JP5947168B2 - Appearance inspection apparatus, control method and program for appearance inspection apparatus - Google Patents

Description

  The present invention relates to an appearance inspection apparatus that takes an image of a product and inspects the appearance of the product, a control method for the appearance inspection apparatus, and a program.

  The appearance of a product such as a connector or a semiconductor element is imaged and pattern matching (image recognition) is performed, whereby a so-called appearance inspection of the product is performed. Depending on the size of the product, there are things that do not fit in the imaging range of the camera, and for such products, an image representing the entire product was created by connecting multiple images, and appearance inspection was performed (patents) References 1, 2).

Japanese Patent Laid-Open No. 06-167321 JP 09-251536 A

  In order to determine the quality of a product (inspection object) that is subject to visual inspection, the end or corner of the terminal of the inspection object is set in advance as a reference point, or a straight line connecting a plurality of terminals is set. Or set as a reference line. This is to inspect whether the distance from the reference point to a certain part is the distance specified by the design, or the distance of the perpendicular line from the reference line to the end of the inspection object is the distance specified by the design. This is to inspect whether it is. In order to accurately detect such a reference point or reference line from an image, it is necessary to apply appropriate illumination to the inspection object. For example, in order to detect the outer contour of the connector, a method of increasing the contrast between the outer contour and other portions using a backlight is used. However, the appearance inspection includes various inspections, and lighting suitable for one inspection may not be suitable for another inspection. For example, one type of inspection uses only the backlight, another type of inspection uses only the front light, and other types of inspection use both the backlight and the front light. The lighting may be switched.

  For example, since the connector has a large number of pins, the entire connector is often divided into a plurality of images for imaging, but the imaging conditions may be changed a plurality of times at each imaging location. When the pin is illuminated and imaged with the front light, the pattern is easily recognized, and when the contour of the outer shape of the housing is raised with the backlight and imaged, the pattern is easily recognized. The problem here is that the distance from the pin to the outer shape of the housing cannot be calculated from a single image. In other words, when trying to calculate the distance between the pin and the corner of the housing, it is easy to recognize the position of the pin in the image obtained by the front light, but it is difficult to recognize the position of the corner of the housing, so the distance cannot be calculated accurately. On the other hand, in the image obtained with the backlight, it is easy to recognize the position of the corner of the housing, but it is difficult to recognize the position of the pin, so the distance cannot be calculated accurately.

  Dimension between multiple images, such as measuring the distance from the reference point of one image to the reference point of another image in multiple images obtained by imaging the inspection object under different illumination conditions. The measurement could not be performed. Therefore, it can be said that there is a demand from the market for a method for accurately obtaining dimensions such as the distance between parts that are difficult to recognize images due to differences in illumination conditions.

  Accordingly, an object of the present invention is to accurately obtain dimensions such as the distance between parts that are difficult to recognize images due to differences in illumination conditions.

The present invention includes, for example, an illumination unit that irradiates a plurality of different illumination lights to an inspection target, an illumination control unit that controls the illumination unit so as to switch illumination light emitted by the illumination unit, and the illumination unit An imaging unit that images the inspection object irradiated with illumination light, and a communication unit that receives an imaging trigger signal that defines the imaging timing of the imaging unit from an externally connected control device ,
An appearance inspection apparatus that determines the quality of the appearance of the inspection object using the image of the inspection object acquired by the imaging unit, and outputs a determination signal indicating the determination result ,
A first image storage unit that stores a first image acquired by the imaging unit imaging a predetermined region of the inspection object irradiated with the first illumination light by the illumination unit;
A second image storage unit for storing a second image acquired by the imaging unit imaging the predetermined region of the inspection object irradiated with the second illumination light by the illumination unit;
A display unit for displaying the first image and the second image;
Using the first image displayed on the display unit, a first setting for extracting a first feature portion that is a starting point when measuring a dimension related to the inspection object from the first image. And the second image displayed on the display unit is used to extract a second feature portion that is an end point when measuring a dimension related to the inspection object from the second image. A setting unit for executing the second setting of
When the imaging unit receives the imaging trigger signal, the imaging unit captures and acquires a new inspection object irradiated with the first illumination light from the illumination unit. The first characteristic portion is extracted based on the first setting by the setting unit, and the imaging unit captures and acquires the new inspection object irradiated with the second illumination light by the illumination unit. An image recognition unit that extracts the second feature portion from the second image based on the second setting by the setting unit;
A dimension calculation unit for calculating a dimension from the first feature part to the second feature part extracted by the image recognition unit ;
A quality determination unit that determines the quality of the appearance of the inspection object based on the dimension calculated by the dimension calculation unit and a predetermined threshold;
Means for outputting a determination signal indicating a determination result by the pass / fail determination unit;
It is characterized by having.

  According to the present invention, different characteristic parts are set for two images having different illumination conditions, and dimension measurement is performed from the characteristic part set in one image and the characteristic part set in the other image. Therefore, it is possible to accurately obtain dimensions such as the distance between parts that are difficult to recognize images due to differences in illumination conditions.

FIG. 1 is a schematic view showing an outline of an appearance inspection apparatus. FIG. 2 is a diagram illustrating an example of a hardware configuration of the appearance inspection apparatus. FIG. 3 is a flowchart showing a basic flow of the appearance inspection process. FIG. 4 is a diagram illustrating a positional relationship among the inspection target, the camera, and the illumination device when setting the parameters. FIG. 5 is a diagram illustrating an image captured by the front light and an image captured by the backlight by dividing the connector that is the inspection object into three parts. FIG. 6 is a diagram illustrating an example of dimension measurement. FIG. 7 is a diagram illustrating an example of dimension measurement regarding the pins in the second row. FIG. 8 is a diagram illustrating a method for obtaining the height of each pin. FIG. 9 is a diagram illustrating an example of obtaining the pin height for each pin in the second row. FIG. 10 is a diagram showing a dialog for setting the number of times of imaging. FIG. 11 is a diagram illustrating a dialog for setting the number of times of imaging. FIG. 12 is a diagram showing a dialog visually showing the number of times of imaging set by the user. FIG. 13 is a diagram showing that a connector category is selected by a mouse from a tool catalog showing a plurality of image processing tool categories, and further, connector position deviation correction is selected as an image processing tool. FIG. 14 is a diagram showing a UI for setting parameters necessary for connector position deviation correction. FIG. 15 is a diagram showing a UI for setting parameters necessary for correcting the connector position deviation. FIG. 16 is a diagram showing a UI for setting parameters necessary for correcting the connector position deviation. FIG. 17 is a diagram illustrating an example of a UI indicating that the setting for correcting the connector displacement has been completed. FIG. 18 is a diagram showing that a connector category is selected by a mouse from a tool catalog showing a plurality of image processing tool categories, and a connector dimension inspection tool is selected as an image processing tool. FIG. 19 is a diagram showing a setting screen for setting the connection points of images. FIG. 20 is a diagram showing a pin row detection parameter setting screen. FIG. 21 is a diagram illustrating a pin model registration screen. FIG. 22 shows a pin detection setting screen. FIG. 23 is a diagram showing a search area, inspection area, and reference point setting screen for the middle part of the connector. FIG. 24 is a diagram showing a search area, inspection area, and reference point setting screen for the right end of the connector. FIG. 25 is a diagram showing a screen indicating that registration of the pin model in the first row (setting for pin row detection) has been completed. FIG. 26 is a diagram illustrating a setting screen for performing settings related to connector end detection. FIG. 27 is a diagram illustrating selection items for selecting an end that is an end detection target. FIG. 28 is a diagram showing a registration screen for a search model for edge detection. FIG. 29 is a diagram showing a search area setting screen. FIG. 30 is a diagram showing a registration screen for a search model for edge detection. FIG. 31 is a diagram showing a search area setting screen. FIG. 32 is a diagram showing a reference point designation screen. FIG. 33 is a diagram showing a reference point designation screen. FIG. 34 is a diagram showing a screen indicating that the work left end and the work right end are normally set. FIG. 35 is a diagram showing a setting screen for setting a reference line. FIG. 36 is a diagram showing selection items of the reference line designation method. FIG. 37 is a diagram showing a screen when “line connecting both ends of the housing” is designated as the reference line from the selection items. FIG. 38 is a diagram showing a main screen for setting dimension measurement parameters. FIG. 39 is a diagram showing an item selection screen for setting dimension measurement parameters. FIG. 40 is a diagram showing a screen for setting parameters used for pass / fail judgment for the selected measurement item. FIG. 41 is a diagram showing a screen for setting parameters used for pass / fail judgment for the selected measurement item. FIG. 42 is a diagram showing an item selection screen for setting dimension measurement parameters. FIG. 43 is a diagram showing a screen for setting parameters used for pass / fail judgment for the selected measurement item. FIG. 44 is a diagram showing a screen for setting parameters used for pass / fail judgment for the selected measurement item. FIG. 45 is a diagram illustrating an example of a mark serving as a reference for connecting a plurality of images. FIG. 46 is a diagram illustrating a setting screen for setting a mark serving as a reference for image connection. FIG. 47 is a diagram showing a setting screen for setting a mark as a reference for image connection. FIG. 48 is a diagram illustrating a setting screen for setting a mark serving as a reference for image connection. FIG. 49 is a diagram showing a front light image and a backlight image connected with a mark as a reference. FIG. 50 is a diagram illustrating a pin detection method using edge detection. FIG. 51 is a diagram illustrating a housing detection method using edge detection. FIG. 52 is a functional block diagram of the appearance inspection apparatus. FIG. 53 is a flowchart showing a series of steps including a parameter setting process related to appearance inspection and non-defective product determination.

  An embodiment of the present invention is shown below. The individual embodiments described below will help to understand various concepts, such as the superordinate concept, intermediate concept and subordinate concept of the present invention. Further, the technical scope of the present invention is determined by the scope of the claims, and is not limited by the following individual embodiments.

  FIG. 1 is a schematic diagram showing an outline of the appearance inspection apparatus 1. The appearance inspection apparatus 1 includes a controller 2, a programmable logic controller (PLC) 3, a camera 4, a lighting device 5 (front light 51 and backlight 52), a mouse 9, a monitor 10, and a program creation support device 11. And have. The inspection object 8 is conveyed by the conveying device 7 such as a belt conveyor controlled by the PLC 3, and the inspection object 8 illuminated by the illumination device such as the front light 51 and the backlight 52 is imaged by the camera 4. For example, the controller 2 switches the illumination device to illuminate the inspection object 8 or causes the camera 4 to perform imaging in accordance with a command from the PLC 3.

  The controller 2 executes various measurement processes such as edge detection and area calculation from the image of the inspection object (work) 8. Image processing is executed using image data obtained from the camera 4, and a determination signal is output as a signal indicating a determination result such as pass / fail of the inspection object 8 to an externally connected control device such as the PLC 3.

  The camera 4 includes a camera module having an image sensor that images the inspection object 8. As the imaging device, for example, a CMOS (complementary metal oxide semiconductor) or a CCD (charge coupled device) can be used. The inspection object 8 is imaged on the basis of a control signal input from the PLC 3, for example, an imaging trigger signal that defines the timing for capturing image data from the camera 4.

  The monitor 10 is a display device such as a liquid crystal panel or a self-luminous panel. An image obtained by imaging the inspection object 8 and a result of measurement processing using the image data are displayed. The monitor 10 may display an image acquired from a non-defective product such as a reference image used for creating comparison data (model image) for pattern matching.

  The mouse 9 is an input device for the user to perform various operations on the monitor 10 (can be omitted if the monitor 10 is a touch panel). The mouse 9 selects each menu item on the monitor 10 and sets a parameter value. The mouse 9 is an example of a pointing device. The user can check the operating state of the controller 2 during operation by visually checking the monitor 10. Further, the user can perform various settings and various edits as necessary by operating the mouse 9 while viewing the monitor 10.

  The illumination device 5 is a device that illuminates the inspection object 8, and in FIG. 1, a front light 51 and a backlight 52 are shown. In addition to these, the illumination device 5 includes, for example, coaxial epi-illumination that highlights gloss, low-angle illumination that highlights scratches and stamped edges, black-light illumination that shines black light, and surface illumination that is used as the backlight 52 (front Illumination devices that perform various types of illumination, such as dome illumination that illuminates from the opposite side of the light 51 and observes the transmitted light or shadow of the inspection object, and diffused light from all sides, are adopted. May be. In particular, the coaxial epi-illumination is an illumination method that illuminates the entire visual field substantially uniformly, and is almost the same as the illumination method in which the camera 4 and the illumination device 5 are arranged in a V shape to receive specularly reflected light from the inspection object 8. There is an advantage that the same effect can be obtained. The low-angle illumination is an illumination technique in which light projecting elements such as LEDs are arranged in a ring shape to illuminate the surface of the inspection object 8 at a shallow angle from the entire circumference. The light that hits the surface of the inspection object 8 is not reflected in the direction of the camera 4, and only the light reflected by the stamped or scratched edge portion is received. That is, since the irradiation angle is a very shallow angle, reflection on the glossy surface is weak, and strong reflection is obtained only with a few scratches or edges on the inspection object 8, and a clear contrast is obtained. In other words, the front light 51 is illumination for extracting features (texture, edges, etc.) of the surface of the inspection object 8, and irradiates illumination light from the camera installation side. The backlight 52 is illumination for extracting the outline (or edge) of the inspection object 8, and irradiates illumination light from the side opposite to the camera installation side. In other words, the backlight 52 irradiates light from the opposite direction to the camera with the inspection object 8 interposed therebetween.

  The program creation support device 11 is a computer (PC) for creating a control program executed by the controller 2. The control program has a plurality of measurement processing modules for executing different measurements relating to the appearance inspection described below. The controller 2 calls and executes various measurement processing modules in the set order. The program creation support apparatus 11 and the controller 2 are connected via a communication network, and control information and setting information such as parameter values generated on the program creation support apparatus 11 are transferred to the controller 2. Conversely, setting information such as a control program and parameter values can be taken from the controller 2 and edited on the program creation support apparatus 11.

  In a factory, a plurality of inspection objects 8 flow on a line of a transfer device 7 such as a conveyor. The controller 2 images the inspection object 8 with the camera 4 installed above (or on the side and below) of the inspection object 8, and uses the captured image as a reference image (for example, an image of a non-defective product) or a reference image. In comparison with the model image created from the above, it is determined whether or not the inspection object 8 has scratches or defects. If it is determined that the inspection object 8 has scratches, defects, or the like, an NG determination is made. On the other hand, when it is determined that there is no scratch or defect on the inspection object 8, an OK determination is made. As described above, the appearance inspection apparatus 1 determines the quality of the appearance of the inspection object 8 using the image obtained by imaging the inspection object 8.

  When the appearance inspection of the inspection object 8 is performed, it is necessary to set the contents (parameter values, etc.) of various parameters used for the inspection. The parameters include, for example, imaging parameters that define imaging conditions such as shutter speed, illumination parameters that specify illumination conditions such as illuminance, and measurement processing parameters that specify inspection conditions indicating what kind of inspection to perform (so-called inspection parameters) ) Etc. In the appearance inspection apparatus 1, the contents of these various parameters are set before quality determination.

  The appearance inspection apparatus 1 is a mode in which an appearance inspection of the inspection object 8 that actually flows one after another on the line of the transport device 7, that is, an operation mode (Run mode) in which the quality of the inspection object 8 is actually judged. And a setting mode (non-run mode) for setting the contents of various parameters used for inspection, and mode switching means for switching between these modes. In the operation mode, the user sets optimum parameter values for various parameters in the setting mode before the pass / fail determination is repeatedly performed on the plurality of inspection objects 8 flowing on the line of the transfer device 7. (adjust. Basically, default values are set for various parameters, and when the user determines that the default values are optimal as the parameter values, there is no need to adjust the parameter values. However, in reality, there may be a case where the determination result desired by the user cannot be obtained with the default values due to differences in the surrounding lighting environment, the mounting position of the camera 4, the posture deviation of the camera 4, focus adjustment, and the like. is there. Therefore, in the setting mode, the operation mode can be switched to the setting mode on the monitor 10 of the controller 2 or the program creation support apparatus 11, and the contents of various parameters can be edited.

<Hardware Configuration of Appearance Inspection Apparatus 1>
FIG. 2 is a diagram illustrating an example of a hardware configuration of the appearance inspection apparatus 1.

  The main control unit 21 performs numerical calculation and information processing based on various programs, and controls each part of the hardware. For example, a CPU 22 as an intermediate processing unit, a work memory 23 such as a RAM that functions as a work area when the main control unit 21 executes various programs, a ROM and a flash that store startup programs, initialization programs, and the like And a program memory 24 such as a ROM or an EEPROM.

  The illumination control unit 26 transmits an illumination control signal to the front light 51, the backlight 52, and the like based on instructions from the CPU 22 and the PLC 3 of the main control unit 21.

  The image input unit 25 includes an application specific integrated circuit (ASIC) that captures image data obtained by imaging with the camera 4. The image input unit 25 may include a frame buffer for buffering image data. Specifically, when receiving an imaging command for the camera 4 from the CPU 22, the image input unit 25 transmits an image data capture signal to the camera 4. Then, the image input unit 25 captures image data obtained by imaging after the camera 4 performs imaging. The captured image data is temporarily buffered (cached).

  The operation input unit 27 receives an operation signal from the mouse 9. The operation input unit 27 functions as an interface (I / F) that receives an operation signal from the mouse 9 based on a user operation.

  On the monitor 10, the user's operation content using the mouse 9 is displayed. Specifically, by operating the mouse 9, the user edits the image processing control program, edits the parameter value of each measurement processing module, and sets the imaging condition of the camera 4 on the monitor 10. Or register a characteristic part of the reference image as a model image, or search the search area and set an area that matches the model image as an inspection area. .

  The display control unit 28 includes a display DSP that displays an image on the monitor 10. The display control unit 28 may include a video memory such as a VRAM that temporarily stores image data when an image is displayed. Based on a display command (display command) sent from the CPU, a control signal for displaying a predetermined image (video) on the monitor 10 is transmitted. For example, a control signal is transmitted to the monitor 10 in order to display image data before or after the measurement process. The display control unit 28 also transmits a control signal for causing the monitor 10 to display the user's operation content using the mouse 9.

  The communication unit 29 is communicably connected to the external PLC 3 and the program creation support device 11. For example, the communication unit 29 is installed in the production line in order to recognize the arrival timing of the inspection object 8, and when there is a trigger input from a sensor (photoelectric sensor or the like, not shown) connected to the PLC 3, the PLC 3 Functions as an interface (I / F) for receiving an imaging trigger signal from the camera. The communication unit 29 also functions as an interface (I / F) that receives the control program of the controller 2 transferred from the program creation support apparatus 11.

  The image processing unit 30 includes a calculation DSP that executes measurement processing such as edge detection and area calculation. The image processing unit 30 may include a memory that stores image data for measurement processing. The image processing unit 30 executes a measurement process on the image data. Specifically, the image processing unit 30 first reads image data from the frame buffer of the image input unit 25 and performs internal transfer to the memory in the image processing unit 30. Then, the image processing unit 30 reads the image data stored in the memory and executes measurement processing.

  The program memory 24 is a control program for controlling each part of the illumination control unit 26, the image input unit 25, the operation input unit 27, the display control unit 28, the communication unit 29, and the image processing unit 30 with commands of the CPU 22. Storing. Further, the control program transferred from the program creation support apparatus 11 is stored in the program memory 24.

  When the CPU 22 receives an imaging trigger signal from the PLC 3 via the communication unit 29, the CPU 22 sends an imaging command (command) to the image input unit 25. Further, the CPU 22 transmits a command for instructing image processing to be executed to the image processing unit 30 based on the control program. In addition, as a device for generating the imaging trigger signal, a trigger input sensor such as a photoelectric sensor may be directly connected to the communication unit 29 instead of the PLC 3.

  These pieces of hardware are communicably connected via an electrical communication path (wiring) such as a bus.

<Measurement module (image processing tool)>
Here, the measurement module that performs the appearance inspection is referred to as an image processing tool. There are various image processing tools. The main image processing tools are edge position measurement tool, edge angle measurement tool, edge width measurement tool, edge pitch measurement tool, area measurement tool, blob measurement tool, and pattern search measurement. Tools, scratch measurement tools, etc.
● Edge position measurement tool: By setting a window for the inspection area where the edge position is to be detected on the screen on which the image of the inspection object 8 is displayed, scanning is performed in any direction within the set inspection area. Thus, a plurality of edges (locations switching from light to dark or locations switching from dark to light) are detected. The specification of one edge is received from the detected plurality of edges, and the position of the edge that has received the specification is measured.
Edge angle measurement tool: Two segments are set in the inspection region for which the setting has been received, and the inclination angle of the inspection object 8 from the edge detected in each segment is measured. The tilt angle can be positive, for example, clockwise.
Edge width measurement tool: Detects a plurality of edges by scanning in an arbitrary direction within the inspection region that has received the setting, and measures the width between the detected edges.
● Edge pitch measurement tool: Scans in any direction within the inspection area that accepts the settings to detect multiple edges. The maximum value / minimum value and average value of the distances (angles) between the detected edges are measured.
Area measurement tool: The image of the inspection object 8 imaged by the camera 4 is binarized to measure the area of the white area or black area. For example, the area of the white region or the black region is measured by accepting the specification of a white region or a black region as a parameter to be measured.
Blob measuring tool: the image of the inspection object 8 imaged by the camera 4 is binarized, and the number, area, and parameters as a set of pixels (blob) having the same luminance value (255 or 0) Measure the center of gravity.
Pattern search measurement tool: A part similar to an image pattern stored in advance in an image of the inspection object 8 that has been stored in advance in an image pattern (model image) to be compared. By detecting this, the position, inclination angle, and correlation value of the image pattern are measured.
Scratch measurement tool: A small area (segment) is moved within the inspection area where the setting is received to calculate an average density value of pixel values, and a position where a density difference equal to or greater than a threshold value is determined to be scratched.
● In addition, OCR recognition tool that recognizes character strings by cutting out character information in the inspection area and collating it with dictionary data, etc., while shifting the window (area) set on the image, at each window position Trend edge tool with a function to repeat edge detection, density tool with a function to measure the average and deviation of the density in the set window, density with a function to measure the average and deviation of the density in the set window There are tools and the like, and the user can select a necessary image processing tool in accordance with the inspection contents. Note that these image processing tools are merely representative examples of typical functions and their implementation methods. An image processing tool corresponding to any image processing can be a subject of the present invention.

<Basic flow of visual inspection>
FIG. 3 is a flowchart showing a basic flow of the appearance inspection process. The appearance inspection process is a setting for setting threshold values such as a model image, an inspection area, a search area, a detection point (hereinafter referred to as a reference point), a reference line, and a tolerance necessary for determining the quality of the inspection object 8. The operation mode is divided into a mode and an operation mode in which the inspection object 8 is actually imaged and pattern matching is executed to determine pass / fail. In order to appropriately set the inspection parameters, it is common to repeatedly execute the setting mode and the operation mode.

  In S <b> 301, the CPU 22 transmits an imaging command to the camera 4 through the image input unit 25 to cause the camera 4 to perform imaging. The CPU 22 displays the image data acquired by the camera 4 on the monitor 10 through the display control unit 28. The user visually confirms the image displayed on the monitor 10 to confirm the posture of the camera 4 and the illumination state of the illumination device 5.

  In S <b> 302, the CPU 22 adjusts exposure conditions such as the shutter speed of the camera 4 based on the instruction input by the mouse 9. Note that the user may manually adjust the posture of the camera 4.

  In S <b> 303, the CPU 22 transmits an imaging command to the camera 4 in order to capture an image of the inspection object 8 arranged at the imaging position of the transport device 7 as a work image. Note that the work image may be a reference image that is stored in the nonvolatile memory and repeatedly used, or may be an image that is captured each time in order to create a model image. Here, the work image is stored in the work memory 23. The model image may be created from the reference image.

  In S304, the CPU 22 executes a misalignment correction setting process. In the image acquired by the camera 4, the position of the image of the inspection object 8 may be shifted from the ideal position. Therefore, the CPU 22 corrects misalignment by rotating or shifting the image of the inspection object 8. Note that the image processing unit 30 may execute the positional deviation correction.

  In S305, the CPU 22 sets the various image processing tools described above. For example, which measurement is performed in the appearance inspection, and a search area, an inspection area, a reference point, and the like necessary for performing the measurement are set.

  In step S <b> 306, the CPU 22 sets parameters (e.g., inspection threshold such as tolerance) necessary for appearance inspection based on an instruction input by the mouse 9. In S307, the CPU 22 switches from the setting mode to the operation mode.

  In S308, the CPU 22 images the inspection object 8 with the camera 4 in accordance with an instruction from the PLC 3, causes the image processing unit 30 to perform pattern matching, determines pass / fail based on the execution result, and outputs the determination result to the PLC 3. Or output to the monitor 10.

  In S309, when a mode switching instruction is input from the mouse 9, the CPU 22 switches from the operation mode to the setting mode. In S310, the CPU 22 resets the parameters based on the instruction input by the mouse 9.

<Parameter settings>
FIG. 4 shows the positional relationship among the inspection object 8, the camera 4, and the illumination device 5 when setting parameters. The inspection object 8 may be an IC (integrated circuit) having a plurality of pins, a BGA (ball grid array) having a plurality of solder balls, etc., but here, for convenience of explanation, a resin housing Assume a connector having 40 and a plurality of pins 41.

  The backlight 52, which is the illumination device 5, is disposed so as to face the camera 4 with the transport device 7 interposed therebetween, and illuminates the inspection object 8 from the rear side of the inspection object 8. The front light 51 is a ring illumination device, is disposed on the same side as the camera 4 when viewed from the transport device 7, and illuminates from the front side of the inspection object 8.

  In the operation mode, the PLC 3 controls the conveyance device 7, and when the inspection object 8 arrives at the imaging position of the camera 4, the conveyance device 7 is stopped and the camera 4 is commanded to image via the controller 2. At this time, the PLC 3 instructs the illumination device 5 to illuminate via the controller 2. In addition, the camera 4 captures an image in a state in which the front light 51 (for extracting the characteristics of the surface of the housing 40) illuminates the connector pins 41 and the housing 40. The camera 4 captures an image in a state where the outline of the housing 40 of the connector is raised by the backlight 52 (for extracting the outline of the housing 40).

  On the other hand, in the setting mode, the transport device 7 is basically stopped, and the controller 2 directly controls the camera 4 and the illumination device 5 to capture a work image of the inspection object 8. The inspection object 8 to be imaged in the setting mode is a work (product) determined as a non-defective product by the user. In general, there is one work image. However, the inspection object 8 is long in the conveyance direction of the conveying device 7, and the entire inspection object 8 may not be imaged by one imaging. In this case, the controller 2 controls the camera 4 so that the inspection object 8 is imaged in a plurality of times. When a plurality of workpiece images are acquired in the setting mode, the same number of images are acquired even in the operation mode and compared with the workpiece image and the model image.

  In both the setting mode and the operation mode, the camera 4 captures an image with the front light 51 illuminating the connector pins 41 and the housing 40. Further, the camera 4 takes an image in a state where the outline of the connector housing 40 is raised by the backlight 52. When imaging the inspection object 8 by dividing it into a plurality of times, the PLC 3 may execute imaging a plurality of times by moving either the inspection object 8 or the camera 4. The movement of the inspection object 8 or the camera 4 may be manual. When the inspection object 8 is divided and imaged at three or more locations, the front light 51 is used for the first imaging location (left end in the drawing) in the transport direction and the last imaging location (right end in the drawing) in the transport direction. Then, the inspection object 8 is illuminated and imaged by the camera 4, and then the inspection object 8 is illuminated by the backlight 52 and imaged by the camera 4. On the other hand, for one or more intermediate imaging locations located between the first imaging location and the last imaging location, the inspection object 8 is illuminated by the front light 51 and imaged by the camera 4, and the backlight 52. The imaging using is omitted. This is because the contour of the intermediate portion of the inspection object 8 is not important for visual inspection. By omitting imaging using the backlight 52 for the intermediate portion, it is possible to improve the work efficiency of the appearance inspection.

  FIG. 5 shows an image captured by the front light 51 and an image captured by the backlight 52 by dividing the connector that is the inspection object 8 into three parts. The image 53 is an image at the left end of the connector imaged by the front light 51. The image 54 is an intermediate image of the connector imaged by the front light 51. The image 55 is an image at the right end of the connector imaged by the front light 51. The image 56 is an image at the left end of the connector imaged by the backlight 52. The image 57 is an image at the right end of the connector imaged by the backlight 52. In each of the images 53 to 55, the black solid portion existing under the connector is an image of the belt conveyor of the transport device 7. When the inspection object 8 is mounted on a jig or the like and conveyed by the conveying device 7, the jig is reflected in the image. As described above, imaging by the backlight 52 is omitted at the center of the connector (hereinafter referred to as the middle).

  Note that the camera 4 captures an image of the connector arranged at a predetermined position with the front light 51, and then captures an image with the backlight 52 while maintaining the position of the connector. That is, the position of the inspection object 8 is the same in the image captured using the front light 51 and the image captured using the backlight 52.

  The same portion of the connector is shown on the right end of the image 53 and the left end of the image 54. Similarly, the same portion of the connector is shown at the right end of the image 54 and the left end of the image 55. The reason that the same portion is duplicated in the adjacent partial images in this way is to create a whole image by connecting these three partial images. That is, the image processing unit 30 connects the two adjacent partial images so that the same pins appearing in common in the two adjacent partial images overlap. Thus, the entire image created includes the entire connector.

<Dimension measurement>
● Distance between Pin and Corner FIG. 6 is a diagram showing an example of dimension measurement. A user interface (UI) 60 displayed on the monitor 10 includes a single image (frontlight image 61) configured by connecting a plurality of partial images captured using the frontlight 51, and a backlight 52. An image (backlight image 62) configured by connecting a plurality of partial images picked up using is shown. The front light image 61 and the backlight image 62 are whole images created by connecting partial images of the same subject captured by switching the front light 51 and the backlight 52. Therefore, the coordinate system of the front light image 61 and the coordinate system of the backlight image 62 are the same. Here, the coordinate system of the frontlight image 61 has the left end of the frontlight image 61 as the origin, and the coordinate system of the backlight image 62 has the left end of the backlight image 62 as the origin. Therefore, the center position of each pin in the front light image 61 and the center position of each pin in the backlight image 62 can be expressed by the same coordinates. Similarly, the position of the corner 64 of the housing 40 in the front light image 61 and the position of the corner 64 of the housing 40 in the backlight image 62 can be expressed by the same coordinates.

  In the UI 60, the front light image 61 and the backlight image 62 are simultaneously displayed so that both ends in the horizontal direction of the front light image 61 coincide with both ends in the horizontal direction of the backlight image 62, respectively. Is done. That is, the front light image 61 and the backlight image 62 are displayed side by side with the horizontal edge of the front light image 61 aligned with the horizontal edge of the backlight image 62. Thus, the user can easily understand that both the front light image 61 and the backlight image 62 are images acquired by imaging the same inspection object 8. Further, the search area, inspection area, and detection point set in the front light image 61 may be set in the backlight image 62 as they are. That is, in the front light image 61 and the backlight image 62, the search area, the inspection area, and the detection point may be displayed in conjunction with each other. Similarly, the search area, inspection area, and detection point set in the backlight image 62 may be set in the front light image 61 as they are, and both may be displayed in conjunction with each other. As a result, a feature portion that is difficult to set in one image can be set using the other image. In the present invention, it is possible to set various measurement objects such as the dimensions of the inspection object 8 by utilizing such characteristics.

  In FIG. 6, the image processing unit 30 calculates the distance from the corner 64 of the housing 40 to the center of the first pin 63. The connector of this example has 17 pins arranged in two rows per row. As for the pin numbers, the leftmost pin is the first pin and the rightmost pin is the 17th pin.

  By the way, when the color of the housing 40 is close to white, the position of the corner 64 may not be accurately detected from the front light image 61. On the other hand, in the backlight image 62, the corner 64 can be detected with high accuracy. This is because the contrast between the background and the housing is high. The image processing unit 30 determines the position (coordinates) of the first pin 63 from the front light image 61 and determines the position of the corner 64 from the backlight image 62. Here, since the coordinate systems of the front light image 61 and the backlight image 62 completely coincide with each other, the position of the first pin 63 obtained from the front light image 61 and the angle 64 obtained from the backlight image 62 are obtained. From this position, the image processing unit 30 can calculate the distance between them. The distance from the 17th pin to the upper right corner of the housing 40 can be calculated from the front light image 61 and the backlight image 62 in the same procedure.

  FIG. 7 is a diagram illustrating an example of dimension measurement regarding the pins in the second row. Regarding the distance from the first pin 65 in the second row to the corner 64, the distance can be calculated in the same procedure as the first pin 63 in the first row.

  Here, pin 1 and pin 17 have been described, but for pin 2 to pin 16, the position of the pin that is a feature point is detected from front light image 61, and the corner that is the feature point from backlight image 62 is detected. Can be detected and the distance between them can be determined.

● Distance from pin to housing outer edge (pin height)
FIG. 8 shows a method for obtaining the height of each pin. In this example, the image processing unit 30 calculates the distance from the center (reference point 82) of each pin to the outer edge on the upper side of the housing 40 as the pin height.

  In order to determine the upper outer edge of the housing 40 from the backlight image 62, the image processing unit 30 detects the position of the upper right corner and the position of the upper left corner, and uses a straight line (straight line equation) connecting the two as a reference line 81. Determine as. The reason why the backlight image 62 is used is that corners can be detected more accurately than the front light image 61. The image processing unit 30 detects the center position of each pin as a reference point 82 from the front light image 61. The image processing unit 30 calculates the distance from the reference point 82 of each pin to the reference line 81 as the height of each pin.

  Here, the reference point 82 is detected from the front light image 61 and the reference line 81 is detected from the backlight image 62. However, when parameters affecting the contrast of the image such as the outer shape, color scheme, and material of the inspection object 8 are different, the image processing unit 30 detects the reference line 81 from the frontlight image 61 and the reference image 81 from the backlight image 62. A point 82 may be detected.

  FIG. 9 shows an example in which the pin height is obtained for each pin in the second row. In order to determine the upper outer edge of the housing 40 from the backlight image 62, the image processing unit 30 detects the position of the upper right corner and the position of the upper left corner by image recognition, and forms a straight line (straight line equation) connecting the two. The reference line 81 is determined. Further, the image processing unit 30 detects the center position of each pin in the second row from the front light image 61 as the reference point 85. The image processing unit 30 calculates the distance from the reference point 85 of each pin to the reference line 81 as the height of each pin.

<Setting the number of imaging>
10 and 11 show a dialog 101 for setting the number of times of imaging. The dialog 101 includes an input field 102 for the number of image divisions and a selection field 103 for the number of cameras and the imaging technique. When the number of divisions is set to 1, it means that the entire inspection object 8 is contained in one image. When the division number is set to 3, it means that the entire inspection object 8 is divided into three images. As shown in FIG. 11, in the selection column 103, an imaging method for imaging by changing illumination only at both ends of the inspection object 8 can be selected.

  FIG. 12 shows a dialog 104 that visually indicates the number of times of imaging set by the user. In this example, imaging is executed at three imaging locations by the illumination A (front light 51), and imaging is executed at two imaging locations except for the middle by the illumination B (backlight 52). It has been shown that. An imaging order 105 is also shown in order to clearly indicate the imaging order to the user. In this example, it is shown that imaging is first performed at the first imaging location using the illumination A, and second imaging is performed at the first imaging location using the illumination B. This makes it easier for the user to understand that the inspection object 8 should be moved when the first and second imaging operations are completed.

<Connector displacement correction>
When an image for pass / fail determination is taken while the inspection object 8 is conveyed by the conveying device 7, the position of the inspection object 8 may be shifted from the ideal position. Here, the ideal position is an arrangement position of a non-defective work when a model image for pattern matching is captured by the camera 4. Since the non-defective workpiece is manually and accurately arranged, it is accurately positioned with respect to the camera 4. Therefore, in order to improve the accuracy of pattern matching and pass / fail judgment when an image of the inspection object 8 is picked up in an actual production line, positional deviation correction such as rotating the picked-up image or shifting left and right and up and down. Is required.

  By the way, as described above, the same part of the inspection object 8 may be imaged a plurality of times by changing the illumination method. For example, in the front light image 61 and the backlight image 62, the same location is imaged without moving the inspection object 8, so that the respective positional deviation amounts should match. Therefore, the correction amount for the backlight image 62 is set using the positional deviation amount obtained from the front light image 61 or the correction amount for the image at each imaging location. When the connector is misaligned, the connector position is detected and the misalignment is corrected in order to correctly inspect in the subsequent inspection.

  As described above, in this embodiment, the image processing unit 30 obtains the correction amount of the positional deviation from one image for a plurality of images obtained by imaging the same portion of the inspection object 8, and the plurality of images are obtained with the obtained correction amount. Each position is corrected, and pass / fail judgment is executed using the corrected image.

  FIG. 13 is a diagram showing that the connector category 108 is selected by the mouse from the tool catalog 111 showing the categories of the plurality of image processing tools, and further, connector position deviation correction is selected as the image processing tool.

  FIGS. 14, 15, and 16 show UIs for setting parameters necessary for connector displacement correction (FIGS. 14 to 16 are UIs corresponding to images 1A to 3A, respectively). Here, two search areas 112 are set by the mouse 9. As shown in the right side of FIG. 14, a model image (search pattern) is registered in advance by the user, and when the search area 112 is set by the mouse 9, the pattern image matches the model image in the search area 112. Search in real time whether there is a part. When a model image is found, it is displayed on the screen as an inspection area 113. That is, in FIG. 14, a model image is found in the search area 112, and the inspection area 113 is displayed. Here, the search area 112 is an area indicating a range in which the model image can be moved. The inspection area 113 is an area including a portion (model image) to be subjected to pattern matching (as described above, an area where a model image is found in FIG. 14). The reason for setting the two search areas 112 is to take into account manufacturing tolerances of the inspection object or to accurately correct the position / tilt deviation during the conveyance of the inspection object. If the positions of the two inspection areas 113 are found in the two search areas 112, the displacement amount of the connector position can be calculated therefrom.

  As shown in FIGS. 14, 15, and 16, two search areas 112 for correcting misalignment and two inspection areas 113 are set at the left end, the middle, and the right end of the connector, respectively. . Since the positional deviation amounts at the left end portion, middle portion, and right end portion of the connector may be the same, the positional deviation amount may be obtained from only one of the left end portion, middle portion, and right end portion. However, in order to improve the accuracy of the misregistration correction, the misregistration correction amount may be obtained from three locations, the left end portion, the middle portion, and the right end portion.

  FIG. 17 is an example of a UI indicating that the setting for correcting the connector displacement has been completed. The image processing unit 30 uses the parameters set by the mouse 9 to execute the positional deviation correction for the non-defective product captured by the camera 4. When the positional deviation correction is successful, the OK mark 114 indicating success is displayed on the monitor 10. May be displayed.

<Addition of connector dimension inspection tool>
FIG. 18 is a diagram showing that the connector category 108 is selected by the mouse from the tool catalog 111 showing a plurality of image processing tool categories, and the connector dimension inspection tool 109 is selected as the image processing tool. When a button indicating the connector dimension inspection tool 109 is clicked, a transition is made to a setting screen. An additional button may be clicked to transition to the setting screen.

● Image connection parameter setting The image connection parameter is one of the setting items of the connector dimension inspection tool. Note that this parameter setting is not necessary when images are not connected.

  FIG. 19 shows a setting screen for setting connection points of images. On this screen, the arrangement order 116 of images and a connection method 117 for specifying a connection method can be selected by the mouse 9. For example, when “left → right” is selected as the image arrangement order 116, it indicates that an image of the inspection object 8 is picked up from the left to the right. When “connect by pin array” is selected as the connection method 117, the image processing unit 30 detects the pin array from each image, and connects a plurality of images so that the pin array becomes one line as a whole.

Setting of Pin Array Detection Parameters FIG. 20 shows a setting screen for pin array detection parameters. In this example, the pin arrangement 118 is a setting item for setting how many rows the pins are arranged. The position 119 of the 1st pin is a setting item for setting which pin is the 1st pin among a plurality of pins arranged in a row. In this example, since the connector that is the inspection object 8 has two rows, “2 rows” is selected by the mouse 9 as the pin arrangement 118.

● Pin model registration The connector has a large number of pins, and all the pins need to be inspected. Therefore, it is necessary to register the model image of each pin to be inspected and its position for non-defective product determination.

  FIG. 21 shows a pin model registration screen. The CPU 22 enlarges and displays the image 53 at the left end of the connector in the registration screen. This is to facilitate registering the pin model. The user sets the inspection area 113 to include the first pin 63. The CPU 22 stores the image of the portion surrounded by the inspection area 113 as a model image in the memory. The position and size of the inspection area 113 are changed by the CPU 22 in conjunction with the operation of the mouse 9.

  FIG. 22 shows a pin detection setting screen. The CPU 22 arranges the reference point 120 at the center of the inspection area 113 and arranges the search area 112 so as to surround the inspection area 113. The position of the inspection area 113, the position of the reference point 120, and the position of the search area 112 are stored in the memory by the CPU 22. As described above, the inspection region 113 indicates a region detected as a result of pattern matching with the model image set in FIG. 21 in the search region 112. That is, in FIG. 22, the inspection area 113 shows a detection result that pattern matches with the model image.

  As described above, the search area 112, the inspection area 113, the reference point 120, etc. must be set for all pins. However, if this setting operation is left to the manual operation, the user will be burdened with a great burden. Therefore, in this embodiment, the user sets the number of pins 121, which is one of the setting items, with the mouse 9, thereby setting the search area 112, the inspection area 113, and the reference point 120 for each pin after the second pin. Can be semi-automated. The CPU 22 copies the search area 112, the inspection area 113, and the reference point 120 in the right direction according to the input number of pins 121, so that the search area 112, the inspection area 113, and the reference area for each of the six pins. Point 120 is set. This is based on the premise that the pins are arranged at equal intervals, and the sizes of the pins also match. The user may finely adjust the search area 112, inspection area 113, and reference point 120 set for each pin by the CPU 22 with the mouse 9. The CPU 22 corrects the coordinates (positions) of the search area 112, the inspection area 113, and the reference point 120 according to the input from the mouse 9, and stores them in the memory.

  23 and 24 show setting screens for the search area 112, the inspection area 113, and the reference point 120 for the middle part and the right end part of the connector. The search region 112, the inspection region 113, and the reference point 120 are set in the middle and right end portions of the connector in the same procedure as the left end portion. That is, the CPU 22 uses the data of the search area 112 of the first pin, the inspection area 113, the reference point 120, and the number of pins 121, and the search area 112, the inspection area 113, A reference point 120 is set.

  FIG. 25 shows a screen indicating that registration of the pin model in the first row has been completed. In this screen, the entire image 122 of the connector is displayed together with the search area 112, the inspection area 113, and the reference point 120. Note that the CPU 22 may display the total number of pins 123 on the monitor 10 by summing up the number of input pins.

  21 to 25, the first row of pins has been described, but the pin registration can be executed for the second row of pins in the same procedure.

● End detection parameter setting When the appearance inspection of the connector is executed, it is necessary to detect the end (corner) position of the connector as described above.

  FIG. 26 is a diagram illustrating a setting screen for performing settings related to connector end detection. The number of ends to be detected varies depending on the type of connector. Therefore, a necessary number of ends are set by pressing the add button 124 with the mouse 9.

  FIG. 27 shows a selection item 125 for selecting an end to be detected. Depending on the type of connector, there may be a case where the metal provided in the connector should end, a case where the connector housing should end, and a case where a jig for conveying the connector should end. Therefore, the user selects an end detection target by operating the mouse 9 from the selection item 125 according to the type of the connector that is the inspection object 8.

  FIG. 28 is a diagram showing a registration screen for a search model for edge detection. In order to set the left end of the workpiece, the front light image (imaging 1A) at the left end of the connector is selected from the image selection unit 126 by the mouse 9. The CPU 22 enlarges and displays the leftmost image to facilitate registration of the search model. An inspection area 127 including an end is set by operating the mouse 9. When the CPU 22 detects that the model registration button 228 has been pressed, the CPU 22 stores in the memory the position of the inspection area 127 set at that time and the model image surrounded by the inspection area 127.

  FIG. 29 shows a setting screen for the search area 128. The CPU 22 sets a rectangular area set in accordance with the operation of the mouse 9 as an end detection search area 128. Here, the CPU 22 performs a search in the search area 128 for the model image of the inspection area 127 registered as a model, and determines whether the model image is found. As shown in FIG. 29, when the contrast between the background and the housing is low, end detection fails.

  Therefore, in this embodiment, the edge detection parameters are set using the backlight image.

  FIG. 30 is a diagram showing a registration screen for a search model for edge detection. The backlight image (imaging 1B) at the left end of the connector is selected by the mouse 9 from the image selection unit 126. The CPU 22 enlarges and displays the leftmost backlight image to facilitate registration of the search model. An inspection area 127 including an end is set by operating the mouse 9. When the CPU 22 detects that the model registration button 228 has been pressed, the CPU 22 stores in the memory the position of the inspection area 127 set at that time and the model image surrounded by the inspection area 127.

  FIG. 31 shows a setting screen for the search area 128. The CPU 22 sets a rectangular area set in accordance with the operation of the mouse 9 as an end detection search area 128. Here, the CPU 22 performs a search in the search area 128 for the model image of the inspection area 127 registered as a model, and determines whether the model image is found. As shown in FIG. 31, when the contrast between the background and the housing is high, the edge detection is successful. When a reference point designation button 130 for designating a reference point is clicked with the mouse 9, the CPU 22 causes the monitor 10 to display a reference point designation screen.

  32 and 33 are diagrams showing a designation screen for the reference point 131. FIG. 32 shows the reference point 131 before fine adjustment, and FIG. 33 shows the reference point 131 after fine adjustment. The CPU 22 arranges the reference point 131 in the middle of the inspection area 127. The user finely adjusts the reference point 131 by operating the mouse 9, and the CPU 22 stores the position of the reference point 131 after the fine adjustment in the memory.

  Note that the inspection region 127, the search region 128, and the reference point 131 are set for the right end of the work, which is the right end of the connector, in the same procedure as that for the left end of the work.

  FIG. 34 is a screen showing that the work left end and the work right end are set normally. In this example, it is shown that the left end of the work and the right end of the work are set using the backlight image.

Setting Reference Line FIG. 35 shows a setting screen for setting the reference line 81. When the add button (edit button) 140 is clicked, a transition is made to a detailed setting screen for setting each parameter of the reference line 81.

  FIG. 36 shows selection items 141 for the method of specifying the reference line 81. Here, the mouse 9 designates which line is used as the reference line 81 for the connector which is the inspection object 8.

  FIG. 37 is a diagram showing a screen when “line connecting both ends of the housing” is designated as the reference line 81 from the selection item 141. As described above, both ends (work left end and work right end) of the housing are set in advance using the backlight image. Therefore, when the “line connecting both ends of the housing” is designated as the reference line 81, the CPU 22 sets the reference line 81 so as to pass through the work left end and the work right end. Note that the left end of the work corresponds to the reference point 131. In addition, although illustration is abbreviate | omitted, the workpiece | work right end is also set with the reference point.

● Dimension measurement parameter setting Here, set which part of the connector dimensions is the measurement target.

  FIG. 38 is a diagram showing a main screen for setting dimension measurement parameters. By clicking the item selection button 150 with the mouse 9, the user can select the dimension to be measured.

  FIG. 39 is a diagram showing an item selection screen for setting dimension measurement parameters. Here, a dimension measurement category 151 that indicates which part of the connector is measured for each category, and a measurement item list 152 that indicates one or more measurement items corresponding to the selected dimension measurement category are displayed. The CPU 22 switches the measurement item list 152 according to the category selected by the mouse 9. In the example shown in FIG. 39, when the housing is selected from the dimension measurement category 151, various measurement items regarding the housing are shown. The housing height is the height of the housing starting from the reference line. The housing distance is a distance from the left end to the right end of the housing. The distance between the housing pins (ends) is the distance from the end of the housing to the end pins. The housing bracket distance is the distance from the housing to the bracket, for example, the distance from the end of the housing to the center of the bracket. A measurement item selected from the measurement item list 152 by the mouse 9 is added as a measurement item to be actually executed.

  FIG. 40 shows a screen for setting parameters used for pass / fail judgment for the selected measurement item. In the parameter setting unit 160, a design standard value and a positive / negative tolerance for the distance between the housing pins (ends) for the pins in the first row are input through the mouse 9. The CPU 22 may calculate an upper limit value and a lower limit value of the dimensions from the standard value and the tolerance and display them on the parameter setting unit 160. Further, the CPU 22 uses the image processing unit 30 to actually measure the dimensions from the work image, and obtains and displays an average value of the measured value for the first pin and the measured value for the 17th pin, The value and the maximum value may be displayed.

  FIG. 41 shows a screen for setting parameters used for pass / fail judgment for the selected measurement item. In the parameter setting unit 161, a design standard value and a positive / negative tolerance for the distance between the housing pins (ends) for the pins in the second row are input through the mouse 9. The CPU 22 may calculate the upper limit value and the lower limit value of the dimensions from the standard value and the tolerance and display them on the parameter setting unit 161. Further, the CPU 22 uses the image processing unit 30 to actually measure the dimensions from the work image, and obtains and displays an average value of the measured value for the first pin and the measured value for the 17th pin, The value and the maximum value may be displayed.

  FIG. 42 is a diagram showing an item selection screen for setting dimension measurement parameters. Here, since the pin height is selected from the dimension measurement category 151, the measurement item list 152 displays the measurement items related to the pin height. The user clicks and selects a measurement item with the mouse 9. Here, it is assumed that the pin height from the reference line is selected.

  FIG. 43 shows a screen for setting parameters used for pass / fail judgment for the selected measurement item. In order to make the measurement items easy to understand, the image display unit 162 displays a reference line, a reference point, a pin number, a mark indicating which part is measured, and the like along with the work image. Here, both a front light image and a backlight image are displayed as work images. In the parameter setting unit 163, a design standard value and a positive / negative tolerance for the pin height for the pins in the first row are input through the mouse 9. The CPU 22 may calculate the upper limit value and the lower limit value of the dimensions from the standard value and the tolerance and display them on the parameter setting unit 163. Further, the CPU 22 uses the image processing unit 30 to actually measure the dimensions from the work image, and obtains and displays the average value of the measured values for each pin from the 1st pin to the 17th pin, or displays the minimum and maximum values. And a value may be displayed. From these numerical values, it will be easy to confirm whether or not the work (product) used for capturing the work image is appropriate.

  FIG. 44 shows a screen for setting parameters used for pass / fail judgment for the selected measurement item. In the parameter setting unit 164, parameters regarding the pin height for the pins in the second row are input. The CPU 22 may calculate an upper limit value and a lower limit value of the dimensions from the standard value and tolerance input to the parameter setting unit 164 and display them on the parameter setting unit 164. Further, the CPU 22 uses the image processing unit 30 to actually measure the dimensions from the work image, and obtain and display the average value of the measured values for each pin from the 1st pin to the 17th pin in the second row, The minimum value and the maximum value may be displayed.

<Modification of image connection>
In the above-described embodiment, imaging is performed so that one or more common pins are included between adjacent images, and two adjacent images are connected by overlapping the common pins, The entire image of the inspection object 8 has been described as being created by the CPU 22 or the image processing unit 30. However, a portion that should be included in common between adjacent images does not have to be a part of the inspection object 8. That is, a mark serving as a reference for connection may be given to a member that moves together with the inspection object 8, such as a jig or a belt conveyor.

  FIG. 45 is a diagram illustrating an example of a mark serving as a reference for connecting a plurality of images. In this example, a connection mark 170 is provided on a jig that conveys the inspection object 8. When “connect at an arbitrary point” is selected as the connection method 117, the CPU 22 can execute connection using the mark 170 as a reference.

  FIG. 46 is a diagram showing a setting screen for setting a mark 170 that is a reference for image connection. The CPU 22 enlarges and displays the left end image of the connector, thereby making it easy to set the reference point 171, the inspection area, and the search area. The user operates the mouse 9 to set the reference point 171 at the center of the mark 170. Further, the CPU 22 sets an inspection area and a search area with the reference point 171 as the center. The CPU 22 may finely adjust the reference point 171 according to the operation of the mouse 9, or may adjust the position and size of the inspection area and the search area.

  FIG. 47 is a diagram showing a setting screen for setting a mark 170 that serves as a reference for image connection. The CPU 22 enlarges and displays the image of the middle part of the connector so that the reference point 172, the inspection area and the search area, which serve as a reference for connecting the left end image and the intermediate part image, the intermediate part image and the right end image It is easy to set a reference point 173, an inspection area, and a search area as a reference for connecting the two. The user operates the mouse 9 to set a reference point 172 at the center of the left mark 170 and a reference point 173 at the center of the right mark 170. Further, the CPU 22 sets an inspection area and a search area around the reference point 172 and the reference point 173, respectively. The CPU 22 may finely adjust the reference points 172 and 173 according to the operation of the mouse 9, or may adjust the positions and sizes of the inspection area and the search area.

  FIG. 48 is a diagram illustrating a setting screen for setting a mark 170 that serves as a reference for image connection. The CPU 22 enlarges and displays the right end image of the connector, thereby making it easy to set a reference point 174, an inspection area, and a search area as a reference for connecting the intermediate image and the right end image. The user operates the mouse 9 to set the reference point 174 at the center of the mark 170. Further, the CPU 22 sets an inspection area and a search area with the reference point 174 as the center. The CPU 22 may finely adjust the reference point 174 according to the operation of the mouse 9, or may adjust the position and size of the inspection area and the search area.

  FIG. 49 shows a front light image and a backlight image connected with the mark 170 as a reference. When connecting the left end image, the intermediate portion image, and the right end image of the connector, the image processing unit 30 connects the left end image and the intermediate portion image so that the reference point 171 and the reference point 172 overlap each other. Further, the image processing unit 30 connects the intermediate portion image and the right end image so that the reference point 173 and the reference point 174 overlap each other. Thereby, a connected image (entire image) in which the entire inspection object 8 is accommodated is created. Since the coordinate systems of the front light image and the backlight image are the same, the backlight image can be similarly connected on the basis of the connection position of the front light image. As shown in FIG. 49, when only the leftmost image and the rightmost image exist for the backlight image, each origin (the upper left corner of the image) of the leftmost image of the frontlight image and the leftmost image of the backlight image. ) Match. In addition, the origins (upper left corner of the image) of the right end image of the front light image and the right end image of the backlight image are also matched. As a result, as shown in FIG. 49, a front light image and a backlight image showing the entire connector are created.

<Pin and housing detection method>
The above-described embodiment is an example in which the housing and the pins of the inspection object 8 are detected using pattern matching using a model image. However, a model image is not always necessary.

  FIG. 50 is a diagram illustrating a pin detection method using edge detection. The image processing unit 30 detects a plurality of horizontal edges and vertical edges by executing edge detection. Then, the detected intermediate position (center position) of the edge in the predetermined direction is automatically set as the reference point 200. The image processing unit 30 calculates the distance from the reference point 200 to each edge, and searches for a combination of edges whose distance is a predetermined distance (allowed if within a certain tolerance range). In this example, an edge 201 corresponding to the left end of the pin 41 and an edge 202 corresponding to the right end are recognized as edges that match the pin detection condition. Similarly, the image processing unit 30 recognizes the edge 203 corresponding to the upper end of the pin 41 and the edge 204 corresponding to the lower end as edges that match the pin detection condition. This is because the distance from the reference point 200 matches the pin detection condition. In this way, the image processing unit 30 can detect the position of each pin by performing edge detection on the entire image.

  FIG. 51 is a diagram illustrating a housing detection method using edge detection. The image processing unit 30 detects a plurality of horizontal edges and vertical edges, and extracts an edge 211 and an edge 212 that are orthogonal to each other at the reference point 210. Here, the edge 211 and the edge 212 are the outermost edges. The reference point (intersection point) 210 is automatically set as a point where the edge 211 and the edge 212 intersect after the edge 211 and the edge 212 are extracted. By setting such detection conditions in advance, the image processing unit 30 may detect the end of the housing using edge detection.

<About internal processing>
Hereinafter, functions realized by the CPU 22 and the image processing unit 30 will be described. Each function may be realized only by the CPU 22, may be realized only by the image processing unit 30, or may be realized by a cooperative operation between the CPU 22 and the image processing unit 30.

  FIG. 52 is a functional block diagram of the appearance inspection apparatus 1. As already described with reference to FIG. 2, the illumination device 5 functions as an illumination unit that irradiates the inspection target 8 with a plurality of different illumination lights. The illumination control unit 26 controls the illumination device 5 so as to switch the illumination light emitted by the illumination device 5 in accordance with an instruction from the CPU 22 or the image processing unit 30. Camera 4 It functions as an imaging unit that images the inspection object 8 irradiated with illumination light by the illumination device 5. The appearance inspection apparatus 1 is an apparatus that executes measurement processing on the image of the inspection object 8 acquired by the camera 4.

  The imaging control unit 301 is a control unit that commands the illumination control unit 26 to switch illumination light such as a front light and a backlight, and commands the camera 4 to perform imaging. The imaging control unit 301 also takes charge of control of the camera 4 and control of the illumination device 5 when the inspection object 8 is divided and imaged.

  The first image storage unit 304 is a predetermined region of the inspection object 8 irradiated with the first illumination light (e.g., front light) by the illumination device 5 (e.g., the entire connector or the left end, middle, and right end of the connector). Etc.) is stored in the first image (e.g., a front light image or a partial image thereof) acquired by the camera 4. The second image storage unit 305 is a second image (the second image (captured by the camera 4) acquired by the camera 4 capturing a predetermined area of the inspection object 8 irradiated with the second illumination light (eg, backlight) by the illumination device 5. Example: Backlight image or its partial image) is stored. The first image storage unit 304 and the second image storage unit 305 are realized by the work memory 23, for example. The display control unit 28 and the monitor 10 function as a display unit that displays the first image and the second image.

  The UI control unit 310 controls a display-type user interface such as the above-described UI 60 and an input-type interface such as a mouse 9 and a keyboard. The UI control unit 310 is further divided into various functional units. As described with reference to FIG. 21 to FIG. 25 and the like, the first measurement target setting unit 311 performs inspection on one of the first image and the second image displayed on the monitor 10. A feature portion (for example, a detection point of a pin) that is a starting point for measuring a dimension related to the object 8 is set.

  As described with reference to FIG. 30 to FIG. 37, the second measurement target setting unit 312 is the inspection target in the other of the first image and the second image displayed on the monitor 10. A feature portion (for example, a detection point for detecting a corner of the housing or a reference line connecting a plurality of detection points) is set as an end point for measuring a dimension related to the object 8. Data such as model images, reference points, and reference lines set by the first measurement target setting unit 311 and the second measurement target setting unit 312 are stored in the feature data storage unit 306. The feature data storage unit 306 is also realized by the work memory 23.

  The dimension calculation unit 302 calculates the dimension from the feature part serving as the starting point set by the first measurement target setting unit 311 to the feature part serving as the end point set by the second measurement target setting unit 312. Here, the dimensions are distance, area, and the like.

  The UI control unit 310 may include a threshold setting unit 313 that sets a threshold for determining whether the appearance of the inspection object 8 is good or bad. The quality determination unit 303 compares the size calculated by the size calculation unit 302 with a threshold value, and determines the quality of the appearance of the inspection object 8. For example, if the pin height obtained for the plurality of pins by the dimension calculating unit 302 exceeds the upper limit value obtained from the tolerance or the standard value by the threshold setting unit 313, the pass / fail judgment unit 303 determines that it is defective.

  As described with reference to FIG. 6 and the like, the UI control unit 310 sets the first image and the first image so that the horizontal end of the first image matches the horizontal end of the second image. Display data of the user interface is created so that the monitor 10 displays the two images at the same time, and is passed to the display control unit 28. That is, the UI control unit 310 causes the monitor 10 to vertically align the first image and the second image by aligning the horizontal end of the first image and the horizontal end of the second image. Create display data to display side by side. In addition, the UI control unit 310 displays an index (for example, a search area, an inspection area, and a detection point) indicating a feature portion that is a starting point set for one of the first image and the second image. Display data may be generated so that the other image of the first image and the second image is displayed in conjunction with the other image (at least the detection point). It is preferable that the cross mark shown is displayed in conjunction with it). Note that the UI control unit 310 uses the first image, the second image, and the index indicating the characteristic portion that is the end point set for the other image of the first image and the second image. Display data for displaying in conjunction with one of the images may be created. As described above, in the UI 60, the search area, the inspection area, the cross mark indicating the detection point, and the like are displayed in conjunction with each other between the front light image and the backlight image. For example, the UI control unit 310 sets and displays the search area, inspection area, and detection points set using the backlight image as they are in the front light image. If the origin of the backlight image is the upper left corner of the backlight image and the origin of the front light image is the upper left corner of the front light image, the search area, inspection area, detection point coordinates set using the backlight image, and the front The search area, inspection area, and detection point coordinates in the light image are matched. Thereby, the UI control unit 310 sets the search area, the inspection area, and the detection point set by using the backlight image, the search area in the frontlight image, and the inspection, when the origin of the frontlight image is the upper left corner of the frontlight image. Areas and detection points can be displayed in conjunction with each other.

  The first measurement target setting unit 311 and the second measurement target setting unit 312 have, as characteristic portions, reference points 82, 85, 120, 131, 171, 172, 173, and 174 that are one or more coordinates on the image. , 200, 210 are set.

  As described with reference to FIG. 21 and the like, the UI control unit 310 may create display data for the monitor 10 to enlarge and display one image. The first measurement target setting unit 311 sets a characteristic part in one image enlarged and displayed on the monitor 10 according to the operation of the mouse 9.

  As described with reference to FIG. 30 and the like, the UI control unit 310 may create display data for the monitor 10 to enlarge and display the other image. The second measurement target setting unit 312 sets a characteristic part in the other image enlarged and displayed on the monitor 10 in accordance with the operation of the mouse 9.

  As described with reference to FIGS. 35 to 37, the second measurement target setting unit 312 sets a reference line 81, which is a straight line connecting two reference points designated for the other image, as a feature portion. May be. In particular, when the corner of the housing 40 is set as a reference point, it becomes difficult to detect the corner in the front light image. Therefore, by setting the corner of the housing 40 as a reference point for the backlight image, the corner position can be accurately detected. This also makes it possible to accurately calculate the distance from the center of the pin to the corner. Furthermore, the accuracy of product quality determination is also improved.

  As described with reference to FIG. 5 and the like, the image linking unit 321 starts from the left in accordance with the order in which a plurality of partial images obtained by the camera 4 dividing and imaging the inspection object 8 a plurality of times are captured. A first image and a second image are created by connecting them side by side to the right. Here, the image connecting unit 321 uses the coordinates of the connecting positions of a plurality of partial images constituting the first image (for example, pin positions and mark positions shown in common) to generate the second image. You may connect the some partial image which comprises. Note that the UI control unit 310 may further include a connection position setting unit 314 that sets a connection position in the first image. The connection position setting unit 314 may set the position of a part (for example, a pin or a metal fitting) of the inspection object that is commonly shown in adjacent partial images as the connection position. As described with reference to FIGS. 45 to 49, the connection position setting unit 314 positions the part of the jig (for example, the mark 170) that fixes the inspection object that is commonly shown in the adjacent partial images. May be set as the connection position.

  The image connecting unit 321 arranges the left end partial image and the right end partial image of the inspection object 8 at both ends of the second image, omits the intermediate image of the inspection object 8, and displays the second image. You may create it.

  In FIG. 52, the image recognition unit 322 compares the model image showing a part of the non-defective product of the inspection object 8 set through the UI control unit with the image of the inspection object 8 flowing through the line acquired by the camera 4. Then, a portion that matches the model image is specified by pattern recognition. The image recognition unit 322 searches for a model image within a preset search area. An image area having the same size as the model image is an inspection area. The center of the inspection area that matches the model image is the reference point. Note that the reference point may be finely adjusted through the UI control unit. The coordinate data of the search area and the coordinate data of the reference point (detection point) are stored in the feature data storage unit 306 and are used by the image recognition unit 322, the image connection unit 321 and the coordinate calculation unit 323. The coordinate calculation unit 323 calculates the coordinates of the search area, the inspection area, and the reference point from the coordinate data indicating the position of the pointer of the mouse 9 and the coordinate data of the backlight image or the frontlight image.

  FIG. 53 is a flowchart showing a series of steps including a parameter setting process related to appearance inspection and non-defective product determination.

  In step S531, the imaging control unit 301 turns on the front light 51 through the illumination control unit 26, images the inspection object 8 (non-defective product), and acquires the front light image 61. The front light image 61 is stored in the first image storage unit 304.

  In step S532, the imaging control unit 301 turns on the front light 51 through the illumination control unit 26 and transfers the backlight 52.

  In step S <b> 533, the imaging control unit 301 acquires the backlight image 62 by imaging the inspection object 8 (non-defective product) illuminated by the backlight 52. The backlight image 62 is stored in the second image storage unit 305.

  Note that when the inspection object 8 (non-defective product) is imaged in multiple times and image connection is executed, S531 to S532 are repeatedly executed. As described above, in the backlight image 62, the intermediate image of the inspection object 8 (non-defective product) may be omitted.

  In step S534, the image connecting unit 321 connects adjacent partial images based on the connection position set by the connection position setting unit 314. As for the front light image, since there are all partial images of the inspection object 8 (non-defective product), the adjacent partial images are connected from left to right according to the order of photographing. When there is no intermediate partial image for the backlight image, the coordinates of the right edge image of the backlight image are set using the coordinates of the right edge image of the front light image. That is, the image connecting unit 321 arranges the partial images of the backlight image so that the left end and the right end of the backlight image coincide with the left end and the right end of the front light image, respectively.

  In S535, the first measurement target setting unit 311 sets a characteristic part (eg, pin) of the front light image. As described above, the first measurement target setting unit 311 sets the coordinates of the search area, the inspection area, and the detection point according to the operation of the mouse 9.

  In S536, the second measurement target setting unit 312 sets a characteristic portion (eg, an end of the housing) of the backlight image. As described above, the second measurement target setting unit 312 sets the coordinates of the search area, the inspection area, and the detection point according to the operation of the mouse 9. Here, the search area, inspection area, and detection point set in the front light image may be directly reflected as the search area, inspection area, and detection point of the backlight. Similarly, the search area, inspection area, and detection point set in the backlight image are directly reflected as the search area, inspection area, and detection point of the front light. Further, the second measurement target setting unit 312 may set the reference line used for the dimension measurement using either the front light image or the backlight image.

  In S537, the threshold setting unit 313 sets a threshold used for non-defective product determination. As described above, threshold values such as an upper limit value and a lower limit value are set from tolerances and standard values set through the mouse 9 or the like. Thereafter, the operation mode is entered.

  In step S538, the imaging control unit 301 controls the camera 4 and the illumination device 5 to image the product conveyed through the line by the conveyance device 7, and the image coupling unit 321 executes image coupling as necessary, and the dimensions The calculation unit 302 calculates dimensions for the front light image and the backlight image using preset search areas, inspection areas, and detection points.

  In S <b> 539, the quality determination unit 303 compares the dimension such as the distance calculated by the dimension calculation unit 302 with the threshold set by the threshold setting unit 313, and determines whether the product is non-defective or defective.

  As described above, according to the present embodiment, the appearance inspection apparatus 1 sets a plurality of different characteristic portions for dimension measurement with respect to the first image and the second image acquired under different illumination conditions. Can be measured. As described above, when calculating the distance between the pin and the corner of the housing, it is easy to recognize the position of the pin in the image obtained by the front light, but it is difficult to recognize the position of the corner of the housing. It was not possible to calculate. On the other hand, in the image obtained with the backlight, it is easy to recognize the position of the corner of the housing, but it is difficult to recognize the position of the pin, so the distance cannot be calculated accurately. Therefore, by detecting the position of the pin from the front light image and detecting the position of the corner of the housing from the backlight image, the distance between the two can be obtained with high accuracy, and the accuracy of the non-defective product determination is improved.

  Moreover, since two images obtained by changing the illumination conditions and imaging the same part of the inspection object 8 are displayed at the same time so that both ends thereof coincide with each other, It will be easier to set a reference point. In particular, when these images are displayed side by side, the end in the horizontal direction of the inspection object 8 in one image matches the end in the horizontal direction of the inspection object 8 in the other image. It will be easier to visually grasp the characteristic part to be measured.

  Further, by displaying an index indicating the characteristic portion set for one of the first image and the second image in conjunction with the other image, It will be possible to accurately set the feature portion while grasping which portion the feature portion corresponds to in another image.

  As described above, the characteristic portion includes a reference point and a reference line used for calculating the dimension. For example, if it is desired to provide a reference point (detection point) at the corner of the housing, it will be easier to set the reference point accurately if a backlight image is used. Similarly, a backlight image may be advantageous when it is desired to set a reference line along the housing outline. However, depending on the housing color, the pin color, and the background color, the front light image may be advantageous. In this case, in the embodiment described above, the front light image and the backlight image are used interchangeably.

  In addition, by enlarging and displaying the first image and the second image, it becomes easier to set the reference point more accurately. That is, the first measurement target setting unit sets a feature portion in one image enlarged and displayed on the monitor 10, and the second measurement target setting unit sets a feature portion in the other image enlarged and displayed on the monitor 10. In doing so, the user will be able to accurately grasp the feature and set the reference point.

  In the present embodiment, the image connecting unit 321 connects a plurality of partial images obtained by the camera 4 dividing and imaging the inspection object 8 a plurality of times from the left to the right according to the order of image capturing. To create a first image and a second image. At this time, it is desirable to use an image having higher connection accuracy among the first image and the second image when the connection position is designated.

  In the present embodiment, the image connecting unit 321 connects a plurality of partial images forming the second image using the coordinates of the connecting positions of the plurality of partial images forming the first image. Since the result of connecting the first images can be used, the amount of calculation required for connecting the second images can be greatly reduced. In particular, when an intermediate partial image is omitted from the second image, the amount of calculation required for connecting the second images can be further reduced.

  In the present embodiment, the illumination device 5 that switches between the front light 51 and the backlight 52 is illustrated. However, the present invention is applicable to any illumination device that irradiates the inspection object 8 with illumination light under different illumination conditions. For example, coaxial epi-illumination that highlights gloss, low-angle illumination that highlights the edges of scratches and marks, black-light illumination that shines black light, surface illumination that is used as a backlight, dome illumination that emits diffuse light from all directions An illumination device may be employed.

  Note that the imaging conditions may be changed, such as changing only the shutter speed of the camera 4 or changing the brightness of the illumination while using a single lighting device. This is because the same effect can be obtained as when the inspection object 8 is irradiated with substantially different illumination light. In this case, the appearance inspection apparatus 1 is a camera that images the illumination apparatus 5 that irradiates the inspection object 8 with illumination light and the inspection object 8 that is irradiated with the illumination light by the illumination apparatus 5 while switching the imaging conditions. 4 and the measurement process is executed on a plurality of images having different shooting conditions. The first image storage unit 304 stores a first image acquired by the imaging unit capturing an image of a predetermined area of the inspection object under the first imaging condition. The second image storage unit 305 stores a second image acquired by the imaging unit capturing an image of a predetermined area of the inspection object under the second imaging condition. Other operations are the same as the processing related to the first image and the second image acquired by changing the illumination condition.

Claims (20)

An illumination unit that irradiates the inspection target with a plurality of different illumination lights, an illumination control unit that controls the illumination unit so as to switch the illumination light emitted by the illumination unit, and the illumination unit irradiates the illumination light. An imaging unit that images the inspection object, and a communication unit that receives an imaging trigger signal that defines imaging timing of the imaging unit from an externally connected control device ,
An appearance inspection apparatus that determines the quality of the appearance of the inspection object using the image of the inspection object acquired by the imaging unit, and outputs a determination signal indicating the determination result ,
A first image storage unit that stores a first image acquired by the imaging unit imaging a predetermined region of the inspection object irradiated with the first illumination light by the illumination unit;
A second image storage unit for storing a second image acquired by the imaging unit imaging the predetermined region of the inspection object irradiated with the second illumination light by the illumination unit;
A display unit for displaying the first image and the second image;
Using the first image displayed on the display unit, a first setting for extracting a first feature portion that is a starting point when measuring a dimension related to the inspection object from the first image. And the second image displayed on the display unit is used to extract a second feature portion that is an end point when measuring a dimension related to the inspection object from the second image. A setting unit for executing the second setting of
When the imaging unit receives the imaging trigger signal, the imaging unit captures and acquires a new inspection object irradiated with the first illumination light from the illumination unit. The first characteristic portion is extracted based on the first setting by the setting unit, and the imaging unit captures and acquires the new inspection object irradiated with the second illumination light by the illumination unit. An image recognition unit that extracts the second feature portion from the second image based on the second setting by the setting unit;
A dimension calculation unit for calculating a dimension from the first feature part to the second feature part extracted by the image recognition unit ;
A quality determination unit that determines the quality of the appearance of the inspection object based on the dimension calculated by the dimension calculation unit and a predetermined threshold;
Means for outputting a determination signal indicating a determination result by the pass / fail determination unit;
An appearance inspection apparatus characterized by comprising: The setting unit uses the first image displayed on the display unit to set a first model image for extracting the first feature portion in the first setting, and the display A second model image for extracting the second feature portion in the second setting is set using the second image displayed in the section;
The image recognition unit searches the first image to extract a portion that matches the first model image as the first feature portion, and searches the second image to search for the second image. The appearance inspection apparatus according to claim 1, wherein a portion that matches a model image is extracted as the second feature portion. The setting unit sets a search area indicating a range for searching for a portion matching the first model image from the first image;
The appearance inspection apparatus according to claim 2, wherein the image recognition unit extracts a portion that matches the first model image in an area corresponding to the search area in the first image. The display unit is configured to enlarge and display the second image;
The setting tough the visual inspection according to claim 2 or 3, characterized in that it is configured to set the second feature portion of the enlarged displayed the second image on the display unit apparatus. The setting tough are claims characterized in that it is configured to set the reference line is a straight line connecting the two reference points specified for the second image as the second characteristic portion of the Item 5. The appearance inspection apparatus according to any one of Items 2 to 4. The display unit simultaneously displays the first image and the second image so that a lateral end of the first image matches a lateral end of the second image. it appearance inspection apparatus according to any one of claims 2 to 5, characterized in that said. The display unit displays the first image and the second image side by side by aligning the horizontal end of the first image and the horizontal end of the second image. is configured to, according to said index indicating a first of said set for the image first characteristic portion of the front Symbol display means displays in conjunction also for the second image Item 7. The appearance inspection apparatus according to Item 6. 7. the display unit, the index indicating the previous SL second the second feature portion which is set for the image, and displaying in conjunction also for the first image 2. An appearance inspection apparatus according to 1. The setting unit uses the second image displayed on the display unit to set an inspection region indicating a range for detecting an edge of the second feature portion in the second setting,
The said image recognition part extracts the edge detected within the area | region corresponding to the said test | inspection area | region among said 2nd images as said 2nd characteristic part. 2. An appearance inspection apparatus according to 1. The display unit is configured to enlarge and display the first image;
The setting tough has any one of claims 1 to 9, characterized in that it is configured to set the first characteristic portion of the enlarged displayed the first image on the display unit 2. An appearance inspection apparatus according to 1. A plurality of partial images obtained by the imaging unit dividing and imaging the inspection object a plurality of times are connected side by side from left to right or from right to left according to the order of imaging, and the first image and appearance inspection apparatus according to any one of claims 1 to 10, further comprising an image connecting unit that creates the second image.   The said image connection part is comprised so that the some partial image which comprises the said 2nd image may be connected using the coordinate of the connection position of the some partial image which comprises the said 1st image. 11. An appearance inspection apparatus according to 11.   The visual inspection apparatus according to claim 12, further comprising a connection position setting unit that sets the connection position in the first image.   The appearance according to claim 13, wherein the connection position setting unit is configured to set a position of a part of an inspection object that is commonly shown in adjacent partial images as the connection position. Inspection device.   The connection position setting unit is configured to set, as the connection position, a position of a part of a jig that fixes an inspection object commonly shown in adjacent partial images. Item 14. An appearance inspection apparatus according to Item 13.   The image connecting unit arranges a left end partial image and a right end partial image of the inspection object at both ends of the second image, omits an intermediate image of the inspection object, and performs the second image. The appearance inspection apparatus according to claim 12, wherein the visual inspection apparatus is configured to create   The illumination unit irradiates a first illumination unit that irradiates a first illumination light for extracting a surface feature of the inspection object and a second illumination light for extracting an outline of the inspection object. The visual inspection apparatus according to claim 1, further comprising a second illumination unit.   The visual inspection apparatus according to claim 17, wherein the first illumination unit irradiates the inspection object with illumination light by coaxial epi-illumination, and the second illumination unit irradiates the inspection object with illumination light by transmitted illumination. . An illumination unit that irradiates the inspection target with a plurality of different illumination lights, an illumination control unit that controls the illumination unit so as to switch the illumination light emitted by the illumination unit, and the illumination unit irradiates the illumination light. An imaging unit that images the inspection object, and a communication unit that receives an imaging trigger signal that defines an imaging timing of the imaging unit from an externally connected control device, and acquired by the imaging unit A method for controlling an appearance inspection apparatus that determines the quality of the appearance of the inspection object using an image of the inspection object and outputs a determination signal indicating the determination result ,
Storing a first image obtained by the imaging unit capturing and acquiring a predetermined area of the inspection object irradiated with the first illumination light by the illumination unit;
Storing a second image acquired by the imaging unit imaging the predetermined region of the inspection object irradiated with the second illumination light by the illumination unit;
Displaying the first image and the second image on a display unit;
Using the first image that is displayed on the display unit, the first setting for extracting a first feature portion as a starting point when measuring the dimensions for the inspection object from the first image And the second image displayed on the display unit is used to extract a second feature portion that is an end point when measuring a dimension related to the inspection object from the second image. Performing a second setting of
When the imaging unit receives the imaging trigger signal, the imaging unit captures and acquires a new inspection object irradiated with the first illumination light from the illumination unit. The first characteristic portion is extracted based on the first setting, and the imaging unit captures and acquires the new inspection object irradiated with the second illumination light by the illumination unit. Extracting the second feature portion from the image based on the second setting;
Calculating a dimension from the extracted first feature portion to the second feature portion ;
Determining the quality of the appearance of the inspection object based on the calculated dimensions and a predetermined threshold;
Outputting a determination signal indicating the determination result of the quality;
A control method for an appearance inspection apparatus, comprising: An illumination unit that irradiates the inspection target with a plurality of different illumination lights, an illumination control unit that controls the illumination unit so as to switch the illumination light emitted by the illumination unit, and the illumination unit irradiates the illumination light. An imaging unit that images the inspection object, and a communication unit that receives an imaging trigger signal that defines an imaging timing of the imaging unit from an externally connected control device, and acquired by the imaging unit A program executed on a computer that determines the quality of the appearance of the inspection object using an image of the inspection object and outputs a determination signal indicating the determination result ,
In the computer,
Storing a first image obtained by the imaging unit capturing and acquiring a predetermined area of the inspection object irradiated with the first illumination light by the illumination unit;
Storing a second image acquired by the imaging unit imaging the predetermined region of the inspection object irradiated with the second illumination light by the illumination unit;
Displaying the first image and the second image on a display unit;
Using the first image displayed on the display unit, a first feature portion for extracting a first feature portion serving as a starting point for measuring a dimension related to the inspection object from the first image . The setting is executed, and a second feature portion serving as an end point when measuring a dimension related to the inspection object is extracted from the second image using the second image displayed on the display unit. Performing a second setting for :
When the imaging unit receives the imaging trigger signal, the imaging unit captures and acquires a new inspection object irradiated with the first illumination light from the illumination unit. The first characteristic portion is extracted based on the first setting, and the imaging unit captures and acquires the new inspection object irradiated with the second illumination light by the illumination unit. Extracting the second feature portion from the image based on the second setting;
Calculating a dimension from the extracted first feature portion to the second feature portion ;
Determining the quality of the appearance of the inspection object based on the calculated dimensions and a predetermined threshold;
Outputting a determination signal indicating the determination result of the quality;
A program characterized by having executed.