JP7231461B2 - Image inspection system and image inspection method - Google Patents

Description

The present invention relates to an image inspection system and an image inspection method for inspecting an inspection object such as a workpiece.

Conventionally, there has been known an image inspection system configured to measure the three-dimensional shape of an inspection object and determine the quality of the inspection object based on the measurement results. The image inspection system disclosed in Patent Document 1 obtains a normal vector of the surface of the inspection object from a plurality of luminance images with different illumination directions, and generates a shape image representing the surface shape of the inspection object based on the normal vector. The so-called photometric stereo principle is used to determine the quality of an object to be inspected using an inspection image showing a shape generated based on the photometric stereo principle. In Japanese Patent Application Laid-Open No. 2002-100000, pattern search is performed on an inspection image to set an inspection area, positional correction of the inspection area is performed, and then presence or absence of a scratch is inspected.

Such an image inspection system includes a control device such as a personal computer, and a camera connected to the control device. In general, it is configured to perform various image processing and determination processing. In such a system, in order to perform appropriate imaging with the camera, it is necessary to change the setting values such as the exposure time and analog gain of the camera according to the situation and the inspection object, and the setting values of these setting items are controlled. It is possible to change from the device through the communication path. However, the correspondence between each setting item and the address of the register that holds the setting value is usually different for each camera model and each camera manufacturer. This entails a lot of trouble, such as modifying the program of the image inspection application and reinstalling the driver software.

Therefore, in recent years, the GenICam standard has been formulated as a standardization standard for the connection interface between the control device and the camera, and the GenICam ( Standardization is taking place between cameras and applications that meet the (registered trademark) standard.

JP 2015-232481 A

By the way, by using an image inspection application compatible with the GenICam standard and a connection interface between the camera, there is an advantage that the degree of freedom in selecting the model of the camera is improved.

In addition, there are cameras that have functions that can internally perform various filtering processes on captured images. It is thought that further improvement of convenience can be aimed at.

However, some filtering processes exhibit effects that are strongly dependent on the directionality of the image. That is, for example, when a captured image is obtained in which a white or black defect appears in a striped background that is continuous in the Y direction, the captured image is subjected to uniform defect extraction filter processing in the X and Y directions. Then, since the defect extraction filter is a filter that extracts a place where there is a large change between adjacent pixels as a defect, the striped pattern in the background is detected as a place where there is a large change in the Y direction, and the background cannot be removed. As a result, the accuracy of defect detection may deteriorate. On the other hand, if a defect extraction filter process that captures changes only in the X direction is executed, the striped pattern in the background will not be detected, and it will be possible to extract only the defect portion. The inspection object may be rotating, and this rotation causes the background striped pattern to continue in directions other than the X direction. In this case, the background cannot be removed even if a defect extraction filter process that captures changes only in the X direction is executed.

The present invention has been made in view of the above points, and its object is to perform filter processing depending on the position of an object to be inspected at the time of imaging with an imaging device conforming to the standardization specifications. , to sufficiently obtain the effect of filtering.

In order to achieve the above object, a first invention provides an imaging device for an image inspection device that captures an image of an object to be inspected; A file in which register information indicating a place where a setting value is stored is described is externally obtained, and the setting value of each setting item set by the user and the register information corresponding to each setting item included in the file are obtained. and a setting device that transmits data indicating to the imaging device and sets the imaging device, wherein the file contains the captured image captured by the imaging device before outputting to the outside Also, as setting items related to the processing applied to the captured image, search settings for correcting the position of the object to be inspected in the captured image and the processing content based on the result of the position correction by the search settings are set. The setting device displays a user interface, setting contents for realizing the search setting and filter setting set by the user on the user interface, and the setting contents. can be transmitted to the imaging device, and the imaging device stores the setting content in the location indicated by the corresponding register information, and performs inspection based on the search setting for the captured image It is characterized in that it is configured to correct the position of the target object and execute filtering based on the result of the position correction according to the search settings.

According to this configuration, if both the imaging device and the setting device connected via the network conform to a common standard, the setting items of the imaging device and the setting values of each setting item are stored. The setting device acquires a file in which register information indicating the location where the image is to be captured is described, for example, from the imaging device. This file is, for example, a Device XML file when the standardization standard is the GenICam standard. This file may be obtained by the setting device from the imaging device, or may be downloaded from a website or the like and designated as a separate file to be referred to by GenApi.

When the user makes search settings and filter settings, the search settings and filter settings set by the user are transmitted to the imaging device together with the register information. As a result, the search setting and filter setting of the imaging device are performed, so that if the imaging device conforms to the standardization, the settings can be freely changed by the setting device, and the degree of freedom in selecting the model of the imaging device is improved. .

The imaging device stores the setting content in the location indicated by the corresponding register information, and performs position correction of the inspection object on the captured image based on the search setting. Correction of the position of the inspection object includes, for example, correction to rotate the inspection object and correction to move the inspection object in the X and Y directions. After the position is corrected by the search settings, the filter processing is executed based on the result of the position correction, so that the effect of the filter processing depending on the position when the inspection object is imaged can be sufficiently obtained.

A second aspect of the invention is characterized in that the imaging device is configured to output to the outside an image that has undergone filtering processing based on a result of position correction based on the search settings.

According to this configuration, after the imaging device executes the filtering process inside it, the image after the execution of the filtering process is output to the outside. The pass/fail judgment of the inspection object in the image after execution of the filter processing can be performed by an external pass/fail judging unit. It should be noted that the quality determination of the inspection object in the captured image may be performed inside the setting device without outputting the captured image to the outside. In this case, the quality determination result of the inspection object is output to the outside.

A third aspect of the invention is characterized in that the filter setting is a filter setting that exerts an effect depending on the directionality of the image.

Filter settings that exert effects dependent on the directionality of an image include, for example, X-direction defect extraction filter processing and Y-direction defect extraction filter processing settings. Since the result of such defect extraction filter processing strongly depends on the directionality of the image, the effect of performing the positional correction of the inspection object in advance becomes even more remarkable.

In a fourth aspect of the invention, the search setting is configured to search for an inspection object in an image captured by the imaging device, and perform position correction by rotating at least an area in which the inspection object is captured. It is characterized by

According to this configuration, by rotating at least the region in which the inspection object is shown, an image suitable for filtering that exhibits an effect dependent on the directionality of the image can be obtained.

A fifth aspect of the present invention is characterized in that the filter setting is a filter setting for extracting a portion where a change in luminance value between pixels adjacent to each other in a predetermined direction of the image is equal to or greater than a predetermined threshold value.

According to this configuration, when there is a background such as a striped pattern, for example, a defective portion of the inspection object is extracted based on a change in luminance value between pixels adjacent to each other in a predetermined direction different from the direction in which the striped pattern continues. and the background can be removed at this time. Therefore, the accuracy of defect detection is improved.

A sixth aspect of the invention is characterized in that the predetermined direction is a filter setting that can be selected from the X direction and the Y direction.

A seventh invention is characterized in that the predetermined direction is a filter setting that can be selected from the X direction and the Y direction. and acquires from the outside a file in which the setting items of the imaging device and the register information indicating the location where the setting value of each setting item is stored, and the setting value of each setting item set by the user. and an image inspection method for inspecting an object to be inspected by preparing a setting device for transmitting data indicating register information corresponding to each setting item included in the file to the imaging device and setting the imaging device. In the file, before the captured image captured by the imaging device is output to the outside, the position of the inspection object in the captured image is stored as a setting item related to the processing to be applied to the captured image. A search setting for performing correction and a filter setting for changing the processing content based on the result of position correction by the search setting are included, and a user interface is displayed on the setting device and used on the user interface. setting contents for realizing search settings and filter settings set by a user and register information indicating a location where the setting contents are stored are transmitted to the imaging device, and the imaging device corresponds the setting contents; In addition to performing positional correction of the inspection object on the captured image based on the search settings, filter processing is performed based on the result of the positional correction based on the search settings.

According to the present invention, filter processing can be performed based on the result of position correction after position correction of the inspection object in the captured image. The effect of treatment can be fully obtained.

1 is a diagram schematically showing the schematic configuration and operation state of an image inspection system according to Embodiment 1 of the present invention; FIG. 1 equivalent view according to Embodiment 2. FIG. 1 equivalent view according to Embodiment 3. FIG. 1 equivalent view according to Embodiment 4. FIG. 4 is a flow chart showing a procedure for generating an inspection image based on the principle of deflectometry; FIG. 16 is a view corresponding to FIG. 1 according to Embodiment 5 in which an inspection image is generated using a photometric stereo method; FIG. 4 is a diagram illustrating a connection interface between an image inspection application and a camera; 4 is a flowchart conceptually showing an example of internal processing of an imaging device; FIG. 10 is a diagram showing a parameter set for generating an inspection image using the principle of deflectometry; FIG. 4 is a diagram showing a parameter set for generating an inspection image by multispectral imaging; FIG. 4 is a diagram showing the configuration of a preprocessing module; FIG. 10 is a diagram for explaining how to identify a certain unit in the flowchart of internal processing of the imaging device; FIG. 10 is a diagram showing a setting user interface; FIG. 10 is a schematic diagram showing a case where setting user interfaces are switched by selecting a parameter set number; FIG. 4 is a diagram for explaining an internal mechanism for switching parameter sets by selecting a parameter set number; It is a figure which shows the processing example of the filter according to direction. FIG. 10 is a diagram illustrating a case where direction-specific filtering is performed after position correction; FIG. 10 is a diagram showing a user interface for filter setting; FIG. 10 is a diagram showing in a list format images that can be filtered. FIG. 10 is a diagram showing a detailed user interface for filtering; FIG. 4 illustrates the directional dependence of a filter; FIG. 10 is a flow chart for executing a directional filter after position correction; FIG. FIG. 10 is a flowchart for executing direction-specific filtering after position correction when a plurality of regions are set; FIG. 10 is a flow chart when following a moving object; FIG. 10 is a diagram showing an example of a form with dependencies; FIG. 25 is a view equivalent to FIG. 25 showing a state in which the order of the last category and register is reversed; FIG. 25 is a view equivalent to FIG. 25 showing a state in which the order of categories and registers one level higher is reversed; FIG. 10 is a diagram showing the appearance order of registers before and after rearrangement and the output form; FIG. 10 is a diagram showing an example of the appearance order of registers in LumiTrax mode; FIG. 29 is a view equivalent to FIG. 29 when the order of registers is not in the order of setting; FIG. 29 is a diagram equivalent to FIG. 29 when registers are rearranged;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail based on the drawings. It should be noted that the following description of preferred embodiments is essentially merely illustrative, and is not intended to limit the invention, its applications, or its uses.

(Embodiment 1)
FIG. 1 schematically shows an operation state of an image inspection system 1 according to an embodiment of the present invention. The image inspection system 1 is configured to inspect the presence or absence of defects in the work W using an image of the work W (inspection object). and a setting device 4 composed of a personal computer (PC) or the like for setting the imaging device 3 . The setting device 4 has a display section 5 , a keyboard 6 and a mouse 7 . The display unit 5 is composed of, for example, an organic EL display, a liquid crystal display, or the like, and displays an image captured by the imaging device 3, an image after processing the image captured by the imaging device 3, various user interfaces (GUIs, etc.). ) can be displayed. Various user interfaces and the like are generated by the main body of the setting device 4 . The horizontal direction of the display portion 5 can be the X direction of the display portion 5 , and the vertical direction of the display portion 5 can be the Y direction of the display portion 5 .

The keyboard 6 and mouse 7 are conventionally well-known devices (operating means) for operating a computer. By operating the keyboard 6 or the mouse 7, various information can be input to the setting device 4, and an image or the like displayed on the display section 5 can be selected. Instead of the keyboard 6 and the mouse 7, or in addition to the keyboard 6 and the mouse 7, for example, a device for computer operation such as a voice input device and a pressure-sensitive touch operation panel can be used.

For example, the setting device 4 is connected to an external control device 8 such as a programmable logic controller (PLC). This external control device 8 may constitute a part of the image inspection system 1 . Also, the external control device 8 does not have to be a component of the image inspection system 1 .

FIG. 1 shows a case where a plurality of works W are placed on the upper surface of a conveying belt conveyor B and conveyed in the direction indicated by the white arrows in FIG. The external control device 8 is a device for sequence-controlling the transport belt conveyor B and the image inspection system 1, and can use a general-purpose PLC.

In the description of this embodiment, the conveying direction of the work W by the conveying belt conveyor B (moving direction of the work W) is defined as the Y direction, and the direction orthogonal to the Y direction in plan view of the conveying belt conveyor B is the X direction. , and the direction orthogonal to the X and Y directions (the direction orthogonal to the upper surface of the conveyor belt B) is defined as the Z direction, but this is only defined for convenience of explanation.

The image inspection system 1 can be used to inspect the appearance of the work W, that is, to inspect the surface of the work W for defects such as scratches, stains, and dents. Acceptance/rejection judgment can also be performed. During operation, the image inspection system 1 receives an inspection start trigger signal that defines the start timing of the defect inspection (good/bad judgment inspection) from the external control device 8 via the signal line. The image inspection system 1 picks up an image of the workpiece W and illuminates it based on the inspection start trigger signal, and obtains an inspection image after predetermined processing. After that, the appearance is inspected based on the inspection image, and the inspection result is transmitted to the external control device 8 via the signal line. In this manner, when the image inspection system 1 is operated, the input of the inspection start trigger signal and the output of the inspection result are repeatedly performed between the image inspection system 1 and the external control device 8 via the signal line. Note that the input of the inspection start trigger signal and the output of the inspection result may be performed via the signal line between the image inspection system 1 and the external control device 8 as described above, or other signals (not shown). You can go through the line. For example, a sensor for detecting the arrival of the work W may be directly connected to the image inspection system 1, and an inspection start trigger signal may be input to the image inspection system 1 from the sensor. In addition, the image inspection system 1 may be configured to automatically generate a trigger signal internally for inspection.

Further, the image inspection system 1 may be configured by dedicated hardware, or by installing software in general-purpose equipment, for example, by installing an image inspection program in a general-purpose or dedicated computer. For example, an image inspection program may be installed in a dedicated computer in which hardware such as a graphic board is specialized for image inspection processing.

(Configuration of illumination device 2)
The lighting device 2 includes a light emitting section 2a and a lighting control section 2b that controls the light emitting section 2a. The light emitting unit 2a and the illumination control unit 2b may be separate or integrated. Also, the illumination control unit 2 b may be incorporated in the setting device 4 . The light-emitting unit 2a can be composed of, for example, a light-emitting diode, a projector using a liquid crystal panel, an organic EL panel, a digital micromirror device (DMD), or the like, and can also be called an illumination unit. Light-emitting diodes, liquid crystal panels, organic EL panels, and DMDs are not shown, but conventionally known structures can be used. The lighting device 2 is connected to the setting device 4 via a signal line 100a, and can be installed separately from the imaging device 3 and the setting device 4. FIG.

The illumination device 2 of Embodiment 1 is configured to perform uniform surface emission. Further, the illumination device 2 is configured to perform illumination capable of realizing the deflectometry process, and therefore has a light emitting unit 2a that irradiates the workpiece W with pattern light having a periodic illumination distribution. are doing. In other words, the illumination device 2 can be a pattern light illumination section that performs pattern light illumination by sequentially irradiating the workpiece W with a plurality of different pattern lights. The illumination device 2 used when generating an inspection image by performing a deflectometry process will be described below.

When a plurality of light emitting diodes are used, pattern light having a periodic illuminance distribution can be generated by arranging the plurality of light emitting diodes in a dot matrix and controlling the current value. For example, in the case of Y-direction pattern light in which the brightness changes in the Y direction, it can also be expressed as pattern light in which striped patterns are repeated in the Y direction. is shifted in the Y direction, it is possible to generate a plurality of Y-direction pattern lights with different phases of the illuminance distribution. The illuminance distribution of the Y-direction pattern light can also be represented by a waveform approximating a sinusoidal waveform. Pattern light, Y-direction pattern light at 180°, and Y-direction pattern light at 270° can be generated.

In addition, in the case of X-direction pattern light in which the brightness changes in the X direction, it can be expressed as pattern light in which a striped pattern is repeated in the X direction. is shifted in the X direction, it is possible to generate a plurality of X-direction pattern lights with different phases of the illuminance distribution. The illuminance distribution of the X-direction pattern light can also be represented by a waveform approximating a sinusoidal waveform. Pattern light, X-direction pattern light at 180°, and X-direction pattern light at 270° can be generated. That is, the illumination device 2 can illuminate the workpiece W in different illumination modes. In the case of performing the deflectometry process, the pattern light irradiated onto the workpiece W can be not only sinusoidal wave but also pattern light such as triangular wave.

It should be noted that it is also possible to irradiate light with a uniform illuminance distribution in the plane by applying a current having the same current value to all the light emitting diodes. If the same current value is applied to all light-emitting diodes and changed, the light-emitting state can be changed from a dark surface-emitting state to a bright surface-emitting state.

In the case of a liquid crystal panel and an organic EL panel, by controlling each panel, the light emitted from each panel can be made to be patterned light having a periodic illuminance distribution. In the case of a digital micromirror device, it is possible to generate and irradiate patterned light having a periodic illuminance distribution by controlling the built-in micromirror surface. The configuration of the illumination device 2 is not limited to that described above, and any device or device that can generate pattern light having a periodic illuminance distribution can be used.

Also, as will be described later, the illumination device 2 used when generating an inspection image using the photometric stereo method can use an illumination device capable of individually irradiating illumination from at least three or more different directions. Moreover, the illumination device 2 may be an illumination device configured to enable multispectral illumination, and the configuration of the illumination device 2 is not particularly limited.

(Configuration of imaging device 3)
The imaging device 3 has a camera 31 , a condensing optical system 32 , and an imaging control section 33 . The imaging device 3 is connected to the setting device 4 via a signal line 100b, and can be installed separately from the lighting device 2 and the setting device 4. FIG.

The camera 31 has an image sensor made up of an imaging device such as a CCD or CMOS that converts the intensity of light obtained through the condensing optical system 32 into an electrical signal. The imaging control unit 33 has a storage device and a signal processing device, and is a part that controls the start and end of exposure of the camera 31, controls the exposure time, adjusts the analog gain, and the like. An internal logic circuit can control the driving of the image sensor and the transfer of image data. Various image processing, filter processing, and the like can be performed by the imaging control unit 33, and the imaging device 3 is a device having a filter processing function.

The condensing optical system 32 is an optical system for condensing light incident from the outside, and typically has one or more optical lenses. In addition, the condensing optical system 32 is configured to be able to perform autofocus. The positional relationship between the imaging device 3 and the illumination device 2 is set so that the light emitted from the illumination device 2 toward the surface of the work W is reflected by the surface of the work W and enters the condensing optical system 32. can do. When the work W is a translucent member such as a transparent film or sheet, the pattern light emitted from the illumination device 2 toward the surface of the work W passes through the work W and is condensed by the imaging device 3. The positional relationship between the imaging device 3 and the illumination device 2 can be set so that the light enters the system optical system 32 . In any of the above cases, the imaging device 3 and the illumination device 2 are arranged so that the specular reflection component and the diffuse reflection component reflected by the surface of the workpiece W enter the condensing optical system 32 of the imaging device 3 . A line camera in which light receiving elements are arranged linearly in the Y direction can be used as the imaging device 3. In the case of this area camera, a form of coaxial illumination is also possible.

(Embodiment 2)
FIG. 2 is a diagram schematically showing the schematic configuration and operation state of an image inspection system according to Embodiment 2 of the present invention. In the second embodiment, the imaging control section 33 is separated from the camera 31 having an imaging device and the condensing optical system 32, and the illumination control section 2b is included in the imaging control section 33 and integrated. The imaging control unit 33 is connected to the setting device 4 via the signal line 100b. The imaging control section 33 and the illumination control section 2b may be separate bodies.

(Embodiment 3)
FIG. 3 is a diagram schematically showing the schematic configuration and operation state of an image inspection system according to Embodiment 3 of the present invention. In the third embodiment, the illumination control section 2b is integrated into the imaging control section 33, and the display section 5 and the mouse 7 are connected to the imaging control section 33. FIG. The setting device 4 is composed of, for example, a notebook computer or the like, and is connected to the imaging control section 33 . The setting device 4 has a keyboard 6 . A mouse may also be connected to the setting device 4 . Various user interfaces and the like generated by the setting device 4 can be displayed on the display section 5 via the imaging control section 33 .

(Embodiment 4)
FIG. 4 is a diagram schematically showing the schematic configuration and operation state of an image inspection system according to Embodiment 4 of the present invention. In Embodiment 4, the illumination device 2 and the imaging device 3 are integrated, and the illumination control section 2b and the imaging control section 33 are also integrated. In Embodiments 2 to 4, the same parts as in Embodiment 1 are denoted by the same reference numerals, and description thereof is omitted.

(deflectometry processing)
In the first to fourth embodiments, the phase data of the surface of the work W is generated based on the principle of deflectometry from a plurality of luminance images captured by the imaging device 3, and the shape of the work W is inspected based on the phase data. It is configured to perform a deflectometry process to generate an image for use.

Generating an inspection image will be described in detail below according to the flowchart shown in FIG. At step SA1 of the flowchart, the illumination device 2 generates pattern light according to predetermined illumination conditions to illuminate the work W. FIG. The illumination conditions at this time are Y-direction pattern light when the phase is 0°, Y-direction pattern light when the phase is 90°, Y-direction pattern light when the phase is 180°, Y-direction pattern light when the phase is 270°, 0 Illumination conditions can be such that the X-direction pattern light at 90°, the X-direction pattern light at 180°, and the X-direction pattern light at 270° are generated in sequence. . Therefore, the illumination device 2 generates a total of eight types of pattern light (four types in the X direction and four types in the Y direction) and sequentially irradiates the workpiece W with the pattern light. The imaging device 3 images the workpiece W each time the pattern light is irradiated. When the imaging device 3 takes an image, since the pattern light is emitted eight times, eight luminance images are obtained. Depending on the shape of the workpiece W, only the patterned light in the X direction or only the patterned light in the Y direction may be irradiated. Also, the number of pattern lights may be three.

After that, the process proceeds to step SA2, and the eight luminance images obtained in step SA1 are subjected to the deflectometry process. By capturing an image of the workpiece W irradiated with the pattern light, each pixel value of the brightness image obtained includes a specular reflection component and a diffuse reflection component (+environmental component). Since the phase of the illuminance distribution of the pattern light is shifted by 90° (π/2) in the X direction and imaged four times, four types of pixel values are obtained by irradiating the pattern light in the X direction. The diffuse reflection component, the specular reflection component, and the specular reflection angle (phase) are phase data and can be calculated by well-known formulas.

Similarly, the phase of the illuminance distribution of the pattern light is shifted by 90° (π/2) in the Y direction and the image is captured four times, so four types of pixel values are obtained by irradiating the Y direction pattern light. The diffuse reflection component, specular reflection component, and specular reflection angle (phase) can be similarly calculated for the Y direction.

For example, a specular reflection component image can be obtained by eliminating the diffusion component by the difference between the opposite phases. They are obtained in the X direction and the Y direction, respectively, and a regular reflection component image is obtained by synthesizing them. Also, the specular reflection angle is obtained by calculating the angle as tan θ=sin θ/cos θ from the specular reflection component shifted by π/2. Also, the average image contains a diffuse component and an environmental component, and is obtained by adding the opposite phases to eliminate the regular reflection component.

At step SA3, contrast correction is performed on the diffuse reflection component image. Further, in step SA4, contrast correction is performed on the regular reflection component image. Each contrast correction can be a linear correction. For example, correction is made so that the average of ROIs becomes the median. In the case of 8 bits, 128 levels should be used as the median value. As a result, the diffuse component after correction and the regular reflection component after correction are obtained.

Further, in step SA5, the difference between the phase component and the reference phase is obtained. At this step SA5, a difference is obtained with respect to the phase of the reference plane. For example, the user designates a spherical shape, a cylindrical shape, a planar shape, etc. as the reference plane, and acquires the difference from these. Alternatively, it may be extracted using a free-form surface. A corrected phase (difference) is obtained for the X direction, and a corrected phase is also obtained for the Y direction. The diffuse reflection component image, specular reflection component image, and reference phase difference image can be output images.

At step SA6, a layered image and a depth contour image are obtained based on the reference phase difference image obtained at step SA5. A layered image is an image that has been repeatedly reduced to 1/2. As a result, layered phase images are obtained in the X and Y directions, respectively.

On the other hand, the depth contour image is an output image in which a portion with a large phase difference is emphasized, and is a concept different from curvature. The depth contour image has the advantage that it can be obtained at a higher speed than the shape image obtained by accumulating shapes, the line flaws on the work W are extremely easy to see, and the contour can be easily extracted.

Next, in step SA7, shape stacking is performed on the layered phase images to generate a shape image. A shape image can be obtained by carrying out accumulation calculation using the Gauss-Jacobi method or the like for the regular reflection angles in the X direction and the Y direction. The shape image becomes the output image.

In general, there are many examples in which the shape is restored by triangulation after unwrapping, but in this embodiment, unwapping is avoided, and local differential values are accumulated using the Gauss-Jacobi method. Thus, the shape is restored without relying on triangulation. A known method can be appropriately used as the shape restoration method. A method that does not use triangulation is preferred. Also, a hierarchical method having multiple stages of reduced images can be used. Alternatively, a method of having a difference between a reduced image and a normal image may be used.

Furthermore, the feature size can also be set as a parameter. The feature size is a parameter for setting the size of a flaw to be detected according to the purpose and type of inspection. For example, when the feature size parameter value is 1, the finest flaws can be detected, and as this value is increased, larger flaws can be detected. As a result, when the feature size is increased, a larger flaw can be easily detected, and flaws and unevenness on the surface of the workpiece W become clear.

At step SA8, simple defect extraction is performed. That is, after generating an image for easily inspecting line scratches (defects) on the work W, an image for easily inspecting dirt (defects) on the work W, and an image for easily inspecting dents (defects) on the work W, The defects that appear are displayed on the defect extraction image. The defect extraction image is an image in which the defects appearing in each image can be extracted and displayed, and the extracted defects can be combined into one image and displayed, so it can also be called a composite defect image. .

For example, the specular reflection component image is an image that facilitates confirmation of dirt that dulls specular reflection, scratches that do not change shape but dull specular reflection, and scratches that do not return specular reflection due to change in shape. The diffuse reflection image is an image from which it is easy to confirm the state of the texture of the surface of the work W (specifically, characters on printed matter, dark stains, etc.). A shape image is an image obtained by accumulating changes in phase while looking at surrounding pixels according to the feature size. Here, if the characteristic size of the shape image is set large, it is possible to capture relatively shallow and wide-area concavities and convexities among the shape changes. On the other hand, if the feature size of the shape image is set small, it is possible to capture line flaws and small-area flaws. Also, defects that are difficult to appear in the shape image (for example, thin scratches and deep scratches) tend to appear in the specular reflection component image. Also, the depth contour image is obtained by calculating a reference plane and imaging the deviation from this reference plane. Line flaws and small-area flaws can be captured from the depth contour image.

A defect extraction image displaying the defects extracted after performing the simple defect extraction in step SA8 is output.

(Embodiment 5)
FIG. 6 is a diagram schematically showing the schematic configuration and operation state of an image inspection system according to Embodiment 5 of the present invention. In the first to fourth embodiments, the inspection image is generated based on the principle of deflectometry, whereas in the fifth embodiment, the inspection image is generated using the photometric stereo method. It is configured.

Hereinafter, a specific method for generating an inspection image using the photometric stereo method will be described with reference to FIG. will be described in detail.

The image inspection system 1 according to the fifth embodiment can be configured in the same manner as the image inspection system disclosed in Japanese Unexamined Patent Application Publication No. 2015-232486, for example. That is, the image inspection system 1 includes an imaging device 3 for imaging the workpiece W from a fixed direction, and an illumination device 200 for illuminating the workpiece W from three or more different illumination directions. At least a display unit 5 , a keyboard 6 and a mouse 7 are provided.

The lighting device 200 is configured to irradiate the workpiece W with light from mutually different directions, and controls the first to fourth light emitting units 201 to 204 and the first to fourth light emitting units 201 to 204. and a lighting control unit 205 that The illumination device 200 is a part that performs multi-directional illumination by irradiating the work W with light from mutually different directions. The first to fourth light emitting units 201 to 204 are arranged so as to surround the workpiece W with a space between them. A light-emitting diode, an incandescent bulb, a fluorescent lamp, or the like can be used for the first to fourth light-emitting units 201-204. Also, the first to fourth light emitting units 201 to 204 may be separate bodies, or may be integrated.

In this embodiment, the first to fourth light emitting units 201 to 204 are sequentially lit, and the imaging device 3 images the work W when any one of the first to fourth light emitting units 201 to 204 is lit. For example, when lighting device 200 receives a first lighting trigger signal, lighting control unit 205 turns on only first light emitting unit 201 . At this time, the imaging device 3 receives the imaging trigger signal and images the work W at the timing when the light is irradiated. When the illumination device 200 receives the second illumination trigger signal, the illumination control unit 205 turns on only the second light emitting unit 202, and the imaging device 3 images the workpiece W at this time. Thus, four luminance images can be obtained. The number of lights is not limited to four, but may be any number as long as it is three or more and can illuminate the work W from mutually different directions.

Then, the normal vector of each pixel to the surface of the workpiece W is calculated using the pixel value of each pixel having a corresponding relationship in the plurality of luminance images captured by the imaging device 3 . Differential processing is performed on the calculated normal vector of each pixel in the X direction and the Y direction to generate a contour image showing the contour of the inclination of the surface of the workpiece W. FIG. Also, from the calculated normal vector of each pixel, the albedo of the same number of pixels as the normal vector is calculated, and from the albedo, a texture rendering image showing a pattern with the inclination state of the surface of the workpiece W removed is generated. . Since this method is a well-known method, detailed description is omitted. A shape image showing the shape of the workpiece W can be generated by synthesizing a plurality of luminance images based on the principle of photometric stereo.

(multispectral lighting)
As another embodiment, the illumination device 2 may be capable of multispectral illumination. Multispectral illumination is to irradiate the work W with light of different wavelengths at different timings, and is suitable for inspecting printed matter (inspection object) for color unevenness, stains, and the like. For example, the illumination device 2 can be configured so as to sequentially irradiate the workpiece W with yellow, blue, and red. The illumination device 2 may be configured with a panel, an organic EL panel, or the like.

The image capturing device 3 captures images of the workpiece W at the timing when the light is irradiated to obtain a plurality of luminance images. Then, an inspection image can be obtained by synthesizing a plurality of luminance images. This is called multispectral imaging. A shape image showing the shape of the workpiece W can be generated by synthesizing a plurality of luminance images by multispectral imaging. In addition, ultraviolet rays and infrared rays can also be included in the light to be irradiated.

(Connection interface between camera and image inspection application)
The camera 31 is a GenICam standard compatible camera compatible with the GenICam standard. The GenICam standard is a standard for standardizing a connection interface between a PC application and the camera 31. As shown in FIG. , and an interface for acquiring an image captured by the camera 31 with the image inspection application 40 of the setting device 4 is standardized. If both the imaging device 3 and the image inspection application 40 of the setting device 4 support the GenICam standard, the camera 31 and the image inspection application 40 of the setting device 4 can be connected. The image inspection application 40 is composed of software installed in a personal computer. In this embodiment, the case where the camera 31 and the setting device 4 are compatible with the GenICam standard as the standardization standard will be described, but the standardization standard is not limited to the GenICam standard, and other standardization standards good.

Assuming that the image inspection application 40 is decomposed into a hierarchical structure, the GenICam layer 41 consists of a higher layer (inspection unit 4A) that actually performs image inspection, defect inspection, etc. in the image inspection application 40, and specific network communication. It is positioned as an intermediate layer positioned between the layer 42 that performs control based on the standard. The GenICam layer 41 can be roughly classified into two parts, GenApi 41a and GenTL (TL: Transport Layer) 41b. The GenApi 41a is a part that converts setting items of the camera 31 and register addresses inside the camera 31 . With this GenApi 41a, the image inspection application 40 does not specify a specific address of the camera 31, but abstractly uses a character string such as "ExposureTime" for the exposure time and "AnalogGain" for the analog gain as an argument. , and by analyzing a file called DeviceXML (details will be described later) obtained from the camera 31 when connecting to the camera 31, the register address corresponding to the character string (called Feature) is obtained. can be allocated.

The GenTL 41b defines an interface (API) for controlling data transfer between the image inspection application 40 and the camera 31. Specifically, the specification of writing to and reading from the registers of the camera 31 and the specifications of the camera 31 defines API specifications for transferring image data transferred from 31 to a higher layer of the image inspection application 40 .

The physical standard for connecting the image inspection application 40 and the camera 31 may be any standard that uses a high-speed network. There is a USB3Vision standard 3B. Therefore, the imaging device 3 and the setting device 4 are connected via a network, and this network may be wired using a high-speed network cable or wireless. Also, the network may be a network using cables other than the Ethernet cable 31a and the USB 3.0 compatible cable 31b, and is not particularly limited.

The GenICam standard does not specifically specify what is to be used as a physical communication standard, and merely defines an abstract specification in the form of GenTL41b. As a layer 42 below GenTL 41b, communication standards are defined using specific communication networks, such as GigE Vision standard 3A and USB 3 Vision standard 3B. Specific communication standards are not limited to the GigE Vision standard 3A and USB3Vision standard 3B, and may be compatible with the GenICam standard. In FIG. 7, in order to explain the concept of GigE Vision standard 3A and USB3Vision standard 3B, cameras 31, 31 corresponding to each standard are shown connected to the setting device 4, but only one camera 31 is set. It can be used by connecting to the device 4 .

The GenICam standard compatible camera 31 stores a file (DeviceXML file) 31c called DeviceXML. The DeviceXML file 31 c is stored in a storage device (internal memory) built into the camera 31 . The image inspection application 40 compatible with the GenICam standard reads the DeviceXML file 31 c from the camera 31 when connecting to the camera 31 to be set by the image inspection application 40 . For reading the DeviceXML file 31c, there is a method of downloading the DeviceXML file 31c from the camera 31, for example. The downloaded DeviceXML file 31c is held in GenApi 41a. The DeviceXML file 31c can be downloaded by request from the image inspection application 40 or automatically when the camera 31 is connected. can be obtained. Alternatively, without downloading from the camera 31 to be connected, the DeviceXML file 31c corresponding to the camera 31 may be obtained by another means (for example, downloaded from a website) and specified to the GenApi 41a at the time of connection. be.

In the DeviceXML file 31c, all setting items held inside the camera 31 and register addresses (register information) in which setting values of the setting items are stored are described in association with each other. Setting items are called Features, and each Feature is assigned a character string for identifying the individual Feature. Each Feature node describes a reference to a node in which a specific register address is described. The register information also includes register addresses and character strings that identify registers.

In the DeviceXML file 31c, for example, there is a Feature named "ExposureTime" (exposure time setting), and its attribute is instructed to refer to a register address "ExposureTimeReg". Suppose it was written. If the camera 31 is different, the address may have a different value, but the Feature name "ExposureTime" is common. In this way, by managing setting items common to many cameras 31 under a unified name, the image inspection application 40 can set the camera 31 without being conscious of differences in the models of the cameras 31 and the differences in manufacturers. 31 can be set and controlled.

As described above, in the GenApi 41a layer, by analyzing the description content of the DeviceXML file 31c, the Feature character string passed as an argument from the image inspection application 40 in the upper layer is converted into a register address. For example, when writing a value (value: 100.5) in a register corresponding to a Feature called "ExposureTime", two values, an address value and a value to be written, are written, such as WriteRegister (address value, 100.5). By executing the function with as an argument, the register value of the camera 31 can be set via a physical communication standard such as GigE Vision standard 3A or USB3 Vision standard 3B.

Note that the DeviceXML file 31c defines what should be included as common features for each manufacturer, but it is also possible to define other vendor-specific features. In the case of the camera 31 having highly unique functions, it is possible to use special functions that do not exist in general-purpose cameras by accessing the camera 31 through dedicated Features.

(Internal processing unit of imaging device 3)
FIG. 8 is a flowchart conceptually showing an example of internal processing of the imaging device 3 . In this flowchart, the group for position correction 1 is described as being composed of the pattern search unit and the position correction unit included in the group, but the groups for position corrections 2 to 5 are not described. This is because the groups of position corrections 2 to 5 have been folded to simplify the drawing.

As shown in this flow chart, it is composed of a combination of a plurality of units. A unit is a unit for controlling imaging and image processing, and by combining the units on the flowchart, the imaging device 3 can be made to perform a desired operation. In the flowchart shown in FIG. 8, the unit is described as one step.

For example, when a function for performing certain processing is enabled by user settings, the processing can be made executable by adding a unit corresponding to that function. A unit can be defined as a combination of a program for executing a process, a parameter necessary for executing the process, and a storage area for storing the data of the process result. Each process can be performed by the imaging control unit 33 shown in FIG. 1, and a storage area can be secured in the storage device of the imaging control unit 33. It should be noted that the concept of the unit itself is not essential when realizing the external specifications of the camera 31 compatible with the GenICam standard.

The flowchart shown in FIG. 8 is obtained by arranging a plurality of units vertically and horizontally in a flowchart format, and can be simply called a flow. A flowchart in which a plurality of units are arranged only in the vertical direction may also be used. Although the processing is sequentially executed from the start to the end of the flowchart shown in FIG. 8, branching can be performed by providing a branch step SB1 in the middle. In the case of branching, a merging step SB2 can be provided until the end, whereby the merging can proceed to the end.

(parameter set)
When the image inspection application 40 is operated in an actual inspection environment, the setting parameters of the imaging device 3 can be dynamically changed when the work W is switched or when a change in the surrounding environment such as brightness is detected. be. If only very limited parameters such as exposure time are to be changed, the image inspection application 40 can directly write the corresponding Feature values.

On the other hand, in the case of the high-performance imaging device 3, the number of items that can be set increases, and the number of parameters that need to be changed at once when the work W is switched also increases. In this case, the time required to change the settings increases in correlation with the number of parameters. In the actual image inspection line, the work W is switched, and the time from when the new work W reaches the imaging field range of the imaging device 3 to when it leaves the field range is often short, so a series of setting changes can be performed at high speed. I have a case I want to do. In such cases, a function called parameter set may be used.

A parameter set is such that a plurality of patterns of combinations of various parameters for imaging with the imaging device 3 are stored in advance, and each pattern can be managed by a parameter set number. For example, if there are three types of workpieces W and it is desired to perform imaging and subsequent processing with different parameters for each type, three parameter sets are prepared.

By using parameter sets, a series of setting changes can be completed in a short period of time by simply designating the parameter set number corresponding to the type of workpiece W to be imaged next before imaging the workpiece W. From the viewpoint of the image inspection application 40, the Feature specifying the parameter set number can be considered as a type of selector, which will be described later.

The selector for switching the target to be set from the image inspection application 40 side and the register for switching the parameter internally referred to by the imaging device 3 during operation can be made independent or the same. The upper limit of the number of parameter sets that can be set is limited to the parameter holding space of the internal memory provided inside the imaging device 3 .

The concept of parameter sets can also be expanded to the imaging device 3 having a filtering function. For example, the parameter Assume that a parameter corresponding to the set Index has been set. By doing so, it is possible to dynamically switch between the imaging and the content to be executed as the filtering process by the parameter set Index.

The GenICam standard supports a function of dynamically switching imaging parameters. The imaging device 3 is configured so that all setting parameters can be accessed by Features in accordance with the GenICam standard. The imaging function includes a function of synchronizing the lighting device 2 and the imaging device 3 and synthesizing images captured in a plurality of patterns by a dedicated algorithm (generating an inspection image using the principle of the above-described deflectometry), It has multiple illumination-imaging control modes, including a multispectral imaging function that acquires multiple images by irradiating different lights. These can be set for each parameter set described above, and it is possible to execute imaging processing, filter processing, synthesis processing, and image output while dynamically switching according to conditions. The image generated by the imaging function may be the image itself acquired by the image sensor while switching the lighting pattern of the illumination, or may include a plurality of images synthesized by the dedicated algorithm described above. .

For filtering, multiple patterns can be set within the same parameter set. For example, in the imaging function described above, multiple patterns of images are generated by performing a series of imaging operations, and it is possible to individually apply filter processing to the different images generated. . In another pattern, it is also possible to set a specific area range (ROI: Region Of Interest) for the same image, and then filter only the inside of each area. The range of the specific area can be set by operating the keyboard 6 or the mouse 7, for example. The specific area may be one or plural. The size of the specific area can be set arbitrarily.

The filter processing may be a multistage filter that can repeatedly set a plurality of types in multiple stages for the same image. For example, dilation filtering may be performed on an image followed by binarization filtering on the image. The filtering process is not limited to multistage filtering, and may be one filtering process.

Using the function of dynamically switching the imaging parameters of the GenICam standard, a plurality of parameter sets can be combined into one flowchart as shown in FIG. In the internal processing flowchart of the imaging device 3, a unit group forming a flowchart from branch step SB1 to merging step SB2 is collectively called a parameter set. The example shown in FIG. 8 has four parameter sets, that is, first to fourth parameter sets. In other words, there are multiple patterns of combinations of multiple selector values for realizing the processing set by the user. The user can set which of the first to fourth parameter sets is selected. For example, when parameter set number 1 is selected, the first parameter set is automatically selected.

After the start of the flow chart shown in FIG. 8, after going through each unit constituting any one of the first to fourth parameter sets in the branch step SB1, they merge in the merging step SB2 to proceed to the end. If parameter set number 1 is selected, the branch number becomes 1 in branch step SB1, and each process of the first parameter set is executed. If the parameter set number 2 is selected, the branch number becomes 2 in the branch step SB1, and each process of the second parameter set is executed. If the parameter set number 3 is selected, the branch number becomes 3 in the branch step SB1, and each process of the third parameter set is executed. If parameter set number 4 is selected, the branch number becomes 4 in branch step SB1, and each process of the fourth parameter set is executed. The number of parameter sets is not limited to four and can be set arbitrarily.

Specific examples of parameter sets are shown in FIGS. 9 and 10. FIG. FIG. 9 shows a parameter set for generating an inspection image using the principle of deflectometry. In order to generate an inspection image using the principle of deflectometry, the unit UA1 executes a plurality of imaging processes with a single trigger signal, the unit UA2 executes dilation filter processing, and the unit UA3 executes an averaging filter. processing, and the unit UA4 executes the grayscale reversal processing. That is, multistage processing defined by a parameter set including filtering is sequentially performed on the captured image captured by the unit UA1. After that, in the unit UA5, the image data that has undergone the multistage processing is transferred to the PC, that is, output to the setting device 4 or the like, which is an external device. Note that the image data may be held internally without being transferred. Depending on the parameter set, it can also be set so that multistage processing is not performed. When transferred to the setting device 4, the inspection unit 4A of the image inspection application 40 shown in FIG. 7 can perform defect inspection and pass/fail judgment. Algorithms for defect inspection and pass/fail judgment can be conventionally well-known.

On the other hand, FIG. 10 is a parameter set for generating an inspection image by multispectral imaging. Unit UB1 executes a plurality of imaging processes with a single trigger signal in order to generate an inspection image by multispectral imaging, and unit UB2 executes binarization filter processing for a certain region (region 0). , in unit UB3, dilation filtering is performed on a region (region 1) different from region 0; That is, in this example as well, multi-stage processing defined by a parameter set including filtering is sequentially performed on the captured image captured by the unit UB1. After that, unit UB4 does not execute the process. The unit UB5 transfers the image data to the PC as in the case shown in FIG. As in this example, a parameter set may contain units that do not perform processing, i.e., disabled units. may

Therefore, the imaging device 3 can sequentially perform multi-step processing on the captured image as set by the setting device 4, so that the user can freely set the sequential procedure. This allows the imaging device 3 to perform complicated image processing. It should be noted that it is also possible to reflect non-sequential settings such as the exposure time.

(type of unit)
There are multiple types of units, for example, a pattern search unit that performs pattern search processing to determine an inspection area, a position correction unit, an image operation unit that internally performs image position correction, color extraction, filter processing, etc. There is also a highly functionalized unit that combines these relatively simple processing units. Each unit is a unit for executing processing applied to a captured image captured by the imaging device 3 before outputting the captured image to the outside. can be arranged to perform multi-stage processing on

The pattern search unit is a unit for searching the workpiece W and the portion of the workpiece W to be inspected from the image including the workpiece W captured by the imaging device 3, and correcting the position of the workpiece W in the captured image. be. For example, when the image inspection system 1 is set, edge detection is performed by a well-known edge detection method on an image of the work W, and the area specified by the detected edge is the work W or the portion of the work W to be inspected. It can be stored in the storage device of the imaging device 3 as a model (search model image). For the edge detection process itself, a conventionally known method can be used. For example, the pixel value of each pixel on the luminance image is obtained, and the change in the pixel value on the luminance image is equal to or greater than the threshold value for edge detection. If an area exists, its boundary is extracted as an edge. The threshold for edge extraction can be arbitrarily adjusted by the user.

After setting the image inspection system 1, when the image inspection system 1 is operated, the workpieces W successively conveyed are imaged to obtain an inspection image, and the pattern search unit detects the workpiece W or the inspection image on the obtained inspection image. A search process is performed based on the stored model to determine whether or not there is an inspection target portion of the work W, and the position and angle of the work W are measured by the search process. The position of the work W can be specified by the X and Y coordinates. Also, the angle of the workpiece W can be an angle around the optical axis of the imaging device W or an angle around the Z-axis shown in FIG.

The position correction unit searches the image including the workpiece W imaged by the imaging device 3 with the pattern search unit for the workpiece W and the inspection target portion of the workpiece W, and after measuring the position and angle of the workpiece W, This is a unit for correcting the position of the work W in the image. When the image inspection system 1 is operated, a plurality of works W are not always conveyed in the same position and orientation, and works W in various positions and in various orientations may be conveyed. . Since the reference portion of the workpiece W can be searched by the pattern search unit and then positionally corrected by the position correction unit, the reference portion of the workpiece W is always at a fixed position, and the workpiece W is positioned at a predetermined position. Position correction is performed by rotating the image or moving the image in the vertical or horizontal direction so as to obtain the posture. A plurality of types of tools such as pattern search can be prepared as position correction tools for performing position correction.

There are a plurality of types of image calculation units, for example, a unit that performs filter processing, a unit that performs synthesis processing for synthesizing a plurality of images captured by the imaging device 3, a unit that performs deflectometric processing, and a photometric stereo method. There are a unit that generates an inspection image using , a unit that performs multispectral imaging, and the like. Since there are a plurality of types of filtering, it is possible to provide a plurality of types of units for performing filtering, such as binarization filters and expansion filters. Since generation of an inspection image by deflectometry processing involves a plurality of processes as described above, a unit may be provided for each process. A unit or the like for generating a reference phase difference image may be provided.

(pretreatment module)
FIG. 11 shows an example of a pretreatment module. The preprocessing module can be a combination of a group consisting of a pattern search unit and a position correction unit, and an image operation unit, but this is an example, and a preprocessing module consisting of only one group. Alternatively, it may be a pretreatment module in which any other unit is combined. The pretreatment modules are numbered and can be distinguished from "pretreatment module 0", "pretreatment module 1", and the like.

Multiple preprocessing modules can be added or set in order to enhance the extensibility of the function, and each is indexed, for example, "PreprocessingModule0" and "PreprocessingModule1". identified.

(selector)
Multiple preprocessing modules can be added to any one parameter set of the flow chart shown in FIG. For example, an arbitrary preprocessing module can be added according to setting parameters for executing pattern search processing and setting parameters for executing filtering processing.

However, if each preprocessing module is provided with individual setting registers, the same register (Feature) appears repeatedly, resulting in redundancy. In this embodiment, the selector is used to dynamically switch the editing target (setting target). Setting objects that can be switched by the selector include, for example, a parameter set (branch number), a preprocessing module number, the number of filter stages in the case of a multi-stage filter (which stage filter is set), and the like. That is, the Feature that specifies the parameter set number, the Feature that specifies the preprocessing module number, and the Feature that specifies the number of filter stages are types of selectors. This selector value is included in the DeviceXML file along with register information indicating where the selector value is stored. Therefore, before outputting a captured image captured by the imaging device 3 to the outside, a plurality of selectors associated with setting items for setting multi-step processing to be applied to the captured image, and The DeviceXML file will contain register information indicating where the values are stored.

(multistage selector)
As shown in FIG. 8, the flowchart branches horizontally according to parameter sets, and within each parameter set, for example, multiple preprocessing modules can be added. Furthermore, multiple stages of preprocessing (filter processing, etc.) can be set in one preprocessing module. The relationship between the Feature associated with the setting item of the setting user interface 50 and the unit to be edited holding the entity of the parameter actually edited by the Feature is specified by going through these selectors in a plurality of stages. be. In this embodiment, the role of the selector is roughly divided into three: specification of parameter set, specification of filtering process, and specification of multi-stage filter. , and the specification of the multi-stage filter is located below the specification of the filtering process, but it is not limited to this. If the parameter set is switched, the selector that specifies the filtering process points to the inside of a different parameter set (flowchart). Furthermore, the specification of the multistage filter specifies the actual filter settings that are set in multiple stages inside the filtering process. By combining the values of multiple selectors in this way, it is possible to identify the unit that holds the parameter associated with that Feature. Note that when the selector that specifies the parameter set is switched, at least the parameter setting target is switched, but whether or not the execution target is also switched depends on the implementation.

For example, in the flowchart shown in FIG. 12, when specifying the unit UX surrounded by double lines, the selector of the highest parameter set is parameter set number 3, and the selector index of the preprocessing module is 1. A unit UX can be identified by determining two values of a certain thing. For example, the pattern search unit, the position correction unit, the image calculation unit (including the filter processing unit), etc. can be determined by the value of the selector. Then, based on the name of the Feature to be accessed, the image inspection application 40 identifies which unit the Feature belongs to, the pattern search unit, the position correction unit, or the image calculation unit. can be reached.

(Configuration of setting device 4)
The setting device 4 transmits to the imaging device 3 the setting values of the setting items set by the user and the data indicating the register information corresponding to the setting items included in the DeviceXML file 31c. It is a device for performing

As shown in FIG. 7, the setting device 4 includes a UI generator 4a. The UI generator 4a is a part that generates various user interfaces, and is configured to be able to generate a setting user interface 50 shown in FIG. 13, for example. Various user interfaces generated by the UI generation unit 4 a are displayed on the display unit 5 .

The setting user interface 50 includes a first area 50a for switching the editing target, a second area 50b for displaying and changing setting items, and an image display area 50c for displaying the image to be edited. is provided. An image switching operation area 51 for switching the image to be edited is provided in the first area 50a. In the image switching operation area 51, identification information of images captured by the imaging device 3 is displayed in a list format. The image identification information may include, for example, the name of the image and the number attached to the image. In this example, the names of four images are displayed vertically, but the number of displayed images and the direction in which they are displayed can be set arbitrarily. The user can select any name from among the names of the images displayed in the image switching operation area 51 by operating the keyboard 6 and the mouse 7 . When a name is selected, the UI generator 4a displays an image corresponding to the selected name in the image display area 50c. The image displayed in the image display area 50c can also be changed.

The second area 50b is provided with a position correction setting area 52 for performing position correction setting as a setting item and a filter setting area 53 for performing filter processing setting as a setting item. In the position correction setting area 52, a portion for selecting whether or not to enable position correction and a portion for selecting the type of position correction tool are assigned and displayed as Features corresponding to the position correction setting. . The selection of the type of the position correction tool can be set only after enabling the position correction. Therefore, the selection of whether to enable the position correction and the selection of the type of the position correction tool are dependent.

In the filter setting area 53, a selected filter type and a part for selecting and adjusting parameters related to filter settings such as extraction size and extraction color are assigned and displayed as Features corresponding to filter processing settings. A single image selected in the image switching operation area 51 may be subjected to a plurality of types of filter processing. assigned and displayed.

Setting work by the user will be described. When the user selects any one of the image names displayed in the image switching operation area 51, the image with the selected name is displayed in the image display area 50c. The display of the second area 50b is automatically switched for those whose parameters for generating the selected image can be edited and set. Internally, for example, when a certain preprocessed image is selected in the image switching operation area 51, the index value corresponding to the preprocessing module used to generate that image is set to the value of the selector that specifies the setting target. is switched, and thereby the setting contents displayed in the second area 50b are switched according to the image. When a value of a selector that specifies a setting target is specified, one or more Features of the preprocessing module indicated by the selector corresponding to the value are read, an image reflecting the setting item is generated, and displayed in the image display area 50c. , and the values set for each parameter are displayed in the position correction setting area 52 and the filter setting area 53 . When the user operates the operation portion displayed in the position correction setting area 52 or the filter setting area 53, the operation is accepted, and the unit corresponding to the preprocessing module corresponding to the selector value specifying the setting target is set. Item is changed.

The following method can be used to identify the unit to be accessed from the value of the selector. That is, the preprocessing module is composed of a plurality of units, and the selector that switches the index of the preprocessing module can use a selector common to these units. , which Feature belongs to which unit can be uniquely determined by the Feature name. As a result, the unit to be accessed can be identified from the combination of the value of the selector and the Feature selected to be edited.

FIG. 14 is a diagram schematically showing a case where the second area 50b of the setting user interface 50 is switched by selecting a parameter set number, and the display of the user interface can be switched as shown in this figure. For example, the setting user interface 50 is provided with a parameter set number selection section 54 as shown on the left side of FIG. Further, as shown on the right side of FIG. 14, a first parameter set setting user interface 50A, a second parameter set setting user interface 50B, a third parameter set setting user interface 50C, a fourth parameter set A user interface 50D for setting the set exists in association with the parameter number. When the parameter set number is switched by the parameter set number selection unit 54, the UI generation unit 4a reconstructs the setting user interface 50 according to the parameter set number. That is, when the user selects the parameter set number 3, the setting user interface 50C for the parameter set number 3 is generated and displayed on the display unit 5. FIG.

FIG. 15 is a diagram conceptually explaining an internal mechanism for switching parameter sets by selecting a parameter set number. The imaging device 3 has a parameter set switching unit 30d that switches parameter sets. A parameter set number (ParameterSetIndex) selected by the user is input to the parameter set switching unit 30d. The parameter set switching unit 30d selects a parameter set corresponding to the input parameter set number, and applies the selected parameter set to the image sensor 3c.

Therefore, the setting device 4 can display the setting user interface 50 on the display unit 5 and receive various settings from the user on the setting user interface 50 . When the user sets so as to perform multi-step processing on the image captured by the imaging device 3, the setting device 4 selects a plurality of selectors for realizing the multi-step processing set by the user. is specified as a multi-stage selector, and the values of the plurality of selectors and register information indicating the locations where the plurality of selectors are stored can be transmitted to the imaging device 3 . Then, the imaging device 3 receives the values of the selectors transmitted from the setting device 4 and the register information indicating the locations where the selectors are stored. It is configured to be stored in the indicated location and to sequentially execute multi-step processing specified by the combination of values of each selector.

Also, when the user changes the setting values of the setting items of the imaging device 3, the setting values of each setting item set by the user and the register information corresponding to each setting item included in the DeviceXML file are stored. The data indicated is transmitted to the imaging device 3, and the imaging device 3 can be set. 3, the degree of freedom in model selection is improved.

In addition, by combining the values of the selectors, it is possible to uniquely specify a part of the multi-stage processing that is sequentially executed. It is possible to perform multi-stage processing such as generation of inspection images using the principle of metric, multispectral imaging, and filtering of the generated inspection images.

(virtual flow chart and virtual unit)
Unlike the flowchart shown in FIG. 8, in an actual operating environment, there is often no regularity in which units appearing in the horizontal or vertical direction can be specified simply by combining selectors. Each time the above unit configuration changes, the DeviceXML must be regenerated inside the camera 3, and the PC software must reacquire the DeviceXML file each time, making it difficult to use.

When a fixed DeviceXML file is used and a flow chart is handled with as few Feature expressions as possible, the flow chart has a pattern determined to some extent in the vertical direction, which is copied horizontally by the concept of parameter sets. desirable. For example, if the flowchart can be expanded vertically, that is, if the number of filter processes that can be set is increased, the entire flowchart will be expanded vertically by the number of filter processes that can be set. Also, if the number of usable parameter sets is increased, the flow chart will expand in the horizontal direction accordingly. However, in an actual operational environment, it is difficult to imagine a case where filtering is used up to the configurable upper limit for all parameter sets. It is expected that most units will be set to "non-execution", such as unit UB4 shown in FIG.

In addition, since there are multiple patterns of combinations of selector values for realizing processing set by the user, the setting device 4 provides the imaging device 3 with multiple patterns of combinations of selector values. can be sent. The imaging device 3 receives the combination of selector values of the plurality of patterns transmitted from the setting device 4, and selects the selector value of any one pattern selected by the user from the combination of the selector values of the plurality of patterns. are configured to specify processing to be executed by a combination of , and dynamically develop setting parameters for executing the specified processing in the internal memory of the imaging device 3 .

Inside the image inspection application 40, an accessible unit is determined by a combination of selectors and the type of Feature, and the unit to which the internal memory area is actually allocated is automatically associated with the Feature. As a result, the user can access without being aware of the shape of the virtual flow chart that is actually internally generated and the arrangement of the virtual units.

(Position correction and filtering)
Among filtering processes, there is a filter that exerts an effect that strongly depends on the directionality of the image. Specifically, it is a directional filter, a difference filter, or the like that extracts a portion where the change in luminance value between pixels adjacent to each other in a predetermined direction of the image is equal to or greater than a predetermined threshold value. The directional filter is a filter that can select the predetermined direction from the X direction and the Y direction.

As shown in FIG. 16, when the photographed image of the workpiece W is an image in which white and black defective portions are reflected on a striped background in the Y direction, the photographed image is If a uniform defect extraction filter process is executed, the defect extraction filter is a filter that extracts areas where there is a large change between adjacent pixels, so changes in the Y direction will also be detected due to the striped pattern in the background. , defects are less noticeable. On the other hand, if filter processing is performed so as to capture changes in the X direction only, the striped pattern in the background disappears and only the defective portion can be extracted, making the presence of the defective portion easier to recognize.

In this way, the workpiece W can be inspected for defects. The defect inspection method is not limited to the method described above, and other defect inspection methods may be used. Further, the defect inspection may be performed by the imaging device 3 or may be performed by the setting device 4 . When the defect inspection is performed by the imaging device 3 , the inspection result can be output from the imaging device 3 to the setting device 4 and displayed on the display unit 5 . When defect inspection is performed by the setting device 4, the image for inspection generated by the imaging device 3 may be output to the setting device 4 and the setting device 4 may perform the defect inspection. It is also possible to output the image picked up by the setting device 4 to the setting device 4 and generate the inspection image by the setting device 4 to perform the defect inspection.

Such a filter that exerts an effect that strongly depends on the directionality of an image cannot obtain an appropriate processing result if the angle of the workpiece W to be imaged changes. Therefore, before executing the filtering process, the imaging device 3 measures the position and angle of the work W by the pattern search unit, and the imaging device 3 performs position correction by the position correction unit, so that the work W is always in a specific direction. You can correct the image to face the This is the position correction function. It should be noted that a plurality of types of tools can be set for the tool used as the search unit so that one suitable for detecting the workpiece W can be selected. The search by the search unit is a search setting for correcting the position of the workpiece W in the captured image, and this setting item is included in the DeviceXML file.

After the position is corrected by the position correction function, a filter process that exerts an effect that strongly depends on the directionality of the image is executed to eliminate the influence of the angle around the optical axis when the workpiece W is imaged, and appropriate processing is performed. You can get results. As shown in FIG. 17, a search model image is specified and stored in the imaging device 3 when the image inspection system 1 is set, and is rotated around the optical axis of the imaging device 3 when the image inspection system 1 is operated. When the workpiece W is imaged, position correction processing is performed based on the search model image so that the workpiece W is oriented in a specific direction. Rotate to face direction. Then, the direction-specific filtering process is performed on the image after the position correction process.

FIG. 18 shows the filter setting user interface 55 . The filter setting user interface 55 is generated by the UI generation unit 4a of the setting device 4 shown in FIG. 7 and displayed on the display unit 5 shown in FIG. 1 and the like. The filter setting user interface 55 includes an image selection section 55a for selecting an image to be filtered, a color extraction setting section 55b for making color extraction settings, a filter setting section 55c for making filter settings, and an image selection section 55c. A region setting unit 55d used when generating a filtered image by cutting out a predetermined region of the image, and a position correction setting unit 55e for setting position correction (search setting) are provided.

FIG. 19 is a diagram showing, in a list format, images that can be filtered. Images that can be filtered differ depending on the method and type of inspection image generation. In the image selection section 55a shown in FIG. 18, any one of the plurality of images shown in FIG. 19 can be selected. Note that the images shown in FIG. 19 are examples, and the present invention is not limited to these images.

FIG. 20 shows a detailed user interface 56 for specifically performing filter settings, which is generated by the UI generation unit 4a of the setting device 4 shown in FIG. 7 and displayed on the display unit 5 shown in FIG. 1 and the like. For example, when the "Add" button of the filter setting section 55c shown in FIG. 18 is pressed, the detailed user interface 56 shown in FIG. 20 can be displayed. The detailed user interface 56 is provided with a selection section 56a for selecting a filter type to be executed and a parameter change section 56b for changing various parameters of the filter selected by the selection section 56a.

FIG. 21 is a diagram showing the types of filters and the directional dependence of each filter. The selection unit 56a shown in FIG. 18 can select any one or a plurality of filters from the plurality of filters shown in FIG. Directionally dependent filters include dilation filters (X/Y directions), erosion filters (X/Y directions), edge extraction filters, real-time grayscale correction filters (X/Y directions), and blurring filters (X/Y directions). Y direction), line defect extraction filter (angle), and the like. These filter processes are filter settings that exert the effect of the directional dependency of each filter by correcting the image in advance using the result of the position correction, and are setting items included in the DeviceXML file.

FIG. 22 shows a flow chart for performing directional filtering after position correction. In step SC1 after the start, imaging is performed. After that, in step SC2, pattern search is performed on the image picked up in step SC1, and then in step SC3, positional correction is performed so that the work W faces a specific direction. Image calculation is performed in step SC4. Directional filter processing is performed as part of this processing. Thereafter, the process proceeds to step SC5 to output the generated image to the setting device 4 or the like.

FIG. 23 shows the direction-specific filter after position correction when a plurality of areas (area 1 and area 2) are set in one image by the area setting section 55d of the filter setting user interface 55 shown in FIG. is a flow chart for executing When imaging is performed in step SD1 after the start, for example, position correction is performed after pattern search is performed for area 1 in step SD2, and position correction is performed for area 2 after pattern search is performed in step SD3. . After that, in step SD4, the generated image is output after performing the image calculation on the area 1, and in step SD5, after performing the image calculation on the area 2, the generated image is output.

That is, the setting device 4 displays the filter setting user interface 55 and the detailed user interface 56, and the setting contents for realizing the search settings and filter settings set by the user on the user interfaces 55 and 56, It is configured to be able to transmit register information indicating the location where the setting content is stored to the imaging device 3 . When the imaging device 3 receives the setting contents and the register information transmitted from the setting device 4, the imaging device 3 stores the setting contents in the location indicated by the corresponding register information, and corrects the position of the work W based on the search setting for the captured image. and filter processing based on the result of position correction by the search setting. This makes it possible to sufficiently obtain the effect of the filter processing that depends on the position of the work W at the time of imaging.

(moving object tracking)
When an inspection image is generated based on the principle of photometric stereo or multispectral, a plurality of imaging processes are executed with a single trigger signal, and the obtained images are combined. In this case, if the workpiece W moves between a plurality of image capturing processes, it is not possible to correctly perform the composition process after the image capturing, so it is necessary to detect the amount of deviation for each image. A search unit can perform a search for detecting the amount of deviation, and after correcting the amount of deviation based on the detected amount of deviation, a process of synthesizing a plurality of images is executed.

Compositing processing can be performed in the image processing unit. Synthesis settings for synthesizing a plurality of captured images are included in the DeviceXML file as setting items. Therefore, before outputting a captured image captured by the imaging device 3 to the outside, the DeviceXML file includes a composition setting for combining a plurality of captured images as a setting item related to processing applied to the captured image. become.

FIG. 24 is a flowchart for tracking a moving object. In the image capturing step of step SE1, the private flowchart is executed to perform image capturing processing multiple times and perform pattern search for each image. After that, in step SE2, a compositing process is performed in consideration of the displacement amount of the plurality of images. At step SE3, the synthesized image is output. Alternatively, as shown in the flowchart on the right, after step Se1, image pickup is performed in step Se2, pattern search is performed in step Se3, and this may be repeated a predetermined number of times.

In addition, the setting device 4 can display a user interface for the user to perform synthesis settings on the display unit 5, and can also display setting contents for realizing the synthesis settings set by the user on the user interface. and register information indicating the location where the setting contents are stored can be transmitted to the imaging device 3 . The imaging device 3 receives the setting content and the register information transmitted from the setting device 4, stores the setting content in the location indicated by the corresponding register information, and executes a synthesizing process of synthesizing a plurality of captured images. is configured as

(Report output)
A form is a text format output of the setting contents of the imaging device 3 that conforms to the GenICam standard. Each setting item is represented by a set of setting name and setting value, which are listed in a list format. In this embodiment, the image inspection application 40 operates on the setting device 4, and the setting device 4 and the imaging device 3 are connected according to the standardization specifications as described above. Therefore, the set contents are read from the imaging device 3 to the setting device 4 side, the read contents are output from the setting device 4 in text format, and managed and saved on the setting device 4 side. It is necessary to be able to write back from the 4 side to the imaging device 3 side.

However, when generating an inspection image based on the principle of photometric stereo or deflectometry, or when generating an inspection image by multispectral imaging, the number of setting items increases and orderliness is required. Become. Therefore, setting items may become complicated. This embodiment is equipped with a function that enables easy form output of the structure of complicated setting items. This is a function realized by the form output unit 4b shown in FIG.

When outputting a form, there is a method of simply writing out the register values in ascending order of the register address, but in order to restore the setting contents correctly, it is necessary to set each register while keeping the setting order between the registers. Therefore, it is preferable to write out in advance in the form in the correct setting order.

The setting order includes the following (1) and (2).
(1) Case where validating a certain register enables setting of other registers For example, a frame rate setting item is an example of a register defined in the GenICam standard. If the user wants to set an arbitrary frame rate, it is first necessary to control a register that allows the frame rate to be set. If the frame rate remains unsettable, the imaging device 3 will operate at an internally determined frame rate. In other words, by enabling the register that enables the frame rate to be set, the frame rate setting item register can be set. Therefore, it is necessary to first write the setting of the register that enables the frame rate to be set, and then write the register for the setting item of the frame rate.

In addition, when the shape image is generated based on the principle of the deflectometry, the feature size of the shape image can be set by enabling the generation of the shape image. Therefore, it is necessary to first write out the register settings that enable generation of shape images, and then write out the shape image feature size setting registers.
(2) When a selector register is rewritten, the entity pointed to by another register is switched In this embodiment, in order to reduce the register consumption of array type parameters, a register (selector register) indicating an array index is separately introduced. there is This is because if there is a selector register, only one parameter body register needs to be defined. For example, the terminal selector that selects a terminal and the selector that sets whether to invert the signal of the selected terminal have the relationship (2) above. Also, the white balance setting selector that can take any value of red, green, or blue and the white balance setting value of the selected color have the relationship (2) above. Therefore, after setting the selector register, it is necessary to set the parameter body register.

In other words, there are a plurality of setting items having a dependency relationship such that the setting of one setting item changes the validity or value of the other setting item. In this embodiment, considering the order shown in (1) and (2) above, the appearance order of the registers in the DeviceXML file is determined in advance so that there is no reverse dependency in the setting order. Then, by writing out the data in the form in that order, the structure of the complicated setting items can be converted into a form in the correct setting order.

FIG. 25 shows an example of a form in which registers having two dependencies, dependency 1 and dependency 2, are described. Each register is defined by grouping related persons into a group called a category. Each category can also be grouped together by higher-level categories. The relationship between each category and each register has a tree structure as shown in FIG. A category is formed by classifying a plurality of setting items that have a dependency relationship in which the setting of one setting item changes the validity or value of the other setting item according to the degree of relevance. easier to understand. A plurality of categories exist in one form.

Classification of registers by category is useful when the user searches for a register corresponding to the content that the user wants to set from the list. Therefore, ignoring the category structure and arranging the register appearance order in the correct order only from the viewpoint of dependency relationship would destroy the category structure and significantly hinder user convenience. will do. Therefore, the form output unit 4b arranges the registers in the correct appearance order while maintaining the classification structure by category.

Specifically, first, sibling nodes are rearranged so that sibling nodes are in the correct order of appearance, starting from the last node of the tree. After that, sibling nodes one level higher on the tree are rearranged so that they are in the correct order of appearance. This is repeated recursively toward the root of the tree. At this time, the appearance order of the registers is rearranged so that the dependency relationship between the two is always in the direction from the rear to the front. Also, the appearance order of the categories is rearranged so that when there is a dependency relationship from a register in category 1 to a register in category 2, category 2 appears first, and then category 1 appears.

Also, when sorting in each layer of the tree, topological sorting is performed by regarding the dependency relationship between registers/categories as directed edges. Topological sorting itself is well-known, and is to arrange each node in order so that any node is positioned before the node ahead of its output edge.

Next, a specific example of the rearrangement procedure will be described with reference to FIGS. 26 and 27. FIG. In this example, categories and setting items have hierarchical dependencies.

Procedure 1: Change the order of category 2-2 and register 2-3 in the last section. This correctly orients dependency 2 from register 3-0 to register 2-3 in the back-to-front direction, as shown in FIG.

Procedure 2: Change the order of category 1-0 and register 1-1 one level higher. This correctly orients dependency 1 from register 2-1 to register 3-0 in the back-to-front direction, as shown in FIG. If there is a higher hierarchy, the order can be changed in the same way.

FIG. 28 is a diagram showing the appearance order of registers before and after rearrangement and the output form. LtrxSh1ImageEnable is a register for setting whether to generate a shape image based on the principle of photometric stereo. LineInverter is a register for setting whether to invert the output signal of the selected terminal. LineSelector is a register that selects a terminal. LtrxSh1FeatureSize is the feature size setting register for the shape image. In the appearance order of the registers before rearrangement, LineInverter appears before LineSelector, and the appearance order is different from the setting order. Therefore, the order of the registers is rearranged so as to match the setting order, and then output as a form. As indicated by "output form", at the selector, the output process is looped by the number of selector indexes.

The register definition order can be automatically rearranged by the form output unit 4b of the setting device 4, for example. The form may be rearranged at the timing of outputting the form, or the device XML file which has been rearranged in advance may be held in the controller, and when the PC is connected, the Device XML file may be returned as it is. The timing of rearranging the register definition order is not particularly limited.

FIG. 29 shows a register structure for setting the imaging mode of photometric stereo. In the photometric stereo mode, a plurality of images are captured while switching the lighting direction of the illumination device 2, and the plurality of image data are combined to create an inspection image emphasizing scratches and irregularities on the surface of the workpiece W. This is the mode to create. The photometric stereo mode also has the moving object tracking function described above. When moving object tracking is enabled, images for calculating the amount of movement of the work W are additionally captured, so that each image can be correctly synthesized even when the work W is moving. The images for calculating the amount of movement of the work W can be, for example, the image at the time when the first image was started and the image at the time when the last image was imaged.

In the photometric stereo mode, on the premise that a certain image has already been captured and exists, it is necessary to select whether or not to output that image to the setting device 4, and so on. , the order of setting between registers must be observed. In other words, as indicated by the arrows in FIG. 30, there is a dependency from the front to the back from the register for whether or not to output to the register for whether or not to image. The setting order (input order) is not set. Therefore, if an attempt is made to input a form that has been output in this order, an error will occur during input and the setting cannot be completed.

In this embodiment, the registers can be automatically rearranged by the procedure described above. FIG. 31 shows a state in which the plurality of registers shown in FIG. 30 are rearranged. Since this definition order is the order (setting order) that should be set by the user, no error occurs when inputting to the output form.

Therefore, the setting device 4 classifies a plurality of setting items having a dependency relationship such that the setting of one setting item changes the validity or value of the other setting item into a plurality of categories formed by classifying them according to the degree of relevance. It is configured to output a form in which each setting item is described according to the order of setting based on the dependency relationship between the setting items. As a result, each setting item is described in the form according to the setting order, based on the dependencies between multiple categories and between each setting item. It is difficult to restore by mistake.

(Action and effect of the embodiment)
As described above, according to this embodiment, a plurality of captured images can be synthesized by the imaging device 3 conforming to the standardization specifications. , multispectral imaging, and the like.

In addition, it is possible to make the image pickup device 3 conforming to the standardization specifications perform multi-stage processing in order, and it is possible to improve the degree of freedom in selecting the model of the image pickup device 3 and improve the accuracy of image inspection. can.

In addition, since the setting parameters for executing the specified processing are dynamically developed in the internal memory of the imaging device 3, flexible function expansion is realized for the imaging device 3 conforming to the standardization specifications. , the memory usage of the imaging device 3 can be suppressed.

In addition, since the filter processing can be performed based on the result of the position correction after performing the position correction of the work W in the captured image, the effect of the filter processing depending on the position and orientation of the work W at the time of imaging can be fully obtained. Obtainable.

Also, a plurality of setting items having a dependency relationship are classified according to the degree of association to form a plurality of categories, and based on the dependency relationship between the plurality of categories and between each setting item, each setting item is arranged according to the setting order. A written form can be output. As a result, it is possible to easily output a form exchanged between the imaging device 3 and the setting device 4 conforming to the standardization specifications, so that the implementation can be facilitated, and the setting contents are not restored by mistake. can be made

The above-described embodiments are merely examples in all respects and should not be construed in a restrictive manner. Furthermore, all modifications and changes within the equivalent range of claims are within the scope of the present invention.

INDUSTRIAL APPLICABILITY As described above, the present invention can be used when inspecting an inspection object such as a workpiece.

1 image inspection system 2 illumination device 2a light emitting unit 2b illumination control unit 3 imaging device 4 setting device 4a UI generation unit 4b form output unit 5 display unit 6 keyboard 7 mouse 8 external control device 31 camera 32 condensing optical system 33 imaging control Unit 40 Image inspection application 50 Setting user interface 200 Illumination devices 201 to 204 First to fourth light emitting units 205 Illumination control unit W Work (inspection object)

Claims (7)

an imaging device for an image inspection device that captures an image of an object to be inspected;
A setting device that sets the image pickup device, is connected to the image pickup device via a network, and includes setting items of the image pickup device and register information indicating a location where setting values for each setting item are stored. Acquiring a file from the outside of the setting device , and transmitting data indicating setting values of each setting item set by a user and register information corresponding to each setting item included in the file to the imaging device , an image inspection system comprising a setting device for setting the imaging device,
In the file, a position of an inspection object in the captured image is stored as setting items related to processing to be applied to the captured image before outputting the captured image captured by the imaging device to the outside of the imaging device. Search settings for correction and filter settings that change the processing content based on the result of position correction by the search settings are included,
The setting device displays a user interface, and includes setting items for realizing search settings and filter settings set by the user on the user interface, and registers indicating locations where setting values of the setting items are stored. information can be sent to the imaging device,
The imaging device stores the setting values of the setting items in the locations indicated by the corresponding register information, performs position correction of the inspection object on the captured image based on the search settings, and uses the result of the position correction based on the search settings. An imaging inspection system configured to perform filtering based on. The image inspection system according to claim 1,
An image inspection system, wherein the image pickup device is configured to output an image that has undergone filtering processing based on a result of position correction based on the search settings to the outside of the image pickup device . In the image inspection system according to claim 1 or 2,
The image inspection system, wherein the filter setting is a filter setting that exerts an effect depending on the directionality of the image. In the image inspection system according to claim 3,
The search setting is configured to search for an inspection object in an image captured by the imaging device, and perform position correction by rotating at least an area in which the inspection object is captured. Image inspection system. In the image inspection system according to claim 4,
The image inspection system, wherein the filter setting is a filter setting for extracting a portion where a change in luminance value between pixels adjacent to each other in a predetermined direction of the image is equal to or greater than a predetermined threshold value. In the image inspection system according to claim 5,
An image inspection system, wherein the predetermined direction is a filter setting that can be selected from an X direction and a Y direction. an imaging device for an image inspection device that captures an image of an object to be inspected;
A setting device that sets the image pickup device, is connected to the image pickup device via a network, and includes setting items of the image pickup device and register information indicating a location where setting values for each setting item are stored. Acquiring a file from the outside of the setting device , and transmitting data indicating setting values of each setting item set by a user and register information corresponding to each setting item included in the file to the imaging device , an image inspection method for inspecting an object to be inspected by preparing a setting device for setting the imaging device,
In the file, a position of an inspection object in the captured image is stored as setting items related to processing to be applied to the captured image before outputting the captured image captured by the imaging device to the outside of the imaging device. Search settings for performing correction and filter settings that change the processing content based on the result of position correction by the search settings are included,
causing the setting device to display a user interface;
transmitting setting items for implementing search settings and filter settings set by a user on the user interface and register information indicating a location where setting values of the setting items are stored to the imaging device;
The imaging device stores the setting values of the setting items in the locations indicated by the corresponding register information, performs position correction of the inspection object on the captured image based on the search settings, and uses the result of the position correction based on the search settings. An image inspection method, characterized in that filtering is performed based on.

Images