SB) 904, and the ASIC (Application Specific Integrated Circuit). It has a 906, a local memory (MEM-C) 907 which is a storage unit, an HDD controller 908, and an HD909 which is a storage unit. Further, the NB 903 and the ASIC 906 are connected by an AGP (Accelerated Graphics Port) bus 921.

Of these, the CPU 901 is a control unit that controls the entire image forming apparatus 1. The NB903 is a bridge for connecting the CPU901, the MEM-P902, the SB904, and the AGP bus 921, and includes a memory controller that controls reading and writing to the MEM-P902, a PCI (Peripheral Component Interconnect) master, and an AGP target. Has.

The MEM-P902 includes a ROM 902a that is a memory for storing programs and data that realizes each function of the controller 910, and a RAM 902b that is used as a memory for developing programs and data and a memory for drawing at the time of memory printing.

The program stored in the RAM 902b is configured to be recorded and provided as a file in an installable format or an executable format on a computer-readable recording medium such as a CD-ROM, CD-R, or DVD. You may.

The SB904 is a bridge for connecting the NB903 to a PCI device and a peripheral device. The ASIC 906 is an IC (Integrated Circuit) for image processing applications having hardware elements for image processing, and has a role of a bridge connecting the AGP bus 921, the PCI bus 922, the HDD 908, and the MEM-C907, respectively.

This ASIC906 is a PCI target and an AGP master, an arbiter (ARB) which is the core of the ASIC906, a memory controller which controls MEM-C907, and a plurality of DMACs (Direct Memory Access Controllers) which rotate image data by hardware logic and the like. , And a PCI unit that transfers data between the scanner unit 931 and the printer unit 932 via the PCI bus 922.

A USB interface or an IEEE 1394 (Institute of Electrical and Electronics Engineers 1394) interface may be connected to the ASIC 906.

The MEM-C907 is a local memory used as a copy image buffer and a code buffer. The HD909 is a storage for accumulating image data, accumulating font data used at the time of printing, and accumulating forms. The HD909 controls reading or writing of data to the HD909 according to the control of the CPU 901.

The AGP bus 921 is a bus interface for a graphics accelerator card proposed to accelerate graphic processing, and the graphics accelerator card can be accelerated by directly accessing the MEM-P902 with a high throughput. ..

Further, the short-range communication circuit 920 includes a communication circuit 920a. The communication circuit 920a is a communication circuit such as NFC and Bluetooth (registered trademark).

Further, the engine control unit 930 has a scanner unit 931 and a printer unit 932. Further, the operation panel 940 displays the current setting value, the selection screen, and the like, and sets the panel display unit 940a such as a touch panel that accepts the input from the operator, and the setting value of the condition related to image formation such as the density setting condition. It has an operation unit 940b including a numeric keypad for receiving and a start key for receiving a copy start instruction.

The controller 910 controls the entire image forming apparatus 1, and controls, for example, drawing, communication, input from the operation panel 940, and the like. The scanner unit 931 or the printer unit 932 includes an image processing portion such as error diffusion and gamma conversion.

The image forming apparatus 1 can sequentially switch and select the document box function, the copy function, the printer function, and the facsimile function by the application switching key of the operation panel 940.

When the document box function is selected, the document box mode is set, when the copy function is selected, the copy mode is set, when the printer function is selected, the printer mode is set, and when the facsimile mode is selected, the facsimile mode is set.

Further, the network I / F950 is an interface for performing data communication using the network. The short-range communication circuit 920 and the network I / F 950 are electrically connected to the ASIC 906 via the PCI bus 922.

Here, the image forming apparatus 1 further includes a lighting unit drive circuit 960, a revolver drive circuit 970, and a pixel array drive circuit 980.

The lighting unit drive circuit 960 is an electric circuit that is electrically connected to the lighting unit 240 and drives the lighting unit 240. The lighting unit drive circuit 960 outputs a drive signal to the lighting unit 240 in response to a control signal from the CPU 901 or the like, so that the intensity and timing of the light emitted by the lighting unit 240 to the recording medium P are controlled.

The revolver drive circuit 970 is an electric circuit that is electrically connected to the revolver motor 221a attached to the revolver 221 and drives the revolver motor 221a. The revolver drive circuit 970 outputs a drive signal to the revolver motor 221a in response to a control signal from the CPU 901 or the like, whereby the revolver 221 is rotationally driven, and the white small diameter roller 222, the white large diameter roller 224, and the black small diameter roller 223 are driven. , And a predetermined roller of the black large diameter rollers 225 can be brought into contact with the other surface of the recording medium P.

The pixel array drive circuit 980 is an electric circuit that is electrically connected to the pixel array 215 and drives the pixel array 215. The image signal by the pixel array 215 is input via the pixel array drive circuit 980, and a predetermined process can be executed or stored in the HD909 or the like.

[First Embodiment]
<Example of functional configuration of the processing unit 260 of the image inspection unit 200>
Next, with reference to FIG. 4, the functional configuration of the processing unit 260 included in the image inspection unit 200 will be described.

The processing unit 260 performs a process of detecting defects in the printed image by comparing the inspection image obtained by capturing the printed image formed on the recording medium sent from the image forming unit 100 with the non-defective image. In the embodiment, the non-defective image is created in advance and stored in HD909 or the like in FIG. The processing unit 260 acquires a non-defective image with reference to HD909 or the like, and performs comparison processing with the inspection image.

FIG. 4 is a block diagram showing an example of the functional configuration of the processing unit 260. As shown in FIG. 4, the processing unit 260 includes a first divided image generation unit 261, a first defect candidate extraction unit 262, a second divided image generation unit 263, a second defect candidate extraction unit 264, and defect detection. It has a unit 265 and an output unit 266.

Each of these parts is a function or a functioning means realized by operating any of the components shown in FIG. 3 by an instruction from the CPU 901 according to a program developed on the RAM 902b from the ROM 902a. Note that FIG. 4 shows the main configurations of the processing unit 260, but the processing unit 260 may have other configurations.

The first divided image generation unit 261 input an image signal including the inspection image captured by the reading device 210, and averaged the luminance values of the pixels included in each region in each region where the inspection image was divided into a plurality of regions. The first processed image is generated. Here, each region divided into a plurality of regions means a region corresponding to each grid when the inspection image is divided into a plurality of regions by a grid of a matrix. The first processed image is a mosaic-like low-resolution image consisting of a plurality of regions in which the luminance values of all the pixels included in one region are replaced with the average values of the luminance values of all the pixels included in one region. be.

Further, the first divided image generation unit 261 averages the luminance values of the pixels included in each region in each region in which the inspection image is divided into a plurality of regions in which at least one of the phase, direction, or size is different from the first processed image. The second processed image is generated.

Here, the phase of a region means the position of a grid that divides an image such as an inspection image into a plurality of regions. Further, the direction of the region means a direction in which the phase of the region is shifted. The direction of the region is basically vertical (column direction of the grid) or horizontal (row direction of the grid), but may include an oblique direction defined by an angle rotated around the center point of the grid.

The size of the area means the area of the area. For example, it is the area of an area represented by the number of pixels such as 10 × 10 pixels.

In a plurality of regions in which at least one of the phase, direction or size is different from the first processed image, the luminance value of all the pixels included in one region is the average value of the luminance values of all the pixels included in one region. Will be replaced. The second processed image is a mosaic-like low-resolution image composed of such a plurality of regions.

The first divided image generation unit 261 generates a first divided image using at least one of the first processed image and the second processed image. For example, the first divided image generation unit 261 generates the first divided image by selecting either the first processed image or the second processed image. Since the first processed image becomes the first divided image and the second processed image becomes the first divided image, the first divided image can be said to be a generic notation of the first processed image and the second processed image. ..

The first defect candidate extraction unit 262 has the luminance value of the pixel in a certain area and the area around the certain area among the plurality of areas included in the first divided image generated by the first divided image generation unit 261. The brightness values of the pixels are compared, and the first defect candidate region is extracted.

For example, the difference in the luminance value of the pixel is calculated between a certain area and the area around the certain area, and when the difference value is equal to or more than a predetermined threshold value, the certain area is extracted as a defect candidate area. By performing the above processing on all of the plurality of regions in the first divided image while changing a certain region, the defect candidate region in the inspection image is extracted as the first defect candidate region. By setting the first defect candidate area as an effective area and the area other than the first defect candidate area as an invalid area in the first divided image, the extracted first defect candidate area can be displayed.

The second divided image generation unit 263 generates a third processed image obtained by averaging the luminance values of the pixels included in each region in each region obtained by dividing the non-defective image acquired with reference to HD909 or the like into a plurality of regions. The third processed image is a mosaic-like low-resolution image consisting of a plurality of regions in which the luminance values of all the pixels included in one region are replaced with the average values of the luminance values of all the pixels included in one region. be.

Further, the second divided image generation unit 263 averages the luminance values of the pixels included in each region in each region in which the non-defective image is divided into a plurality of regions in which at least one of the phase, direction or size is different from the third processed image. The fourth processed image is generated.

In a plurality of regions in which at least one of the phase, direction or size is different from the third processed image, the luminance value of all the pixels included in one region is the average value of the luminance values of all the pixels included in one region. Will be replaced. The fourth processed image is a mosaic-like low-resolution image composed of such a plurality of regions.

The second divided image generation unit 263 generates the second divided image by using at least one of the third processed image and the fourth processed image. For example, the second divided image generation unit 263 generates the second divided image by selecting either the third processed image or the fourth processed image. Since the third processed image becomes the second divided image and the fourth processed image becomes the second divided image, the second divided image can be said to be a generic notation of the third processed image and the fourth processed image. ..

The second defect candidate extraction unit 264 has the luminance value of the pixel in a certain area and the area around the certain area among the plurality of areas included in the second divided image generated by the second divided image generation unit 263. The brightness values of the pixels are compared, and the second defect candidate region is extracted.

For example, the difference in the luminance value of the pixel is calculated between a certain area and the area around the certain area, and when the difference value is equal to or more than a predetermined threshold value, the certain area is extracted as a defect candidate area. By performing the above processing on all of the plurality of regions in the second divided image while changing a certain region, the defect candidate region in the non-defective image is extracted as the second defect candidate region. By setting the second defect candidate area as the effective area and the area other than the second defect candidate area as the invalid area in the second divided image, the extracted second defect candidate area can be displayed.

The defect detection unit 265 compares the first defect candidate region and the second defect candidate region, and detects the defect region. Specifically, the defect detection unit 265 removes (non-extracts) the defect candidate region common to both the first defect candidate region and the second defect candidate region as a feature of the non-defective product. Then, the region extracted only by the first defect candidate region is detected as a positive defect region in the inspection image but not in the good product image. Further, the region extracted only by the second defect candidate region is detected as a negative defect region in the non-defective product image but not in the inspection image.

In the first defect candidate region and the second defect candidate region used by the defect detection unit 265 for comparison, the phases of the regions (lattices) applied when the first divided image and the second divided image, which are the sources thereof, are generated. It is preferable that the directions or sizes are the same.

The defect detection unit 265 uses the detection result as the defect detection result of the printed image and outputs the detection result to an external device such as a PC (Personal Computer) via the output unit 266.

<Operation example of image inspection unit 200>
Next, the operation of the image inspection unit 200 will be described with reference to FIG. FIG. 5 is a flowchart showing an example of the operation of the image inspection unit 200. FIG. 5 shows the operation of the image inspection unit 200 triggered by the timing at which the recording medium P on which the printed image is formed is sent from the image forming unit 100 to the image inspection unit 200. However, the trigger for the inspection by the image inspection unit 200 is not limited to this, and the start operation of the inspection performed by the user using the operation panel 101 may be a trigger.

First, in step S51, the reading device 210 captures a printed image formed on the recording medium P as an inspection image.

Subsequently, in step S52, the first divided image generation unit 261 inputs an image signal including the inspection image captured by the reading device 210, and divides the inspection image into a plurality of regions which are grids of a matrix.

Subsequently, in step S53, the first divided image generation unit 261 generates the first processed image by averaging the luminance values of the pixels included in each region in each region in the divided plurality of regions.

Subsequently, in step S54, the first divided image generation unit 261 changes at least one of the phases, directions, or sizes of each region in the divided plurality of regions with respect to the first processed image.

Subsequently, in step S55, the first divided image generation unit 261 is used in each region in a plurality of regions in which at least one of the phases, directions, or sizes of each region is changed with respect to the first processed image. The second processed image is generated by averaging the brightness values of the included pixels.

Subsequently, in step S56, the first divided image generation unit 261 generates a first divided image using at least one of the first processed image and the second processed image.

Subsequently, in step S57, the first defect candidate extraction unit 262 has the luminance value of the pixel in a certain region and the luminance value of the pixel in the region surrounding the region among the plurality of regions included in the first divided image. Is compared to extract the first defect candidate region. While changing a certain region, the above comparison processing is performed on all of the plurality of regions, and the first defect candidate region is extracted from the entire first divided image.

Subsequently, in step S58, the second divided image generation unit 263 acquires a non-defective image with reference to HD909 and the like.

Subsequently, in step S59, the second divided image generation unit 263 divides the non-defective image into a plurality of regions which are grids of a matrix.

Subsequently, in step S60, the second divided image generation unit 263 generates a third processed image in which the luminance values of the pixels included in each region are averaged in each region in the divided plurality of regions.

Subsequently, in step S61, the second divided image generation unit 263 changes at least one of the phases, directions, or sizes of each region in the divided plurality of regions with respect to the third processed image.

Subsequently, in step S62, the second divided image generation unit 263 is used in each region in a plurality of regions in which at least one of the phases, directions, or sizes of each region is changed with respect to the third processed image. A fourth processed image is generated by averaging the brightness values of the included pixels.

Subsequently, in step S63, the second divided image generation unit 263 generates the second divided image using at least one of the third processed image and the fourth processed image.

Subsequently, in step S64, the second defect candidate extraction unit 264 has the luminance value of the pixel in a certain region and the luminance value of the pixel in the region surrounding the region among the plurality of regions included in the second divided image. To extract the second defect candidate region. While changing a certain region, the above comparison processing is performed on all of the plurality of regions, and the second defect candidate region is extracted from the entire second divided image.

The operations of steps S51 to S57 and the operations of steps S58 to S64 may be performed in a different order, or both may be executed in parallel.

Subsequently, in step S65, the defect detection unit 265 compares the first defect candidate region and the second defect candidate region.

Subsequently, in step S66, the defect detection unit 265 uses the defect region obtained as a result of the comparison as the defect detection result of the printed image.

Subsequently, in step S67, the defect detection unit 265 outputs the defect detection result of the printed image to an external device such as a PC via the output unit 266.

In this way, the image inspection unit 200 can detect defects in the printed image.

<Example of processing result by processing unit 260>
Next, the processing results by each functional component will be described with reference to FIGS. 6 to 9. First, FIG. 6 is a diagram showing an example of a processing result by the first divided image generation unit 261. 6 (a) is a diagram showing an example of an inspection image input by the first divided image generation unit 261, FIG. 6 (b) is a diagram showing an example of a first processed image, and FIG. 6 (c) is a second. It is a figure which shows an example of a processed image.

As shown in FIG. 6A, the inspection image 60 includes a defect 70 and a non-defective feature 80.

As shown in FIG. 6B, in the first processed image 61, the inspection image 60 is divided into a plurality of lattice regions to form a mosaic-like image. The first processed image 61 includes a defect region 71 corresponding to the defect 70 and a non-defective product feature region 81 corresponding to the non-defective product feature 80.

As shown in FIG. 6 (c), in the second processed image 62, the phase and size of each region of the plurality of regions obtained by dividing the inspection image 60 into a grid are different from those of the first processed image 61. It is a coarser mosaic-like image as compared with the processed image 61. The second processed image 62 includes a defect region 72 corresponding to the defect 70 and a non-defective product feature region 82 corresponding to the non-defective product feature 80.

Next, FIG. 7 is a diagram showing an example of a processing result by the first defect candidate extraction unit 262. 7 (a) is a diagram showing an example of a first defect candidate region extracted from the first processed image 61 of FIG. 6 (b), and FIG. 7 (b) is a second processed image of FIG. 6 (c). It is a figure which shows an example of the 1st defect candidate region extracted from 62.

In FIG. 7, the area other than the first defect candidate area is displayed in black as an invalid area. As shown in FIG. 7A, the first defect candidate region 63 includes a first defect candidate region 73 corresponding to the defect 70 and a first defect candidate region 83 corresponding to the non-defective product feature 80. Further, as shown in FIG. 7B, the first defect candidate region 64 includes a first defect candidate region 74 corresponding to the defect 70 and a first defect candidate region 84 corresponding to the non-defective product feature 80.

Since the first defect candidate region 63 has a smaller grid area and higher resolution than the first defect candidate region 64, it includes many first defect candidate regions such as a substrate pattern included in the inspection image 60. I'm out. The base pattern is a ground pattern of a recording medium on which a printed image is formed.

Next, FIG. 8 is a diagram showing an example of a processing result by the second divided image generation unit 263. 8 (a) is a diagram showing an example of a non-defective image input by the second divided image generation unit 263, FIG. 8 (b) is a diagram showing an example of a third processed image, and FIG. 8 (c) is a fourth. It is a figure which shows an example of a processed image.

As shown in FIG. 8A, the non-defective image 160 includes a defect 170 and a non-defective feature 180. The defect 170 is not included in the inspection image 60 (see FIG. 6) and corresponds to a negative defect that the inspection object does not have the characteristics of a good product as it should be.

As shown in FIG. 8B, in the third processed image 161 the non-defective image 160 is divided into a plurality of lattice regions to form a mosaic-like image. The third processed image 161 includes a defect region 171 corresponding to the defect 170 and a non-defective product feature region 181 corresponding to the non-defective product feature 80.

As shown in FIG. 8 (c), in the fourth processed image 162, the phase and size of each region of the plurality of regions obtained by dividing the non-defective image 160 into a grid are different from those of the third processed image 161. It is a coarser mosaic-like image as compared with the processed image 161. The fourth processed image 162 includes a defect region 172 corresponding to the defect 170 and a non-defective product feature region 182 corresponding to the non-defective product feature 80.

Next, FIG. 9 is a diagram showing an example of the processing result by the second defect candidate extraction unit 264. 9 (a) is a diagram showing an example of a second defect candidate region extracted from the third processed image 161 of FIG. 8 (b), and FIG. 9 (b) is a fourth processed image of FIG. 8 (c). It is a figure which shows an example of the 2nd defect candidate region extracted from 162.

In FIG. 9, the area other than the second defect candidate area is displayed in black as an invalid area. As shown in FIG. 9A, the second defect candidate region 163 includes a second defect candidate region 173 corresponding to the defect 170 and a second defect candidate region 183 corresponding to the non-defective product feature 180. Further, as shown in FIG. 9B, the second defect candidate region 164 includes a second defect candidate region 174 corresponding to the defect 170 and a second defect candidate region 184 corresponding to the non-defective product feature 180.

Since the second defect candidate region 163 has a smaller grid area and higher resolution than the second defect candidate region 164, it includes many second defect candidate regions such as a substrate pattern included in the non-defective image 160. I'm out.

The defect detection unit 265 compares, for example, the first defect candidate region 64 in FIG. 7 (b) and the second defect candidate region 164 in FIG. 9 (b), and the defect candidate region common to both is a feature of the non-defective product. Remove as (make it non-extracted). Then, the region extracted only by the first defect candidate region 64 is detected as a positive defect region in the inspection image 60 but not in the non-defective image 160. The region extracted only by the second defect candidate region 164 is detected as a negative defect region in the good product image 160 but not in the inspection image 60.

In this way, the image inspection unit 200 can detect both positive defects and negative defects.

<Example of action and effect of image inspection unit 200>
As described above, in the present embodiment, the first processed image 61 or the first processed image 61 in which the image of the printed image is imaged as the inspection image 60 and the brightness values of the pixels are averaged in each region of the inspection image 60 divided into a plurality of regions. In each region where the inspection image 60 is divided into a plurality of regions in which at least one of the phase, direction, or size is different from the first processed image 61, at least one of the second processed images 62, in which the brightness values of the pixels are averaged, is used. Generate the first divided image. For example, the second processed image 62 is used to generate the first divided image.

Then, among the plurality of regions included in the second processed image 62 as the first divided image, the luminance value of the pixel in a certain region is compared with the luminance value of the pixel in the region surrounding the region, and the first defect candidate is compared. The region 64 is extracted. As a result, the first defect candidate region 64 of the inspection image 60 can be extracted at high speed and with high sensitivity by using a mechanism in which a human unconsciously recognizes a difference from the normal state.

Further, in each region of the non-defective image 160 divided into a plurality of regions, the non-defective image is in a plurality of regions whose phase, direction or size is different from the third processed image 161 in which the luminance values of the pixels are averaged or the third processed image 161. In each region where the 160 is divided, a second divided image is generated using at least one of the fourth processed image 162, which is the average of the luminance values of the pixels. For example, the second divided image is generated by using the fourth processed image 162.

Then, among the plurality of regions included in the fourth processed image 162 as the second divided image, the luminance value of the pixel in a certain region is compared with the luminance value in the region around the region, and the second defect candidate region 164 is compared. To extract. As a result, the second defect candidate region 164 of the non-defective image 160 can be extracted at high speed and with high sensitivity by utilizing the mechanism by which a human unconsciously recognizes the difference from the normal state.

By comparing the first defect candidate region 64 and the second defect candidate region 164, the region extracted only by the first defect candidate region 64 is a positive defect region (first defect region) in the inspection image 60 and not in the good product image 160. Defect area) is detected. Further, the region extracted only by the second defect candidate region 164 is detected as a negative defect region (second defect region) in the non-defective product image 160 but not in the inspection image 60. In this way, it is possible to obtain a defect detection result including a defect that the characteristic of the non-defective product should not be in the inspection target.

Further, in the present embodiment, a non-defective image is created by processing the image data which is the original data for forming the printed image. For example, in the method of reading a sample image with the reading device 210 to create a good product image, when the printed image is changed frequently, the reading device 210 reads the sample image every time the printed image is changed. This may reduce the efficiency of the inspection.

On the other hand, in the method of processing the image data to create a non-defective image, the labor of reading the sample image by the reading device 210 can be omitted, so that the inspection efficiency can be further improved.

In addition, we know how to suppress erroneous detection of defects by creating multiple non-defective product images and determining that features common to multiple non-defective product images are not defects but "features of non-defective products". Has been done. However, with this method, it may be difficult to create a plurality of non-defective images when a printed image or the like whose type can be easily changed by changing the image data is used as an inspection target.

In the present embodiment, erroneous detection of defects is suppressed by extracting a second defect candidate region of a digital non-defective image created by processing image data. As a result, even when a printed image or the like that is frequently changed is used as an inspection target, it can be easily dealt with.

[Second Embodiment]
Next, the image forming apparatus 1a according to the second embodiment will be described. The same component numbers as those described in the first embodiment are designated by the same part numbers, and duplicate explanations will be omitted as appropriate.

<Example of functional configuration of the processing unit 260a of the image inspection unit 200a>
The image forming apparatus 1a has an image inspection unit 200a. Further, the image inspection unit 200a has a processing unit 260a. FIG. 10 is a block diagram showing an example of the functional configuration of the processing unit 260a.

As shown in FIG. 10, the processing unit 260a includes a first divided image generation unit 261a, a first defect candidate extraction unit 262a, a second division image generation unit 263a, a second defect candidate extraction unit 264a, and defect detection. It has a unit 265a and a voting unit 267.

Each of these parts is a function or a functioning means realized by operating any of the components shown in FIG. 3 by an instruction from the CPU 901 according to a program developed on the RAM 902b from the ROM 902a.

The first divided image generation unit 261a inputs an image signal including an inspection image captured by the reading device 210, and divides the inspection image into a plurality of regions, and averages the luminance values of the pixels in each region to obtain a first divided image. N pieces are generated while differentiating at least one of the phases, directions, or sizes of the plurality of regions. Here, N represents an integer of 2 or more. The first divided image generation unit 261a can generate N first divided images having different mosaic roughness, for example, in a mosaic-like image.

The first defect candidate extraction unit 262a is each of the N first divided images generated by the first divided image generation unit 261a, and the luminance value of the pixel in a certain region among the plurality of regions included in the first divided image. And, a process of comparing the luminance values of the pixels in the area around the certain area is performed, and N first defect candidate areas are extracted.

The second divided image generation unit 263 acquires a good product image with reference to HD909 or the like, and divides the good product image into a plurality of regions, and averages the luminance values of the pixels in each region to obtain the second divided image in the plurality of regions. N pieces are generated while differentiating at least one of the phases, directions, or sizes of. The second divided image generation unit 263a can generate N second divided images having different mosaic roughness, for example, in a mosaic-like image.

The second defect candidate extraction unit 264a is each of the N second divided images generated by the second divided image generation unit 263a, and the luminance value of the pixel in a certain region among the plurality of regions included in the second divided image. And, a process of comparing the luminance values of the pixels in the area around the certain area is performed, and N second defect candidate areas are extracted.

The defect detection unit 265a compares N first defect candidate regions with N second defect candidate regions corresponding to the first defect candidate regions, and detects N defect regions. The defect detection unit 265a removes (non-extracts) the defect candidate region common to both the first defect candidate region and the second defect candidate region, for example, as a feature of the non-defective product.

Then, the region extracted only by the first defect candidate region is detected as N positive defect regions in the inspection image but not in the good product image. Further, the region extracted only by the second defect candidate region is detected as N negative defect regions in the non-defective product image but not in the inspection image.

The voting unit 267 generates at least one of a first voting defect region in which N positive defect regions are voted and a second voting defect region in which N negative defect regions are voted.

Specifically, the voting unit 267 prepares a voting space for positive defects and a voting space for negative defects. The voting unit 267 votes N positive defect regions into a voting space for positive defects to generate a first voting defect region. Further, N negative defect regions are voted in the voting space for negative defects to generate a second voting defect region.

Here, the voting process is a process in which the first defective area and the second defective area are used as voting source information, and voting (integration) is performed in the voting space set corresponding to the inspection image. The voting space may be configured with a resolution equal to the original resolution of the inspection image and the good product image, and voting may be performed for each pixel of the voting space. It is also possible to perform a weighted voting process in which areas are weighted according to predetermined rules when voting.

The voting unit 267 can use at least one of the first voting defect area and the second voting defect area as the defect detection result and output it to an external device such as a PC (Personal Computer) via the output unit 266.

<Operation example of image inspection unit 200a>
Next, the operation of the image inspection unit 200a will be described with reference to FIG. FIG. 11 is a flowchart showing an example of the operation of the image inspection unit 200a.

FIG. 11 shows the operation of the image inspection unit 200a triggered by the timing at which the recording medium P on which the printed image is formed is sent from the image formation unit 100 to the image inspection unit 200a. However, the trigger for the inspection by the image inspection unit 200 is not limited to this, and the start operation of the inspection performed by the user using the operation panel 101 may be a trigger.

First, in step S111, the reading device 210 captures a printed image formed on the recording medium P as an inspection image.

Subsequently, in step S112, the first divided image generation unit 261a substitutes 1 for the counter n.

Subsequently, in step S113, the first divided image generation unit 261a inputs an image signal including the inspection image captured by the reading device 210, and divides the inspection image into a plurality of regions which are grids of a matrix.

Subsequently, in step S114, the first divided image generation unit 261a generates the first divided image _Pn by averaging the luminance values of the pixels included in each region in each region in the divided plurality of regions.

Subsequently, in step S115, the first defect candidate extraction unit 262a determines the luminance value of the pixel in a certain region and the pixel in the region around the certain region among the plurality of regions included in the first divided image _Pn . The first defect candidate region _Un is extracted by comparing the luminance values. While changing a certain region, the above comparison processing is performed on all of the plurality of regions, and the first defect candidate region _Un is extracted from the entire first divided image P _n .

Subsequently, in step S116, the first divided image generation unit 261a changes at least one of the phases, directions, or sizes of each region in the divided plurality of regions with respect to the first divided image P _n , and causes each region. The first divided image P _{n + 1} is generated by averaging the luminance values of the included pixels.

Subsequently, in step S117, the first divided image generation unit 261a determines whether or not n is equal to N.

If it is determined in step S117 that n is not equal to N (step S117, No), in step S118, the first divided image generation unit 261a adds 1 to the counter n. After that, the processing after step S114 is performed again.

On the other hand, when it is determined in step S117 that n is equal to N (step S117, Yes), in step S119, the second divided image generation unit 263a acquires a non-defective image with reference to HD909 and the like.

Subsequently, in step S120, the second divided image generation unit 263a substitutes 1 for the counter n.

Subsequently, in step S121, the second divided image generation unit 263a divides the non-defective image into a plurality of regions which are grids of a matrix.

Subsequently, in step S122, the second divided image generation unit _263a generates a second divided image Qn obtained by averaging the luminance values of the pixels included in each region in each region in the divided plurality of regions.

Subsequently, in step S123, the second defect candidate extraction unit 264a determines the luminance value of the pixel in a certain region and the pixel in the region around the certain region among the plurality of regions included in the second divided image _Qn . The second defect candidate region V _n is extracted by comparing the luminance values. While changing a certain region, the above comparison processing is performed on all of the plurality of regions, and the second defect candidate region V _n is extracted from the entire second divided image Q _n .

Subsequently, in step S124, the second divided image generation unit 263a changes at least one of the phases, directions, or sizes of each region in the divided plurality of regions with respect to the second divided image Q _n , and causes each region. A second divided image Q _{n + 1} is generated by averaging the brightness values of the included pixels.

Subsequently, in step S125, the second divided image generation unit 263a determines whether or not n is equal to N.

If it is determined in step S125 that n is not equal to N (step S125, No), in step S126, the second divided image generation unit 263a adds 1 to the counter n. After that, the processing after step S122 is performed again.

On the other hand, when it is determined in step S125 that n is equal to N (step S125, Yes), in step S127, the defect detection unit 265a substitutes 1 for the counter n.

The operation for the inspection image in steps S111 to S118 and the operation for the non-defective image in steps S119 to S126 may be performed first, or both may be performed in parallel.

Subsequently, in step S128, the defect detection unit 265a compares the first defect candidate region _Un with the second defect candidate region V _n corresponding to the first defect candidate region _Un .

Subsequently, in step S129, the defect detection unit 265a detects a positive defect region W _n and a negative defect region X _n by comparison.

Subsequently, in step S130, the defect detection unit 265a determines whether or not n is equal to N.

If it is determined in step S130 that n is not equal to N (step S130, No), in step S131, the defect detection unit 265a adds 1 to the counter n. After that, the processing after step S128 is performed again.

On the other hand, when it is determined in step S130 that n is equal to N (step S130, Yes), in step S132, the voting unit 267 has voted _N positive defect regions Wn for the first voting defect. At least one of the region or the second voting defect region obtained by voting the N negative defect regions _Xn is generated as the defect detection result.

Subsequently, in step S133, the defect detection unit 265a outputs the defect detection result of the printed image to an external device such as a PC via the output unit 266.

In this way, the image inspection unit 200a can detect defects in the printed image.

<Example of processing result by processing unit 260a>
Next, FIG. 12 is a diagram showing an example of processing by the processing unit 260a. FIG. 12 shows an example of processing results for each step in which the processing unit 260a inputs the inspection image 60 and the non-defective product image 160 and performs the processing.

The _first grid-divided images P'1 to P'N show _N images obtained by dividing the inspection image 60 by changing the size of the grid-like region. The second grid-divided images _Q'1 to Q'N show _N images obtained by dividing the non-defective image 160 by changing the size of the grid-like region.

The _first divided images P1 to PN show _N images in which at least one of the phases or directions of the grid-like regions in each of the _first lattice divided images P'1 to _P'N is changed. .. The second divided images _Q1 to QN show _N images in which at least one of the phases or directions of the grid-like regions in each of the second lattice divided images _Q'1 to _Q'N is changed. ..

The first defect candidate regions U ₁ to _UN indicate the first defect candidate regions extracted from each of the first divided images P ₁ to _PN . The second defect candidate regions V ₁ to _VN indicate the second defect candidate regions extracted from each of the second divided images Q ₁ to Q _N.

By comparing the _first defect candidate regions U1 to _UN and the _second defect candidate regions V1 to _VN corresponding to the _first defect candidate regions _U1 to UN, respectively, the positive defect regions _W1 to _WN Alternatively, at least _one of the negative defect regions X1 to _XN can be detected.

Here, the image inspection unit 200 shown in the first embodiment corresponds to the case of N = 2 in the image inspection unit 200a. Specifically, the first processed image 61 in FIG. 6 corresponds to the _first divided image P1 with n = 1, and the second processed image 62 in FIG. 6 corresponds to the first divided image P2 when n = ₂ . handle. Further, the first defect candidate region 63 in FIG. 7 corresponds to the first defect candidate region U ₁ , and the first defect candidate region 64 corresponds to the first defect candidate region U ₂ .

Similarly, the third processed image 161 of FIG. 8 corresponds to the _second divided image Q1 of n = ₁ , and the fourth processed image 162 of FIG. 8 corresponds to the second divided image Q2 when n = 2. .. Further, the second defect candidate region 163 in FIG. 9 corresponds to the second defect candidate region V ₁ , and the second defect candidate region 164 corresponds to the second defect candidate region V ₂ .

That is, in the present embodiment, N _first divided images P1 to _PN including the example of the first processed image 61 or the second processed image 62 of FIG. 6 are generated, and the first defect candidate region 63 of FIG. 7 is generated. And _{N first} defect candidate regions U1 to UN including 64 examples can be extracted.

Further, _N second divided images _Q1 to QN including the example of the third processed image 161 or the fourth processed image 162 of FIG. 8 are generated, and the examples of the second defect candidate regions 163 and 164 of FIG. 9 are included. _{N second} defect candidate regions V1 to VN can be extracted.

Based on these, the defect detection unit _265a can detect at least _one of the positive defect regions W1 to _WN or the negative defect regions X1 to _XN .

Next, FIG. 13 is a diagram showing an example of a defect detection result of the image inspection unit 200a. FIG. 13A is a diagram showing a first voting defect region 360, and FIG. 13B is a diagram showing an example of a second voting defect region 370.

The _first voting defect region 360 in FIG. 13A is a representation in which the positive defect regions _W1 to WN are voted and the integrated value of the brightness of the pixels by the voting process is normalized and displayed. As shown in FIG. 13 (a), the positive defect 361 is clearly detected.

The second voting defect region 370 in FIG. 13B is obtained by voting the negative defect regions _X1 to _XN and normalizing and displaying the integrated value of the brightness of the pixels by the voting process. As shown in FIG. 13 (b), the negative defect 371 is clearly detected.

<Action and effect of image inspection unit 200a>
As described above, in the present embodiment, the _first voting defect region 360 in which _N positive defect regions W1 to WN are voted, or the Nth negative defect regions X1 to XN in which _N negative defect regions X1 to _XN are voted are voted. 2 Generate at least one of the voting defect areas 370.

By performing the voting process, the positive defects commonly detected in the _N positive defect regions W1 to _WN are integrated and the pixel brightness becomes larger than that of the other regions, so that they are more conspicuous. Become. Similarly, in the _N negative defect regions X1 to _XN , the negative defects commonly detected are integrated and the brightness of the pixel becomes larger than that of the other regions, so that they become more conspicuous. As a result, it is possible to suppress erroneous detection of defects and detect defects with high accuracy.

In the present embodiment, an example in which N positive defect regions W ₁ to W _N and N negative defect regions X ₁ to X _N are voted respectively is shown, but the present invention is not limited to this. .. _One of _N positive defect regions W1 to WN is selected as a positive defect detection result, or _one of _N negative defect regions X1 to XN is selected as a negative defect detection result. You can also do it.

For example, when _N positive defect regions W1 to _WN and _N negative defect regions X1 to _XN are detected by different sizes of a plurality of regions, the smaller the region size, the higher the resolution. Become. In this case, in a recording medium having a large surface roughness of the substrate such as plain paper, the substrate is likely to be erroneously detected as a _defect region, so that the region size is large and the resolution is low. When _n is selected, erroneous detection can be suitably suppressed.

On the other hand, in a recording medium having a small surface roughness of the base material such as glossy paper, the base material is unlikely to be erroneously _detected as a _defect region. This will increase the resolution and allow smaller defects to be detected.

In this way, by selecting either the positive defect region _{Wn or the negative defect region Xn ,} appropriate defect detection can be performed according to the characteristics of the inspection object.

Although the embodiments have been described above, the present invention is not limited to the above-described embodiments specifically disclosed, and various modifications and changes can be made without departing from the scope of claims. ..

Although the printed image formed by the electrophotographic method is exemplified in the above-described embodiment, the image inspection device according to the embodiment can be applied to the printed image formed by another method such as the inkjet method.

Further, the image inspection device according to the embodiment is a device that performs inspection based on an image (inspection image) of an inspection object, and the inspection object is not limited to an image such as a printed image. For example, parts and the like can be inspected.

In addition, the numbers such as the ordinal number and the quantity used above are all exemplified for the purpose of concretely explaining the technique of the present invention, and the present invention is not limited to the exemplified numbers. Further, the connection relationship between the constituent elements is exemplified for concretely explaining the technique of the present invention, and the connection relationship for realizing the function of the present invention is not limited to this.

Further, each function of the embodiment described above can be realized by one or a plurality of processing circuits. Here, the "processing circuit" as used herein is a processor programmed to perform each function by software, such as a processor implemented by an electronic circuit, or a processor designed to execute each function described above. It shall include devices such as ASIC (Application Specific Integrated Circuit), DSP (digital signal processor), FPGA (field programmable gate array) and conventional circuit modules.

The embodiment also includes an image inspection method. For example, the image inspection method is an image inspection method using an image inspection device for detecting defects in the inspection object, and is a step of capturing an inspection image of the inspection object and a step of capturing the inspection image and dividing the inspection image into a plurality of regions. In the first processed image obtained by averaging the brightness values of the pixels in the region, or in each region where the inspection image is divided into a plurality of regions in which at least one of the phase, direction or size is different from the first processed image, the pixel A step of generating a first divided image using at least one of a second processed image obtained by averaging the brightness values, a brightness value of a pixel in a certain region among a plurality of regions included in the first divided image, and the said. The brightness of the pixels in the step of comparing the brightness values of the pixels in the area around a certain area and extracting the first defect candidate area, and in each area where the non-defective image to be compared with the inspection image is divided into a plurality of areas. The third processed image in which the values are averaged, or the good product image is divided into a plurality of regions in which at least one of the phase, direction, or size is different from the third processed image, and the brightness values of the pixels are averaged in each region. A step of generating a second divided image using at least one of the four processed images, a brightness value of a pixel in a certain region among a plurality of regions included in the second divided image, and a region around the certain region. A step of comparing the brightness values of the pixels in the image and extracting the second defect candidate region, and a step of comparing the first defect candidate region with the second defect candidate region and detecting the defect region are performed. By such an image inspection method, the same effect as the above-mentioned image inspection apparatus can be obtained.

1 Image forming device 60 Inspection image 160 Good product image 100 Image forming unit 103 Image forming unit 200 Image inspection unit (an example of image inspection device)
210 Reading device (example of image pickup unit)
220 Background part 221 Revolver 222 White small diameter roller 223 Black small diameter roller 224 White large diameter roller 225 Black large diameter roller 230 Contact glass 240 Lighting unit 260 Processing unit 261, 261a First division image generation unit 262, 262a First defect candidate extraction unit 263, 263a 2nd divided image generation unit 264, 264a 2nd defect candidate extraction unit 265, 265a Defect detection unit 266 Output unit 267 Voting unit

Japanese Patent No. 5821708 Japanese Unexamined Patent Publication No. 2017-211206

Claims (9)

An image inspection device that detects defects in the inspection object.
An image pickup unit that captures an inspection image of the inspection object,
In each region obtained by dividing the inspection image into a plurality of regions, the first processed image obtained by averaging the luminance values of the pixels, or the plurality of regions having at least one of the phase, direction, or size different from the first processed image. A first divided image generation unit that generates a first divided image using at least one of a second processed image obtained by averaging the brightness values of pixels in each region where the inspection image is divided.
A first defect that extracts a first defect candidate region by comparing the brightness value of a pixel in a certain region with the brightness value of a pixel in a region surrounding the region among a plurality of regions included in the first divided image. Candidate extraction unit and
In each region where the non-defective product image to be compared with the inspection image is divided into a plurality of regions, the third processed image obtained by averaging the brightness values of the pixels, or at least one of the phase, direction, or size is the third processed image. Is a second divided image generation unit that generates a second divided image by using at least one of a fourth processed image obtained by averaging the brightness values of pixels in each region obtained by dividing the non-defective image into a plurality of different regions.
A second defect that extracts a second defect candidate region by comparing the luminance value of a pixel in a certain region with the luminance value of a pixel in a region surrounding the region among a plurality of regions included in the second divided image. Candidate extraction unit and
An image inspection apparatus including a defect detection unit that compares the first defect candidate region and the second defect candidate region and detects the defect region. The defect detection unit is a first defect region in the first defect candidate region and not in the second defect candidate region, or a second defect region in the second defect candidate region and not in the first defect candidate region. The image inspection apparatus according to claim 1, wherein at least one of the above is detected. An image inspection device that detects defects in the inspection object.
An image pickup unit that captures an inspection image of the inspection object,
The first divided image obtained by averaging the luminance values of the pixels in each of the divided regions of the inspection image is N with different phases, directions, or sizes of the plurality of regions included in the inspection image. The first divided image generation unit to generate pieces and
In each of the N first divided images, the brightness value of the pixel in a certain area and the brightness value of the pixel in the area around the certain area among the plurality of regions included in the first divided image are compared. A first defect candidate extraction unit that performs processing and extracts N first defect candidate regions,
The second divided image obtained by averaging the luminance values of the pixels in each region obtained by dividing the good product image to be compared with the inspection image into a plurality of regions is obtained by at least the phase, direction or size of the plurality of regions included in the good product image. A second divided image generator that generates N images while making one different,
In each of the N second divided images, the brightness value of the pixel in a certain area and the brightness value of the pixel in the area around the certain area among the plurality of regions included in the second divided image are compared. A second defect candidate extraction unit that performs processing and extracts N second defect candidate regions,
It has N defect candidate regions and N defect detection units corresponding to the first defect candidate regions and detecting N defect regions by comparing each of the second defect candidate regions. Image inspection equipment.
(N represents an integer of 2 or more.) The defect detection unit has N first defect regions in the first defect candidate region and not in the second defect candidate region, or N in the second defect candidate region and not in the first defect candidate region. The image inspection apparatus according to claim 3, wherein at least one of the second defect regions is detected. 4. Claim 4 further having a voting unit that generates at least one of a first voting defect area in which N first defective areas have been voted or a second voting defect area in which N second defective areas have been voted. The image inspection device described in. The image inspection apparatus according to any one of claims 1 to 5, wherein the non-defective image is an image obtained by capturing an image of the inspection object of a non-defective product. The inspection object is a printed image formed on a recording medium based on image data.
The image inspection device according to any one of claims 1 to 5, wherein the non-defective image is an image created based on the image data. An image forming apparatus having the image inspection apparatus according to any one of claims 1 to 7. It is an image inspection method using an image inspection device that detects defects in the inspection object.
The process of capturing an inspection image of the inspection object and
In each region obtained by dividing the inspection image into a plurality of regions, the first processed image obtained by averaging the luminance values of the pixels, or the plurality of regions having at least one of the phase, direction, or size different from the first processed image. A step of generating a first divided image using at least one of a second processed image obtained by averaging the brightness values of pixels in each region where the inspection image is divided.
A step of comparing the luminance value of a pixel in a certain region with the luminance value of a pixel in a region surrounding the region among a plurality of regions included in the first divided image, and extracting a first defect candidate region.
In each region where the non-defective product image to be compared with the inspection image is divided into a plurality of regions, the third processed image obtained by averaging the luminance values of the pixels, or at least one of the phase, direction or size is the third processed image. Is a step of generating a second divided image using at least one of a fourth processed image obtained by averaging the brightness values of pixels in each region obtained by dividing the good product image into a plurality of different regions.
A step of comparing the luminance value of a pixel in a certain region with the luminance value of a pixel in a region surrounding the region among a plurality of regions included in the second divided image, and extracting a second defect candidate region.
An image inspection method for performing a step of comparing the first defect candidate region and the second defect candidate region and detecting the defect region.

Images