Image inspection apparatus, control unit for image inspection apparatus, image inspection method, image inspection program, computer-readable recording medium, and recorded device.

Abstract

To easily capture an image for inspection.SOLUTION: An image inspection apparatus 100 includes: an image score calculation unit 22 which calculates image scores of work images generated in different image generation conditions; and an image generation condition setting unit 23 which receives, while thumbnail images corresponding to the work images having higher image score are displayed on a display unit 4, a selection of one thumbnail image, to receive settings for image generation condition corresponding to the selected thumbnail image. The image generation conditions include a single condition of generating an image captured in a single image generation condition and a composition condition of generating a composite image from images captured in different image generation conditions. The display unit 4 includes a reference information display column 205 for displaying reference information to identify whether each of the thumbnail images having higher image scores has been generated in the single condition or the composition condition when displaying the thumbnail images.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to an image inspection apparatus, a control unit for an image inspection apparatus, an image inspection method, an image inspection program, a computer-readable recording medium, and a recording device.

Background Art

[0002] An image inspection apparatus such as an image sensor that captures an inspection object such as a workpiece and determines the quality of the inspection object based on the obtained image is used. Image inspection performed by such an image inspection apparatus is generally performed by registering a master image and setting various image inspection tools on the registered master image.

[0003] When there are a wide variety of combinations of imaging conditions for an image sensor such as lighting and a camera, it has been difficult for the user to empirically find the optimal imaging conditions. In contrast, there is known an image inspection apparatus that determines the optimal imaging conditions by changing various imaging conditions and evaluating the images (for example, Patent Document 1).

[0004] However, since the image suitable for the inspection differs depending on the inspection content that the user wishes to perform, it has been difficult to automatically determine the optimal imaging conditions with an image inspection apparatus. For example, when it is desired to inspect the shape of a workpiece to be inspected, an image in which the edges related to the shape are sharply reflected is generally the optimal image. On the other hand, when it is desired to inspect the color of the workpiece to be inspected, an image in which the color difference is clear is generally the optimal image. Also, when it is desired to perform inspections from various viewpoints, an image in which various feature amounts are reflected in a balanced manner is the optimal image.

[0005] Furthermore, optimizing images for inspection may require combining multiple images. Examples include HDR processing, which expands the dynamic range by combining multiple images taken with varying exposure times, and processing that combines images illuminated from multiple different directions. However, improving image quality through these processes worsens the cycle time due to the acquisition of multiple images. For users, the truly optimal image must be selected considering not only image quality but also the cycle time required to obtain that image. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2017-223458 [Overview of the project] [Problems that the invention aims to solve]

[0007] One of the objectives of the present invention is to provide an image inspection apparatus, a control unit for the image inspection apparatus, an image inspection method, an image inspection program, a computer-readable recording medium, and a recording device that can easily capture images for inspection. Means for solving the problem and effects of the invention.

[0008] According to one embodiment of the present invention, an image inspection apparatus comprises: an illumination unit that irradiates an object to be inspected with illumination light; a camera unit that receives light irradiated from the illumination unit and reflected by the object to be inspected, and generates a work image; a display unit that displays the work image generated by the camera unit; an image generation condition control unit that controls the camera unit to generate multiple work images by changing the image generation conditions when generating the work image to multiple different conditions; an image score calculation unit that calculates an image score, which is an evaluation value of each work image, based on multiple feature quantities that characterize each work image, from the multiple work images generated by the image generation condition control unit under different image generation conditions; and thumbnails corresponding to the multiple work images having the highest image scores. The image inspection apparatus includes an image generation condition setting unit that, while displaying an image on the display unit, accepts the selection of one thumbnail image from among a plurality of thumbnail images, and accepts the setting of image generation conditions corresponding to the selected thumbnail image, wherein the plurality of image generation conditions include a single condition for generating an image captured under a single image generation condition, and a composite condition for combining images captured under multiple different image generation conditions, and the display unit, when displaying a plurality of thumbnail images having the highest image scores, is provided with a reference information display field that displays reference information to identify whether each thumbnail image was generated under the single condition or under the composite condition. With the above configuration, when allowing the user to select a desired image from a plurality of thumbnail images, it is possible to easily select an appropriate image by taking into account the reference information along with the image score, and thereby easily set the image generation conditions for generating the image.

[0009] Furthermore, according to another embodiment of the present invention, an image inspection apparatus comprises: an illumination unit that irradiates an object to be inspected with illumination light; a camera unit that receives light reflected from the object to be inspected and generates a work image; an image score calculation unit that controls the camera unit to change the image generation conditions when generating the work image to a plurality of different conditions, and calculates an image score, which is an evaluation value of each work image, based on a plurality of feature quantities that characterize each work image; a display unit that displays a plurality of thumbnail images corresponding to the work image having the highest image score among the plurality of work images based on the image score calculated by the image score calculation unit; and an image generation condition setting unit that accepts the setting of inspection conditions corresponding to a thumbnail image selected from the plurality of thumbnail images, wherein the image score calculation unit is configured to calculate the image score of the work image for each feature quantity for a plurality of different feature quantities, and to determine the plurality of thumbnail images based on the plurality of image scores.

[0010] Furthermore, according to another embodiment of the present invention, an image inspection apparatus comprises: an illumination unit that irradiates an object to be inspected with illumination light; a camera unit that receives light irradiated from the illumination unit and reflected by the object to be inspected, and generates a work image; a display unit that displays the work image generated by the camera unit; and a processor unit that controls the camera unit to generate multiple work images by changing the image generation conditions when generating the work image to multiple different conditions, calculates an image score which is an evaluation value for each work image based on multiple feature quantities that characterize each work image, displays thumbnail images corresponding to the multiple work images having the highest image scores on the display unit, and accepts the selection of one thumbnail image from the multiple thumbnail images displayed on the display unit, thereby accepting the setting of image generation conditions corresponding to the selected thumbnail image, wherein the multiple image generation conditions include a single condition that generates an image captured under a single image generation condition, and a composite condition that combines images captured under multiple different image generation conditions. The display unit includes a reference information display field that, when displaying multiple thumbnail images having the highest image scores, displays reference information to identify whether each thumbnail image was generated under the single condition or under the combined condition. With this configuration, when allowing the user to select a desired image from multiple thumbnail images, the reference information, along with the image score, can be taken into consideration to make it easier to select an appropriate image, thereby making it easy to set the image generation conditions for generating the image.

[0011] Furthermore, according to another embodiment of the present invention, a control unit for an image inspection apparatus is connected to an illumination unit that irradiates an object to be inspected with illumination light, a camera unit that receives light irradiated from the illumination unit and reflected by the object to be inspected and generates a work image, and a display unit that displays the work image generated by the camera unit, and comprises: an image generation condition control unit that controls the camera unit to generate multiple work images by changing the image generation conditions when generating the work image to multiple different conditions, an image score calculation unit that calculates an image score which is an evaluation value of each work image based on multiple feature quantities that characterize each work image, and a higher-level... The system includes an image generation condition setting unit that, while displaying thumbnail images corresponding to multiple work images having image scores on the display unit, accepts the selection of one inspection image from among the multiple thumbnail images and sets inspection conditions corresponding to the selected inspection image. The multiple image generation conditions include a single condition that generates an image captured under a single image generation condition and a composite condition that combines images captured under multiple different image generation conditions. When the display unit displays multiple thumbnail images having the highest image scores, the system is configured to display reference information in a reference information display field to identify whether each thumbnail image was generated under the single condition or under the composite condition. With this configuration, when allowing the user to select a desired image from among multiple thumbnail images, it is possible to easily select an appropriate image by taking into account the reference information along with the image score, thereby easily setting the image generation conditions for generating the image.

[0012] Furthermore, according to another embodiment of the present invention, an image inspection method is performed by irradiating an object to be inspected with illumination light from an illumination unit, receiving the light reflected from the object to be inspected, and capturing a work image with a camera unit, wherein the image inspection method includes the steps of: generating a plurality of image generation conditions in an image generation condition control unit, which include a single condition for generating an image captured with a single image generation condition and a composite condition for combining images captured with a plurality of different image generation conditions, and controlling the camera unit with each of the plurality of image generation conditions to generate a plurality of work images; calculating an image score, which is an evaluation value of each work image, based on a plurality of feature quantities that characterize each work image, using an image score calculation unit; displaying thumbnail images corresponding to the plurality of work images having the highest image scores on a display unit; and receiving the selection of one inspection image from the plurality of thumbnail images displayed on the display unit, thereby setting inspection conditions corresponding to the selected inspection image. This makes it easy to set image generation conditions that generate an appropriate image by having the user select a desired image from a plurality of thumbnail images.

[0013] Furthermore, according to another embodiment of the present invention, an image inspection program is provided for performing image inspection by irradiating an object to be inspected with illumination light from an illumination unit, receiving the light reflected from the object to be inspected, and capturing a work image with a camera unit, wherein the image inspection program provides a computer with the following functions: an image generation condition control unit generates multiple image generation conditions, including a single condition for generating an image captured with a single image generation condition and a composite condition for combining images captured with multiple different image generation conditions, and controls the camera unit according to the multiple image generation conditions to generate multiple work images; an image score calculation unit calculates an image score, which is an evaluation value for each work image, based on multiple feature quantities that characterize each work image, for the work images generated with the multiple different image generation conditions; a display unit displays thumbnail images corresponding to multiple work images having higher image scores; and a computer can be provided with the computer with the ability to set inspection conditions corresponding to the selected inspection image by accepting the selection of one inspection image from the multiple thumbnail images displayed on the display unit. With the above configuration, it is possible to easily set image generation conditions for generating an appropriate image by having the user select the desired image from among multiple thumbnail images.

[0014] Furthermore, a computer-readable recording medium or storage device according to another embodiment of the present invention stores the above-mentioned program. The recording medium includes magnetic disks such as CD-ROM, CD-R, CD-RW, flexible disk, magnetic tape, MO, DVD-ROM, DVD-RAM, DVD-R, DVD+R, DVD-RW, DVD+RW, Blu-ray, HD DVD (AOD), UHD (all product names), optical disks, magneto-optical disks, semiconductor memory, and other media capable of storing programs. In addition to programs stored and distributed on the above-mentioned recording medium, programs also include those distributed by download via network lines such as the Internet. Furthermore, storage devices include general-purpose or dedicated devices on which the above-mentioned program is implemented in an executable state in the form of software or firmware. Moreover, each process and function included in the program may be executed by program software executable on a computer, or each part of the processing may be implemented in a form that combines predetermined gate arrays (FPGA, ASIC) or other hardware, or partial hardware modules that realize some elements of program software and hardware. [Brief explanation of the drawing]

[0015] [Figure 1] This is a schematic diagram showing the configuration of an image inspection apparatus according to an embodiment of the present invention. [Figure 2] This diagram shows the hardware configuration of the image inspection device. [Figure 3] This is a block diagram showing the functions of the processor section. [Figure 4] This is a schematic diagram showing an example of a user interface screen for an image inspection program. [Figure 5] Figure 4 is a schematic diagram showing an example of switching the thumbnail image display area. [Figure 6] This is a schematic diagram illustrating another example of a user interface screen for an image inspection program. [Figure 7] This flowchart shows the procedure for displaying candidate images in order of image score. [Figure 8] It is a schematic diagram showing an image generation condition setting screen. [Figure 9] It is a schematic diagram showing a region designation screen. [Figure 10] It is a schematic diagram showing a screen of "AI is creating imaging conditions". [Figure 11] It is a schematic diagram showing an image generation condition setting screen. [Figure 12] It is a schematic diagram for creating an image score map from the image scores of a candidate image group. [Figure 13] It is a flowchart showing a procedure for determining rankings from the image scores of a candidate image group. [Figure 14] It is a schematic diagram showing a procedure for generating an ideal image. [Figure 15] It is a schematic diagram showing an example of generating an ideal image of color. [Figure 16] It is a schematic diagram showing the reason for separating colored and non-colored. [Figure 17] It is a schematic diagram showing a calculation example of an image score. [Figure 18] It is a schematic diagram showing a procedure for generating ideal images of color and shape.

Mode for Carrying Out the Invention

[0016] Embodiments of the present invention will be described below with reference to the drawings. However, the embodiments shown below are illustrative examples of an image inspection apparatus, image inspection method, image inspection program, computer-readable recording medium, and recorded device for realizing the technical concept of the present invention, and the present invention does not limit the image inspection apparatus, image inspection method, image inspection program, computer-readable recording medium, and recorded device to the following. Furthermore, this specification does not limit the components shown in the claims to the components of the embodiments. In particular, the dimensions, materials, shapes, relative arrangements, etc. of the components described in the embodiments are not intended to limit the scope of the present invention to those, unless otherwise specifically stated, but are merely illustrative examples. Note that the size and positional relationships of the components shown in each drawing may be exaggerated for clarity of explanation. Furthermore, in the following description, the same name and reference numerals indicate the same or similar components, and detailed explanations are omitted as appropriate. Furthermore, each element constituting the present invention may be configured such that multiple elements are made of the same component, with one component serving multiple elements, or conversely, the function of one component may be shared among multiple components. [Embodiment 1]

[0017] A schematic diagram of an image inspection device according to Embodiment 1 of the present invention is shown in Figure 1. An image inspection device is a device for determining the quality of an object to be inspected, such as various parts or products, based on an image captured of the object. It is also called an image sensor and can be used in production sites such as factories. The object to be inspected may be the entire object or only a part of it. Furthermore, one object to be inspected may contain multiple objects. Moreover, a single image may contain multiple objects to be inspected.

[0018] This section describes an example of an image inspection device that captures images of the external appearance of an object to be inspected and performs a pass / fail judgment according to predetermined inspection conditions. For pass / fail judgment, for example, predetermined pass / fail judgment conditions for determining whether an item is good or defective are set at the time of setup. During operation, images of the object to be inspected are captured, and the pass / fail judgment of the object is determined in light of the pass / fail judgment conditions.

[0019] The image inspection device 100 comprises a control unit 2 (which serves as the main unit), an imaging unit 3, a display unit 4, a personal computer 5, and an operation unit 6. An image inspection program for operating the image inspection device 100 is installed on the personal computer 5. The user interface screen of the image inspection program can be displayed on the monitor of the personal computer 5 or on the display unit 4. The personal computer 5 is not mandatory and can be omitted. In this case, the control unit 2 performs the image inspection. Alternatively, the control unit 2 may be configured to execute the image inspection program.

[0020] Furthermore, a personal computer display can be used instead of the display unit 4. Also, in Figure 1, the control unit 2, imaging unit 3, display unit 4, personal computer 5, and operation unit 6 are shown as separate components as an example of the configuration of the image inspection apparatus 100, but any multiple of these can be combined and integrated. For example, the control unit 2 and imaging unit 3 can be integrated, or the control unit 2 and display unit 4 can be integrated. In addition, the control unit 2 can be divided into multiple units and some parts incorporated into the imaging unit 3 or display unit 4, or the imaging unit 3 can be divided into multiple units and some parts incorporated into other units. Furthermore, the operation unit 6 can be provided separately, or it can utilize the input devices of the personal computer, or the display unit can be integrated into other components such as a touch panel.

[0021] In the example shown in Figure 1, the control unit 2 is connected to the imaging unit 3, the display unit 4, and the personal computer 5 via cables. However, the present invention is not limited to wired connections for each component; wireless connections via media such as Wi-Fi, public communication lines, NFC, radio waves, infrared light, or light may also be used. Furthermore, standardized general-purpose communication standards such as Ethernet, IEEE 802.1x, USB, Bluetooth, and ZigBee (all registered trademarks or product names), as well as dedicated protocols and interfaces, can be used as appropriate.

[0022] The hardware configuration of the image inspection apparatus 100 according to Embodiment 1 of the present invention is shown in the block diagram of Figure 2. The image inspection apparatus 100 shown in this figure comprises a control unit 2, an imaging unit 3, a display unit 4, and a personal computer 5. (Control Unit 2)

[0023] The control unit 2 comprises a main board 13, a connector board 16, a communication board 17, and a power supply board 18. The main board 13 is equipped with a processor unit 20 and memory 133. The memory 133 consists of RAM, ROM, etc.

[0024] The connector board 16 receives power from an external power source via a power connector provided on the power interface 161. The power supply board 18 supplies the supplied power to each board. In this embodiment, the camera unit 14 is supplied with power via the main board 13. The motor driver 181 of the power supply board 18 supplies drive power to the motor 141 of the camera unit 14, thereby enabling autofocus.

[0025] The communication board 17 transmits OK / NG signals (judgment signals) indicating the pass / fail judgment result of the object to be inspected, as well as image data, etc., output from the main board 13, to the display unit 4. The display unit 4, upon receiving the judgment signal, displays the judgment result. In this embodiment, the judgment signal is output via the communication board 17, but it may also be configured to output the judgment signal via, for example, the connector board 16. (Operation unit 6)

[0026] The image inspection device 100 also includes an operation unit 6 that accepts user input. The operation unit 6 can utilize existing input devices such as a keyboard, mouse, or touch panel. In the example shown in Figure 2, the communication board 17 is configured to accept various user operations input from the touch panel 41 of the display unit 4 or the keyboard 51 of the personal computer 5. The touch panel 41 of the display unit 4 is, for example, a known touch-type operation panel equipped with a pressure sensor, which detects user touch operations and outputs them to the communication board 17. The personal computer 5 is equipped with a keyboard 51, as well as a mouse and a touch panel, and is configured to accept various user operations input from these operation devices. Communication may be wired or wireless, and either communication method can be realized by conventionally known communication modules.

[0027] The illumination unit 15 is equipped with multiple LEDs 11 that illuminate the imaging area for imaging the object to be inspected. Lenses and reflectors can be provided for the LEDs 11. The lenses can be replaced as short-range or long-range lens units. In this specification, illumination light mainly refers to the light emitted by the illumination unit 15, but it is also used to include ambient light that exists independently of the illumination unit, such as natural light.

[0028] The imaging unit 3 comprises a camera unit 14 and an illumination unit 15. The camera unit 14 can control autofocus operation by being driven by a motor 141. This camera unit 14 captures an image of the object to be inspected in response to an imaging instruction signal from the main board 13. In this embodiment, a CMOS substrate 142 is provided as the image sensor. The captured color image is converted into an HDR image by the CMOS substrate 142 based on conversion characteristics that widen the dynamic range, and output to the processor unit 20 of the main board 13.

[0029] The main board 13 controls the operation of each connected board. For example, it sends a control signal to the LED driver 151 to control the on / off state of multiple LEDs 11 in the lighting unit 15. The LED driver 151 adjusts the on / off state, light intensity, etc., of the LEDs 11 according to the control signal from the processor unit 20. It also sends a control signal to the motor 141 of the camera unit 14 via the motor driver 181 of the power supply board 18 to control the autofocus operation. Furthermore, it sends an imaging instruction signal to the CMOS board 142. (Storage unit 19)

[0030] The control unit 2 is equipped with a storage unit 19, such as a hard disk drive or semiconductor memory. The storage unit 19 stores program files, configuration files, images, judgment results, etc., which enable various controls and processes to be executed by the hardware. The program files and configuration files can also be stored on a portable storage medium such as a USB memory stick or optical disc, and these stored on the storage medium can be read into the control unit 2. (Processor section 20)

[0031] The processor unit 20 of the main board 13 is a control circuit and control element that processes given signals and data, performs various calculations, and outputs the calculation results. The processor unit 20 is not limited to a general-purpose PC processor such as a CPU, MPU, GPU, or TPU, but can be composed of a gate array such as an LSI, FPGA, or ASIC customized for a specific application, a microcontroller, or a chipset or package such as an SoC. The processor unit 20 realizes several functions, which will be described later. Note that the present invention is not limited to an example in which the processor unit is physically composed of one, but may be composed of multiple CPUs, etc. Multiple CPUs may be physically multiple CPUs, or they may be so-called multi-core MPUs that incorporate multiple CPU cores into one package. In this case, in addition to realizing each function with multiple CPUs or CPU cores, different functions may be assigned to and executed for each CPU or CPU core. Furthermore, the processor unit may be composed of a combination of a CPU and a GPU. In this case, in addition to performing the functions of the display control unit described above, the GPU may be configured to execute some or all of the functions assigned to the processor unit.

[0032] In the example shown in Figure 2, the processor section 20 of the main board 13 is composed of an FPGA and a DSP. The FPGA performs lighting control and imaging control, as well as image processing on the acquired image data. The DSP performs edge detection processing, pattern search processing, etc., on the image data. As a result of the pattern search processing, a judgment result indicating whether the object to be inspected is good or bad is output to the communication board 17. The calculation processing results, etc., are stored in the memory 133. In the example above, the FPGA performs lighting control, imaging control, etc., but the DSP may also perform these functions. Alternatively, instead of the FPGA and DSP combination, a single main control circuit or main control unit may be provided. For example, a single CPU may act as the main control unit, performing functions such as sending control signals to the LED driver 151 to control the on / off state of multiple LEDs 11, sending control signals to the motor 141 of the camera section 14 to control autofocus operation, and sending imaging instruction signals, etc., to the CMOS board 142.

[0033] A block diagram of the processor unit 20 is shown in Figure 3. As shown in this figure, the processor unit 20 implements the functions of the image generation condition control unit 21, the image score calculation unit 22, and the image generation condition setting unit 23.

[0034] The image generation condition control unit 21 controls the camera unit 14 to generate multiple work images by changing the image generation conditions when generating work images to several different conditions. This image generation condition control unit 21 also functions as an imaging condition control unit, controlling the camera unit 14 according to imaging conditions, which are one aspect of the image generation conditions, when the camera unit 14 captures a work image. Note that "work image" refers to an image of a work object. (Image score calculation unit 22)

[0035] The image score calculation unit 22 calculates an image score, which is an evaluation value for each work image, based on multiple feature quantities that characterize each work image, from multiple work images generated by the image generation condition control unit 21 under different image generation conditions, and displays the multiple images with the highest image scores as thumbnail images on the display unit 4. If multiple different feature quantities are set, for example, a first feature quantity and a second feature quantity, the image score calculation unit 22 calculates image scores for the first feature quantity and the second feature quantity for each work image. Then, for the image score of the first feature quantity and the image score of the second feature quantity, the work image with a high image score for only one of the feature quantities is displayed on the display unit 4 as the thumbnail image with the highest image score. In this way, by individually calculating image scores for multiple different feature quantities and making work images with a high score for only one of the feature quantities into thumbnail images, images with special features can be displayed as thumbnail images and selected.

[0036] Alternatively, the image with the highest overall image score may be used as the thumbnail image. In this case, the image score calculation unit 22 calculates the image scores for the first and second features for each work image, and displays the work images with high image scores for both as thumbnail images with the highest image scores on the display unit 4. This allows for the calculation of image scores for multiple different features individually, and by using work images with high scores for both features as thumbnail images, inspections from different inspection perspectives can be performed based on the same image.

[0037] The image score calculation unit 22 calculates an image score for a specific inspection target area within the image of the object to be inspected, based on the features contained within this inspection target area. For example, if the shape of the image is specified as the first feature and the color of the image is specified as the second feature, the image score calculation unit 22 calculates an image score based on the edges of the object to be inspected contained within the inspection target area. one In addition to calculating an image score related to the shape, which is a feature, the color information of the object being inspected is used to determine the number two The system calculates an image score related to the color, which is a feature. By having the user specify a particular area to be inspected before calculating the image score, it is possible to avoid situations where the image score is affected by information outside the inspected area. This enables stable image inspection that is not influenced by color or shape information contained in unnecessary areas. (Image generation condition setting unit 23)

[0038] The image generation condition setting unit 23 accepts the selection of one thumbnail image from among the multiple thumbnail images displayed on the display unit 4, and sets the image generation conditions corresponding to the selected thumbnail image. (User interface screen of the image inspection program)

[0039] Here, Figure 4 shows an example of the user interface screen of the image inspection program displayed on the display unit 4. This figure shows the AI-generated imaging condition list screen 200 of the image inspection program, with an image display area 201 on the left and a thumbnail image display area 202 on the far right. The image display area 201 is the area that displays the image captured by the camera unit 14. The image display area 201 constitutes the inspection target area specification section for specifying the inspection target area. (Thumbnail image display area 202)

[0040] The thumbnail image display area 202 is an area that displays multiple thumbnail images in a list. In the example in Figure 4, four thumbnail images LI1 to LI4 are displayed on one screen, and other thumbnail images can be displayed by switching screens using the left and right arrow buttons 203 and 204. For example, pressing the right arrow button 203 on the screen in Figure 4 switches to the screen in Figure 5, where four thumbnail images LI5 to LI8 are displayed in the thumbnail image display area 202B. Pressing the left arrow button 204 on the screen in Figure 5 returns to the screen in Figure 4. Note that known screen switching methods such as scroll bars or flick gestures can be used instead of arrow buttons as appropriate.

[0041] The thumbnail image display area 202 also functions as a thumbnail image selection unit, allowing the user to select a desired work image. The user can select a desired work image from among multiple thumbnail images displayed in the thumbnail image display area 202 using an input device such as a mouse or touch panel. The selected thumbnail image becomes the selected image for image inspection, and image generation conditions corresponding to this selected thumbnail image are set. For example, if the image inspection is a quality determination to distinguish between good and defective products, it is necessary to register a master image of a good product. By registering the master image, the characteristic quantities of the selected thumbnail image, such as color, edges, image brightness and focus, or image generation conditions such as the amount of illumination light, exposure time, and focal position during imaging, are extracted from the thumbnail image or from data stored in association with the thumbnail image, and registered as settings for image inspection. In addition to the master image of a good product, a master image of a defective product can also be registered to automatically extract and set the conditions that define the boundary between good and defective products.

[0042] To display a list of such thumbnail images, it is necessary to generate multiple thumbnail images. Here, the image generation condition control unit 21 automatically generates multiple different image generation conditions and operates the camera unit 14 and the lighting unit 15 for each image generation condition to acquire work images. The image score calculation unit 22 then calculates an image score, which is an evaluation value of the work image, for the acquired work images, and extracts multiple work images with the highest image scores as thumbnail images, which are then displayed in the thumbnail image display area 202 of the display unit 4. At this time, a predetermined number of work images are extracted as thumbnail images in descending order of image score, and are also displayed in the thumbnail image display area 202 in descending order of image score. (Single conditions and combined conditions)

[0043] Here, the multiple image generation conditions include single conditions and composite conditions. A single condition is a condition for generating a work image captured under a single image generation condition. A composite condition is a condition for combining work images captured under multiple different image generation conditions. Thus, work images consist of a single image captured under a single condition and a composite image captured under a composite condition. A composite image is created by combining multiple optical images of the same object to be inspected, each captured under different image generation conditions. Here, we will describe an example where this image synthesis is performed sequentially. In other words, the composite image is captured in real time, and the multiple optical images captured to generate the composite image are not saved as data but are discarded. However, the present invention is not limited to a configuration in which the composite image is captured in real time; for example, single images captured in advance may be saved, and these may be combined retrospectively to obtain a composite image.

[0044] The image synthesis conditions can include a first synthesis process that combines images illuminated from multiple directions to emphasize the edges of the workpiece shape, and a second synthesis process that combines images taken with multiple different exposure times to expand the dynamic range. The second synthesis process is sometimes called HDR.

[0045] Furthermore, the thumbnail image display area 202 will display both single-condition images with high image scores and composite-condition images with high image scores as thumbnail images. Simply comparing image scores tends to favor composite-condition images. Therefore, single-condition images with high image scores will also be included. For example, if the top-ranked image is the image with the highest image score regardless of whether it is a single-condition or composite-condition image, then the top-ranked image is likely to be an image taken under the composite-condition. Therefore, the second-ranked image will be the image with the highest image score among images taken under the single-condition. (Reference information display column 205)

[0046] Furthermore, when the display unit 4 displays multiple thumbnail images with higher image scores, it may also display reference information for each thumbnail image. In the example in Figure 4, the display unit 4 is provided with a reference information display area 205 for displaying reference information. The reference information display area 205 is displayed alongside each thumbnail image displayed in the thumbnail image display area 202. In the example in Figure 4, a reference information display area 205 is provided in the lower left corner of each thumbnail image. (Reference information)

[0047] Reference information is used to identify whether the thumbnail image displayed in the thumbnail image display area 202 was generated under single conditions or under composite conditions. Examples of reference information include the time required to generate the work image, the image score value, and, if the work image is a composite image, the number of images used for synthesis, the number of shots taken, and the type of synthesis such as "HDR" or "light separation." Alternatively, flags such as "single" or "composite" may also be used. In the example in Figure 4, "×4" is displayed overlaid on the camera icon as reference information, indicating that four images were captured and synthesized. Furthermore, the thumbnail image selected in the thumbnail image display area 202 is highlighted, making it easier for the user to understand which thumbnail image is currently selected.

[0048] Furthermore, the thumbnail image selected in the thumbnail image display area 202 may be enlarged and displayed in the image display area 201. In this case, a reference information display area 205B may also be provided in the image display area 201. In the example in Figure 4, the thumbnail image LI1 displayed in the upper row "1" of the thumbnail image display area 202 is selected, and this thumbnail image LI1 is enlarged and displayed in the image display area 201, along with reference information displayed in the reference information display area 205B in the upper right. Here, as more detailed reference information, the exposure time is also displayed in addition to the number of images. When a user selects a desired image from among multiple thumbnail images by referring to such reference information, an environment is provided that makes it easier to select an appropriate image by considering the reference information along with the image of the thumbnail image.

[0049] Furthermore, the thumbnail image display area 202 may be configured to display the top-ranked thumbnail image among multiple thumbnail images relatively larger than the thumbnail images ranked second and below. An example of this is shown in Figure 6. Here, the thumbnail image LI1, which has the highest image score ("1"), is displayed slightly larger than the thumbnail images LI2 to LI4 ("2" and below). This makes it easy for the user to visually understand which thumbnail image is currently selected. Also, when another thumbnail image is selected from the state where thumbnail image LI1 ("1") is selected, the selected thumbnail image is displayed slightly larger, and thumbnail image LI1 ("1") returns to its normal size. This makes it easy for the user to visually understand which thumbnail image is selected. Thus, the enlarged display of thumbnail images in the thumbnail image display area 202 indicates that it is the currently selected thumbnail image. In addition, in the initial stage when thumbnail images are extracted, the default setting is to select thumbnail image LI1 ("1"), which has the highest image score. This means that, in the initial state, the top thumbnail image will inevitably be highlighted.

[0050] Furthermore, the display unit 4 may update the image displayed in the image display area 201 in real time as it sequentially captures work images by varying the image generation conditions. When the image score calculation unit 22 determines multiple thumbnail images, the top thumbnail image is displayed relatively larger in the initial state. Additionally, the image generation condition control unit 21 may be configured to set the brightness of the work image displayed on the display unit 4 and the image generation conditions including the focus of the camera unit 14.

[0051] In conventional image sensors, the combinations of image generation conditions have been extremely diverse, and users have had to empirically search through hundreds of combinations to find the best conditions depending on the object being inspected (workpiece). On the other hand, even with image sensors that have built-in lighting, progress has been made to produce better images by adding additional lighting or changing the lighting method, further increasing the complexity of the combinations of shooting conditions. Therefore, the image inspection apparatus 100 according to this embodiment is configured to present multiple image generation conditions that are considered appropriate and allow the user to select one of them. Here, multiple workpiece images are captured, and an ideal image is generated from the multiple workpiece images obtained. Furthermore, the image generation conditions are determined by scoring the candidate images. In scoring, multiple different evaluation indicators are set, and multiple ideal images that take the maximum of each evaluation index are created. Candidate images refer to workpiece images with the highest image scores among the workpiece images. Thumbnail images refer to thumbnail images corresponding to the candidate images.

[0052] The image score calculation unit 22 also calculates an image score based on the degree of agreement between the ideal image and each work image. Here, an ideal image is created, and then the image score is calculated based on the distance from that ideal image. Alternatively, multiple ideal images may be generated for each feature, and the image score for each feature may be calculated. (Ideal image)

[0053] As mentioned above, features include the shape (edges) and color of the image. Therefore, the ideal image can be an ideal image for shape or an ideal image for color. When generating an ideal image for color, a priority can be set for the conditions. For example, chromatic, achromatic, and saturated colors can be used in order of priority. It may also be possible to allow the user to specify which features to use from among multiple features. In this case, a feature specification unit is provided that accepts the specification of which features to use for inspection from among multiple features. As an example of a feature specification unit, for example, the image inspection program displays a feature specification screen on the display unit 4, and the user operates an input device such as a mouse to specify which features to use for inspection from among multiple features. Accordingly, the image score calculation unit 22 calculates an image score based on the specified features, and the thumbnail images with the highest calculated image scores are displayed on the display unit 4. In this way, by selecting and specifying specific features, it is possible to display candidate images with the desired features.

[0054] Here, we will explain the procedure for displaying candidate images in order of image score, based on the flowchart in Figure 7 and Figures 8 to 11. In this procedure, a good product to be inspected is registered as a master image using an image inspection program, and inspection conditions for image inspection are set to determine its contours, area, edges, and other features. Once the master image is registered and the settings are complete, it is assumed that in operation, the difference between the inspected product and the master image is used to determine whether the product is good or bad.

[0055] First, in step S701 in Figure 7, autofocus (AF) and automatic exposure (AE) are temporarily performed so that the object to be inspected is visible. For example, in the image inspection program, a dialog box "Brightness & Focus Automatic Adjustment in Progress" is displayed. Next, in step S702, the user specifies the area they want to focus on. The area can be specified in the inspection target area specification section. For example, pressing the "AI Imaging" button 223 in the operation area 222 located to the right of the image generation condition setting screen 220 shown in Figure 8 brings up the area specification screen 230 in Figure 9, which is one aspect of the inspection target area specification section. In the image display area 201 located to the left of the area specification screen 230, a live image of the object to be inspected captured by the camera unit 14 is displayed. The live image is displayed as a continuously updated video. In this state, the user is instructed to specify the inspection target area in the image display area 201 using an input device such as a mouse. Here, the target area is specified so as to surround the entire object to be inspected displayed in the image display area 201, or the part of the object to be inspected that the user wants to focus on. Furthermore, in step S703, autofocus is performed again on the area specified in the region to adjust the focus.

[0056] Next, in step S704, candidate images are captured while varying the brightness by combining image generation conditions, in this case imaging conditions, such as turning the inner lighting ON / OFF, turning the outer lighting ON / OFF, and taking multiple images. As shown in Figure 10, a dialog box "AI is creating imaging conditions" is displayed, and the captured work images are updated in real time in the image display area 201. Furthermore, in step S705, all candidate images are selected from one condition, such as a combination of lighting ON / OFF, and these are combined to create the "ideal color image." Here, the selected candidate images are called the candidate image group. Next, in step S706, the degree of agreement between the candidate image group and the "ideal color image" is set as the "color image score." Also in step S707, the "ideal shape image" is created from the candidate image group and the "ideal color image." Furthermore, in step S708, the degree of agreement between the candidate image group and the "ideal shape image" is set as the "shape image score." Then, in step S709, the "overall image score" is calculated from the "color image score" and the "shape image score". In step S710, the ranking is calculated from the candidate image group and the "overall image score".

[0057] By calculating the image score in this way, the order in which the candidate images are displayed in the thumbnail image display area 202 is determined based on the calculated ranking of the candidate images. Then, in step S711, the candidate images are displayed in the thumbnail image display area 202 as shown in Figure 4. In the image display area 201, a work image with adjusted brightness and focus to match the image generation conditions of the top-ranked candidate image is displayed in the image display area 201. At this time, the thumbnail image ranked first may be displayed relatively larger, as shown in Figure 6.

[0058] Furthermore, when displaying thumbnail images, reference information may also be displayed (step S712). In the example in Figure 4, icons are displayed in the reference information display area 202 for images synthesized from multiple images. In this way, the user is allowed to select the desired thumbnail image as the selected image. This switches the user from the AI ​​generation imaging condition list screen 200 (as shown in Figures 4 and 5) to the image generation condition setting screen 240 (as shown in Figure 11). (Ranking Determination Procedure)

[0059] Next, the procedure for determining the ranking from the image scores of the candidate image groups will be explained based on the schematic diagram in Figure 12 and the flowchart in Figure 13. In Figure 12, let's assume there are four candidate image groups, A to D. First, in step S1301, an "ideal color image" is created from the candidate image groups. For example, the "ideal color image" E is created from the four candidate image groups A to D. Here, as shown in Figure 14, a group of color-coded images is created from the candidate image groups first. Then, the ideal color image is created from the candidate image groups and the color-coded image groups. Furthermore, in step S1302, the degree of agreement between the candidate image groups and the "ideal color image" is calculated and used as the "color image score". Here, the color image score is calculated from the candidate image groups, the color-coded image groups, and the ideal color image.

[0060] On the other hand, in step S1303, an "ideal shape image" is generated from the candidate image group and the "ideal color image". For example, the "ideal shape image" F is created from the four candidate image groups A to D. Then, in step S1304, the degree of agreement between the candidate image group and the "ideal shape image" is calculated and used as the "shape image score". Note that these calculations for the "color image score" and the "shape image score" may be performed in parallel, or the "shape image score" may be calculated first and then the "color image score".

[0061] Then, in step S1305, the "overall image score" is calculated from the "color image score" and the "shape image score". For example, Figure 12 shows an image score map SM plotted with the "color image score" on the horizontal axis and the "shape (edge) image score" on the vertical axis. In Figure 12, G represents the image scores of candidate images A, B, C, and D, where 0 is A, 1 is B, 2 is C, and 3 is D. Then, in step S1306, the ranking is calculated from the group of candidate images and the "overall image score". For example, in the image score map G, the best candidate image for each item (color, shape) is selected and its ranking is determined. (Ideal image in terms of color)

[0062] The way an object being inspected appears changes depending on the image generation conditions, specifically the shooting conditions. For example, when comparing multiple candidate images taken under different shooting conditions, some will be good and others will be bad in terms of color. From this group of candidate images with different shooting conditions, the good parts are extracted and synthesized to create the best possible image, i.e., an ideal image in terms of color. In this process, "color sorting" and "color priority" are used as methods to determine quality from a color perspective. (Color coding, color priority)

[0063] The priority order for colors is chromatic, achromatic, and saturated, in descending order. Each candidate image is color-coded, and these are considered correct according to the priority order of colors. The most likely correct colors are then collected from the group of candidate images to synthesize an ideal color image. Based on this synthesized ideal color image, each candidate image is then evaluated. Specifically, the image score for each candidate image is defined such that a higher value indicates a closer match to the ideal image, and this is calculated by the image score calculation unit 22. (Method for generating an ideal color image)

[0064] Here, we will explain the procedure for generating an ideal color image, referring to Figure 15. Here, we consider an example of generating an ideal color image from two candidate images, A and B. First, each candidate image is color-coded. Here, each candidate image is color-coded into chromatic, achromatic, and saturated. Specifically, the chromatic color of A becomes C, the achromatic color becomes E, and the saturated color becomes G. Similarly, the chromatic color of B becomes D, the achromatic color becomes F, and the saturated color becomes H. Then, from each of the obtained images, we collect the pixels with the highest priority to generate an ideal color image. Here, from candidate images A and B, image I is generated as the ideal color image. (Color-coded judgment)

[0065] Here, we will explain the reason for distinguishing between chromatic and non-chromatic colors based on Figure 16. Even when images of the same object being inspected are captured, the reason why they change depending on the shooting method is mainly due to two factors: light intensity and component. First, changes in the light intensity of the illumination light cause images to become underexposed or overexposed. For example, in Figure 16, image B is underexposed as image A, and overexposed as image C. On the other hand, the component refers to the component of the reflected light when the illumination light is reflected by the object being inspected, and includes specular reflection and diffuse reflection. For example, in image E, the specularly reflected image is D, and the diffusely reflected image is F. Thus, captured images can change from chromatic to other colors depending on the light intensity and component of the illumination light. On the other hand, non-chromatic objects basically do not change into chromatic colors. Therefore, when the same object being inspected appears as both chromatic and other colors depending on the shooting method, it can be said that the chromatic image is correct.

[0066] Thus, by prioritizing chromatic colors, it becomes possible to make highly reliable judgments. Also, since chromatic colors become achromatic when photographed with a monochrome camera, achromatic colors are the next most reliable after chromatic colors. On the other hand, saturated colors are relatively less reliable than chromatic or achromatic colors because all color information is lost. However, they are not worthless because they may contain information other than color, such as the presence or absence of gloss or the continuity of three-dimensional shapes. As a result, the priority of colors, in descending order, is chromatic, achromatic, and saturated. In the examples in Figures 17E to 17T, the images shown at the top have higher priority. (Color image)

[0067] Here, we will explain a color classification method that categorizes each pixel in an image as either chromatic, achromatic, or saturated. First, the conditions for classifying an image as chromatic are that it has high saturation, the surrounding hue is constant, and it is unsaturated. (Achromatic image)

[0068] Next, the conditions for determining an image as achromatic are low saturation, constant ambient brightness, and unsaturation. Here, while saturation is excluded from the definition of achromatic, crushed blacks are not excluded. This is because, as can be seen when considering a glossy black surface, the information obtained from saturation is only the presence or absence of gloss, which is unrelated to color, whereas crushed blacks can sometimes indicate the correct color information of black. (Saturated image)

[0069] Furthermore, the conditions for determining saturation are those that satisfy all the conditions for chromatic and achromatic colors mentioned above, except for the condition for non-saturation. (Calculation of image score for each candidate image)

[0070] Next, the procedure for calculating the image score for each candidate image will be explained based on Figure 17. Figure 17 shows the image scores for eight combinations of the ideal image and candidate images. The image score is defined so that the closer the image is to the ideal image, the higher the score. Here, the image score is expressed as a percentage from 0 to 100%. For pixels where the color coding of the ideal image and the candidate image matches, as shown in Figure 17, the score is set to 100% if the colors match and 0% if they do not match, and the average of all pixels is used as the image score.

[0071] For example, as shown in Figure 17, if the ideal image is chromatic and the candidate image is a perfect match with the same chromatic colors, the image score will be 100%. Also, as shown in Figure 2, if the ideal image is chromatic and the candidate image differs in saturation or brightness, the image score will be determined according to the difference, and in this case it will be 75%. Here, the image score is calculated as {1 - (absolute value of the difference in the maximum RGB values)} / 255. Furthermore, as shown in Figure 3, if the ideal image is chromatic and the candidate image is chromatic but differs in hue, the image score will be 0%. Moreover, as shown in Figure 4, if the ideal image is chromatic and the candidate image is achromatic, the color coding is different, so the image score will also be 0%.

[0072] On the other hand, as shown in Figure 17, if the ideal image is achromatic and the candidate image is also achromatic, the image score will be 100% because it is a perfect match between two achromatic colors. Although chromatic colors have a higher priority than achromatic colors when generating the ideal image, the weight of chromatic and achromatic colors is the same when calculating the image score. Also, if the image score for two achromatic images does not reach 100%, the achromatic object to be inspected will be determined solely by its shape. Furthermore, as shown in Figure 17, if the ideal image is achromatic and the candidate image is achromatic but has a different brightness, the image score will be determined according to the difference in brightness, and in this case it will be 75%.

[0073] On the other hand, as shown in Figure 17 at point 7, if the ideal image is a saturated image and the candidate image is also a saturated image in terms of saturation, then the saturated images match, and the image score becomes 100%. As will be discussed later, there may be cases where achromatic is better and cases where saturated is better. The image scores for colored and saturated images are given the same weight. Also, as shown in point 8, if the ideal image is a saturated image and the candidate image is a saturated image with some differences, the image score is determined according to the differences, and in this case it becomes 50%. (Ideal image of the shape)

[0074] Next, we will explain how to generate an ideal image of the shape. The shape contained in each work image, such as the edge information obtained from the shape, may be excessive or insufficient. When calculating the image score, it is possible that there is insufficient edge information, in which case it is necessary to supplement the edge information. Conversely, if there is extraneous edge information, it may be necessary to select or discard it. For example, if it is too bright and saturated, fine details will disappear and only the large shapes will remain. Also, if the lighting reflects off a glossy surface, shapes unrelated to the object will appear in the work image. In addition, the shadow of a three-dimensional object may be cast by the lighting, causing its shape to be doubled.

[0075] In such cases, even when it is not possible to determine whether there is excess or deficiency from an image generated under a single image generation condition, valuable shape information can be extracted by comparing work images generated under multiple image generation conditions. For example, shapes that are common to work images generated under multiple image generation conditions are more reliable than shapes that are only present in a work image generated under a single image generation condition. Applying this to the example of shadows on three-dimensional objects mentioned above, the shape of the shadow when illuminated from one direction will not appear when illuminated from another direction. (Procedure for generating ideal images in terms of color and shape)

[0076] Here, the procedure for generating ideal color and ideal shape images will be explained based on Figure 18. First, an ideal color image is generated from the original work image. Here, let's assume that C is obtained as the ideal color image from the original work images A and B.

[0077] On the other hand, edge information is extracted from the original work images. Existing algorithms can be used as appropriate for the process of extracting the image contour, i.e., edge information, from the work images. For example, the Canny edge detector, differential filters, Prewitt filters, and Sobel filters can be used. Here, let's assume that D to E are obtained as edge images from each of the work images A to B. Then, only the edges of pixels whose color coding matches the ideal color image are made valid, and the maximum value of all edge images is taken to obtain the ideal shape image. Here, let's assume that for each candidate image A to B, the regions whose color coding matches the ideal color image C are as shown in F to G. H to I are obtained by making only the regions F to G valid from D to E mentioned above. Furthermore, J, which is the maximum value of H to I, is the ideal shape image. As shown in J, in the ideal shape image, fine textures that appear differently in each candidate image are suppressed, and it can be seen that the shape common to all candidate images, in this example the engraved part, is emphasized. (Removal of texture information)

[0078] Next, we will explain the procedure for removing texture information. If we were to create an ideal shape image by combining images while retaining texture information without selecting edges, the resulting image would be an ideal shape image. As a result of texture information remaining in this ideal shape image, the image score of the image containing texture information would be high, raising concerns that images containing unnecessary texture information might be selected as the ideal image, leading to lower accuracy. For example, work images like those shown in Figure 18A and B have high edge intensity due to texture information. Furthermore, since spatial frequencies differ depending on the object being inspected, it is difficult to remove texture information using filters based on edge intensity and spatial frequency for each work image. Therefore, in order to suppress such texture information, we perform processing using information from all images rather than each individual work image. We also use evaluation metrics other than edge intensity and spatial frequency for judgment. Here, we use an ideal color image to sort out the edge information. (Edge detection method)

[0079] When sorting edge information using an ideal color image, edge extraction is performed. Here, the Sobel filter is used as the edge extraction method, and the sum of the absolute values ​​in the X and Y directions is used. [Industrial applicability]

[0080] The image inspection apparatus, control unit for the image inspection apparatus, image inspection method, image inspection program, computer-readable recording medium, and recorded device of the present invention can be suitably used for applications such as determining the quality of an object to be inspected based on an image of the object being inspected. [Explanation of symbols]

[0081] 100…Image inspection device 2…Control Unit 3…Imaging Unit 4...Display section 5…Personal computer 6...Operation unit 11…LED 13…Main board 14…Camera Club 15…Lighting Department 16…Connector board 17…Communication board 18…Power supply board 19...Storage section 20…Processor section 21...Image generation condition control unit 22…Image score calculation unit 23...Image generation condition setting section 41…Touch panel 51...Keyboard 100…Image inspection device 133...Memory 141…motor 142... Circuit board 151... Driver 161…Power Interface 181...Motor Driver 200... AI-generated imaging conditions list screen 201…Image display area 202, 202B... Thumbnail image display area 203... Right arrow button 204... Left arrow button 205, 205B…Reference information display column 220...Image generation condition setting screen 222…Operation area 223... "AI Imaging" button 224... Trigger condition button 230…Area specification screen 240...Image generation condition setting screen LI1~LI8...Thumbnail images SM...Image Score Map

Claims

1. A lighting unit that illuminates the object to be inspected, A camera unit that receives light emitted from the aforementioned illumination unit and reflected by the object to be inspected, and generates a workpiece image, A display unit that displays the work image generated by the camera unit, An image generation condition control unit controls the camera unit to generate multiple work images by changing the image generation conditions when generating the aforementioned work image to multiple different conditions, A target area designation unit that accepts the designation of the target area, An image score calculation unit calculates an image score, which is an evaluation value for each work image, based on a plurality of feature quantities that characterize the target region of each work image, from a plurality of work images generated by the image generation condition control unit under different image generation conditions. An image generation condition setting unit displays thumbnail images corresponding to multiple work images having higher image scores on the display unit, accepts the selection of one thumbnail image from among the multiple thumbnail images, and accepts the setting of image generation conditions corresponding to the selected thumbnail image. Equipped with, The target area designation unit displays a live image captured by the camera unit of the object to be inspected, and accepts the designation of the target area based on operations on the live image. The image inspection apparatus is configured such that the image generation condition control unit, after receiving the designation of the target area, controls the camera unit to generate multiple work images relating to the same object to be inspected.

2. An image inspection apparatus according to claim 1, The image generation condition setting unit is an image inspection device that controls autofocus based on the target area.

3. An image inspection apparatus according to claim 1, The image score calculation unit calculates image scores for each work image, for each of the multiple different feature quantities, namely the first feature quantity and the second feature quantity. An image inspection apparatus configured to display work images with high image scores for only one of the first and second feature quantities as thumbnail images with higher image scores on the display unit.

4. An image inspection apparatus according to claim 1, The image score calculation unit calculates image scores for each work image, for each of the multiple different feature quantities, namely the first feature quantity and the second feature quantity. An image inspection apparatus configured to display work images with high image scores for both the first feature and the second feature as thumbnail images with higher image scores on the display unit.

5. An image inspection apparatus according to any one of claims 1 to 4, The aforementioned multiple feature quantities include a shape set as the first feature quantity and a color set as the second feature quantity. The aforementioned image score calculation unit, Based on the edges of the object to be inspected, an image score relating to the shape, which is the first feature quantity, is calculated, An image inspection device configured to calculate an image score related to color, which is the second feature quantity, based on the color information of the object to be inspected.

6. An image inspection apparatus according to any one of claims 1 to 5, further, It is equipped with a feature selection unit that accepts the selection of features to be used for inspection from among multiple features. The image score calculation unit calculates an image score based on the specified feature quantities. An image inspection apparatus configured such that thumbnail images having the higher image scores calculated by the image score calculation unit are displayed on the display unit.

7. An image inspection apparatus according to any one of claims 1 to 6, The image inspection device is configured such that the image score calculation unit calculates an image score based on the degree of agreement between the ideal image generated for inspection based on the feature quantities and each workpiece image.

8. An image inspection apparatus according to any one of claims 1 to 7, The aforementioned multiple different conditions include: A single condition for generating an image captured under a single image generation condition, A synthesis condition that combines images captured under multiple different image generation conditions, and It includes, The aforementioned display is an image inspection device that displays both a work image under a single condition having a high image score and a thumbnail image corresponding to a work image under a composite condition having a high image score.

9. An image inspection apparatus according to any one of claims 1 to 8, The aforementioned multiple different conditions include: A single condition for generating an image captured under a single image generation condition, A synthesis condition that combines images captured under multiple different image generation conditions, and It includes, An image inspection apparatus comprising a reference information display field that displays, as reference information, the number of times the camera unit took images to generate the thumbnail image generated by the aforementioned synthesis conditions.

10. An image inspection apparatus according to any one of claims 1 to 9, The aforementioned multiple different conditions include: A single condition for generating an image captured under a single image generation condition, A synthesis condition that combines images captured under multiple different image generation conditions, and It includes, The aforementioned synthesis conditions are, A first synthesis process combines images illuminated from multiple directions to emphasize the edges of the workpiece's shape, The second compositing process expands the dynamic range by combining images taken with multiple different exposure times. An image inspection device comprising the following:

11. An image inspection apparatus according to any one of claims 1 to 10, The display unit is configured to display the topmost thumbnail image among the plurality of thumbnail images in a relatively larger size than the second and subsequent thumbnail images.

12. A lighting unit that illuminates the object to be inspected, A camera unit that receives light reflected from the object being inspected and generates a workpiece image, An image score calculation unit that displays a live image of the object to be inspected captured by the camera unit, accepts the designation of a target area based on operations on the live image, and after accepting the designation of the target area, controls the camera unit to change the image generation conditions for generating the work image to several different conditions to generate multiple work images relating to the same object to be inspected, and calculates an image score, which is an evaluation value of each work image, based on several feature quantities that characterize the target area of ​​each work image, A display unit that displays multiple thumbnail images corresponding to work images with higher image scores among multiple work images, based on the image scores calculated by the image score calculation unit, An image generation condition setting unit that accepts the setting of inspection conditions corresponding to a thumbnail image selected from the aforementioned plurality of thumbnail images, An image inspection device comprising, The aforementioned image score calculation unit, For multiple different features, the image score of the work image is calculated for each feature. Based on the above-mentioned multiple image scores, the above-mentioned multiple thumbnail images are determined. An image inspection device configured in such a way.

13. A lighting unit that illuminates the object to be inspected, A camera unit that receives light emitted from the aforementioned illumination unit and reflected by the object to be inspected, and generates a workpiece image, A display unit that displays the work image generated by the camera unit, The camera unit displays a live image of the object to be inspected, and accepts the designation of a target area based on operations on the live image. After receiving the designation of the target area, the camera unit is controlled to generate multiple work images by changing the image generation conditions for generating the work image to multiple different conditions, thereby generating multiple work images relating to the same object to be inspected. The image score, which is an evaluation value for each work image, is calculated based on a plurality of feature quantities that characterize the target region of each work image, and thumbnail images corresponding to the work images with the highest image scores are displayed on the display unit. A processor unit that accepts the selection of one thumbnail image from among the plurality of thumbnail images displayed on the display unit, and accepts the setting of image generation conditions corresponding to the selected thumbnail image, An image inspection device equipped with the following features.

14. An image inspection apparatus according to any one of claims 1 to 13, The aforementioned target area is the area to be image-inspected, The image inspection device is an image sensor that performs a pass / fail determination based on an image region including the target region.

15. A lighting unit that illuminates the object to be inspected, A camera unit that receives light emitted from the aforementioned illumination unit and reflected by the object to be inspected, and generates a workpiece image, A display unit that displays the work image generated by the camera unit, A control unit connected to, An image generation condition control unit controls the camera unit to generate multiple work images by changing the image generation conditions when generating the aforementioned work image to multiple different conditions, A target area designation unit that accepts the designation of the target area, An image score calculation unit calculates an image score, which is an evaluation value for each work image, based on a plurality of feature quantities that characterize the target region of each work image, from a plurality of work images generated by the image generation condition control unit under different image generation conditions. An image generation condition setting unit displays thumbnail images corresponding to multiple work images having higher image scores on the display unit, accepts the selection of one inspection image from among the multiple thumbnail images, and sets inspection conditions corresponding to the selected inspection image. Equipped with, The target area designation unit displays a live image captured by the camera unit of the object to be inspected, and accepts the designation of the target area based on operations on the live image. The image generation condition control unit is configured to receive the designation of the target area and then control the camera unit to generate multiple work images relating to the same object to be inspected, thereby forming a control unit for an image inspection apparatus.

16. An image inspection method in which an illumination unit irradiates an object to be inspected with illumination light, the light reflected from the object to be inspected is received, and a camera unit captures an image of the workpiece, thereby performing image inspection. The image generation conditions when generating the aforementioned work image are as follows: A single condition for generating an image captured under a single image generation condition, Image synthesis conditions for combining images captured under multiple different image generation conditions. The process involves generating multiple image generation conditions, including the above, in the image generation condition control unit, and controlling the camera unit according to the multiple image generation conditions to generate multiple work images, The process of receiving the designation of the target area, The process involves calculating an image score, which is an evaluation value for each work image, based on a plurality of feature quantities that characterize the target region of each work image, using an image score calculation unit, for work images generated under the plurality of different image generation conditions. A step of displaying thumbnail images corresponding to multiple work images having high image scores on the display unit, The process of receiving the selection of one inspection image from among the multiple thumbnail images displayed on the display unit, and setting inspection conditions corresponding to the selected inspection image, Includes, In the process of receiving the designation of the target area, the camera unit displays a live image of the object to be inspected, and the designation of the target area is received based on operations on the live image. An image inspection method in which, in the step of generating the plurality of work images, the image generation condition control unit receives the designation of the target area and then controls the camera unit to generate the plurality of work images relating to the same object to be inspected.

17. An image inspection program for performing image inspection by irradiating an object to be inspected with illumination light from an illumination unit, receiving the light reflected from the object to be inspected, and capturing a workpiece image with a camera unit, The image generation conditions when generating the aforementioned work image are as follows: A single condition for generating an image captured under a single image generation condition, Image synthesis conditions for combining images captured under multiple different image generation conditions. The image generation condition control unit generates multiple image generation conditions, and the camera unit is controlled according to each of the multiple image generation conditions to generate multiple work images. A function that accepts the specification of the target area, The image score calculation unit calculates an image score, which is an evaluation value for each work image, based on a plurality of feature quantities that characterize the target region of each work image, for work images generated under the plurality of different image generation conditions. A function to display thumbnail images corresponding to multiple work images with high image scores in the display unit, The function includes accepting the selection of one inspection image from among the multiple thumbnail images displayed on the display unit, and setting the inspection conditions corresponding to the selected inspection image. This is to enable computers to do this. In the function for receiving the designation of the target area, the camera unit displays a live image of the object to be inspected, and the designation of the target area is received based on operations on the live image. An image inspection program comprising the function for generating multiple work images, wherein the image generation condition control unit is configured to receive the designation of the target area and then control the camera unit to generate multiple work images relating to the same object to be inspected.

Images