Automatic target image capture and calibration system applied for inspection - Patents.com

The automatic target image capture and calibration system addresses the inefficiencies of manual camera adjustments by using robotic arms and integrated software for precise, automated inspection of industrial products.

JP7578965B2Active Publication Date: 2024-11-07KAPITO INC +3
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2023133430
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2022-08-24
Filing Date
2023-08-18
Publication Date
2024-11-07
Estimated Expiration
2043-08-18

AI Technical Summary

Technical Problem

Conventional automatic optical inspection systems for industrial products require manual adjustment of camera parameters and light sources for different product models, which is time-consuming and prone to errors.

Method used

An automatic target image capture and calibration system using robotic arms and cameras with integrated control software to automatically adjust camera positions, capture multiple images, and perform defect inspection by recognizing laser labels for precise positioning and image processing.

Benefits of technology

Ensures accurate defect inspection without manual parameter tuning, improving efficiency and reducing errors by automatically capturing and calibrating images of industrial products.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007578965000001
    Figure 0007578965000001
  • Figure 0007578965000002
    Figure 0007578965000002
  • Figure 0007578965000003
    Figure 0007578965000003
Patent Text Reader

Abstract

To provide an automatic target image capture and calibration system applied for inspection.SOLUTION: A control and processing device 15 controls a first robot arm 11 so as to reverse an article 2 according to reversing order, and controls a first video camera 12 so as to sequentially acquire a plurality of two dimensional images in a process of reversing the article. After finding at least one two dimensional image including an identification feature 21 from the plurality of two dimensional images, the control and processing device determines a relative position and a relative three dimensional coordinate of the object according to the plurality of two dimensional images, and then, a second robot arm 13, by holding the second video camera 14 and moving the second video camera with a plurality of times, control N article images so as to correspondingly acquire the N article images as the second video camera is made to face a large number of surfaces of articles. Finally, the control and processing device performs a defect inspection on the article images to determine whether each article image contains at least one defect.SELECTED DRAWING: Figure 2A
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to the technical field of machine vision, and more particularly to an automatic target image capture and calibration system applied for inspection. [Background technology]

[0002] Generally, semi-finished products or finished products produced by industrial technology need to undergo quality inspection to determine whether they meet industrial standards, and the quality inspection includes cosmetic inspection and function test. Generally, an industrial product has six surfaces, namely, upper surface, lower surface, front surface, rear surface, left surface and right surface. When performing the cosmetic inspection, the quality inspector will carry out the most detailed cosmetic inspection on some surfaces of the six surfaces of the industrial product according to the relevant inspection regulations. However, according to practical experience, when the quality inspector performs the cosmetic inspection of the industrial product, there are often cases where the inspection is overlooked or the inspection is erroneous. In view of the above, automated optical inspection (AOI) systems using machine vision technology have been developed and introduced into automated production lines, and are used to perform visual inspection on industrial semi-finished or finished products as a substitute for human labor.

[0003] For example, Fig. 1 is a perspective view showing a screw defect inspection system 1a according to the related art. As shown in Fig. 1, the screw defect inspection system 1a includes a transport platform 11a, a plurality of photographing devices 12a, and an image processing device 13a, among which each of the photographing devices 12a is mounted on an adjustable support mechanism 14a. It should be understood that the adjustable support mechanism 14a can be manually adjusted to allow each of the photographing devices 12a to have an installation height and an installation angle, and be disposed at an imaging distance from the transport platform 11a. With this arrangement, the thread defect inspection system 1a can perform visual inspection on some of the six surfaces of the type A thread.

[0004] If an AOI engineer is skilled in developing and manufacturing defect inspection systems, he or she will know that when the thread defect inspection system 1a is used to perform visual inspection on the model A thread, the engineer must tune the installation height, installation angle, and imaging distance of each individual imaging device 12a. In addition, if necessary, the engineer must also design a light source. In other words, the manufacturer requests the AOI engineer to change the equipment parameters of the thread defect inspection system 1a so that it can be used to perform visual inspection on the model B thread, and in this case, the engineer must repeatedly tune the installation height, installation angle, imaging distance, and light source until completion.

[0005] Practical experience has shown that tuning the installation height, installation angle, imaging distance and light source is not easy and may require considerable effort and time to complete. In view of the above, the development of a new automatic optical inspection system that can automatically change the installation height, installation angle and imaging distance of the photographing equipment according to the specified industrial product, adjust the light source intensity, and capture the target image of the photographing equipment by inverting the industrial product should be considered. Summary of the Invention [Problem to be solved by the invention]

[0006] As can be seen from the above description, conventional automatic optical inspection systems used to perform visual inspection on industrial products still have deficiencies and improvements are desired. In view of the above, the inventor of the present invention has made extensive research and development, and finally completed an automatic target image capture and calibration system applicable for inspection according to the present invention.

[0007] The primary objective of the present invention is to provide an automatic target image acquisition and calibration system adapted for inspection. In accordance with the design of the present invention, electronic equipment controls the first robot arm to flip the article according to a flip order, and controls the first video camera to capture a plurality of 2D images in sequence during the process of flipping the article. Continuously, finds at least one 2D image including an identification feature (e.g., a laser label or a recognizable feature of the article) from the plurality of 2D images, and then electronic equipment confirms the relative position and relative three-dimensional coordinates of the object according to the plurality of two-dimensional images, and then the second robot arm holds the second video camera and controls the second video camera to move in a plurality of steps to face the plurality of surfaces of the object in a plurality of steps, thereby correspondingly acquiring N to N object images among the plurality of surfaces of the object. electronic equipment performs a defect inspection on the product images to determine whether each product image contains at least one defect.

[0008] In short, the present invention uses the method of identifying a laser label to establish the relative position and relative three-dimensional coordinates of an article, and then only by controlling the carrying and moving of a video camera by a robot arm, the video camera can be accurately faced with the inspected surface of the article, realizing automatic target image capture of the article, and ensuring the accuracy of defect inspection. Furthermore, when the defect inspection system of the present invention is applied to inspect other articles for defects, the technician does not need to manually adjust the installation height and installation angle of the video camera, and does not need to tune the light source direction and intensity. [Means for solving the problem]

[0009] In order to achieve the above object, the present invention provides an embodiment of an automatic target image capture and calibration system, which includes a first robot arm, a first video camera, and at least one second robot arm; electronic equipment a first robot arm is disposed adjacent to the first conveyor belt and the second conveyor belt at the same time, the first conveyor belt conveys a plurality of articles, and at least one surface of the articles has an identification feature; a first video camera is disposed adjacent to the first robot arm; at least one second robot arm is disposed adjacent to the second conveyor belt, and each second robot arm holds a second video camera; electronic equipmentthe at least one second robot arm and the at least one second video camera, and includes a microprocessor and a memory, wherein an application program compiled using a programming language is stored in the memory, and the microprocessor executes the application program through access to the memory to start the following functions: a function of controlling the first robot arm to pick up one of the articles from the first conveyor belt, hold the article, move it within a first image capture range of the first video camera, and then turn over the article according to a turn-over order; a function of controlling the first video camera to sequentially capture a plurality of two-dimensional images of the article when turning over the article according to the turn-over order; and a function of performing a label recognition process on the plurality of two-dimensional images to find at least one of the two-dimensional images including an identification feature image. the first robot arm dynamically performs the functions of: determining a relative position and relative three-dimensional coordinates of the article according to the plurality of two-dimensional images including at least one of the identifying feature images, and controlling the first robot arm to clamp the article and move it onto the second conveyor belt; controlling the second robot arm to move the second video camera in a plurality of steps so that the plurality of surfaces of the article are within a second image capturing range of the second video camera in a plurality of steps; controlling the second video camera to correspondingly capture N article images from N (N is a predetermined positive integer) of the plurality of surfaces of the article in the process in which the second robot arm moves the second video camera in a plurality of steps; and determining whether each of the article images contains at least one defect by performing a feature extraction process on the article images to obtain corresponding feature images, and then performing a feature matching process between each of the feature images and a feature template.

[0010] In a possible embodiment, the automatic target image acquisition and calibration system of the present invention described above comprises: electronic equipmentand is positioned adjacent to the first video camera, electronic equipment a first light source that supplies a first inspection light in response to control from the first light source to brightly illuminate the first image capture range; electronic equipment and is disposed adjacent to the second video camera, electronic equipment and at least one second light source configured to provide second inspection light in response to control from the second light source to illuminate the second image capture range.

[0011] In one possible embodiment, the identification feature is a label, and the label is any one selected from the group consisting of: a 2D barcode tag, a QR code tag, a matrix barcode tag, a tag containing graphic patterns, a tag containing text messages, a tag containing alphabets, and a tag containing numeric figures.

[0012] In one embodiment, the plurality of basic information of the item carried on the label includes a vendor ID (VID), a product ID (PID), a lot number, and a serial number.

[0013] In another embodiment, the identifying feature is one of the recognizable features on the article.

[0014] In one embodiment, the electronic equipment By operating the image processing program, a three-dimensional image of the object is reconstructed according to a plurality of two-dimensional images including at least one of the identification feature images, and then the relative position and the relative three-dimensional coordinates of the object are confirmed.

[0015] In one embodiment, the image processing program includes one image processing algorithm selected from the group consisting of a Delaunay triangulation algorithm, a Crust triangulation algorithm, and a Triangulated Irregular Network (TIN) algorithm.

[0016] In one embodiment, the application program includes a plurality of subprograms, the plurality of subprograms including a first mechanism control program used to control the first robot arm to pick up one of the articles from the first conveyor belt, to clamp the article and move it within the first image capturing range of the first video camera, and then to flip the article according to the flipping sequence; a first video camera control program used to control the first video camera to capture images of the article and obtain the plurality of two-dimensional images when the article is flipped according to the flipping sequence; a label recognition program used to execute the label recognition process on the plurality of two-dimensional images to find at least one of the two-dimensional images including the identification feature image; and a label recognition program used to reconstruct the three-dimensional image of the article according to the plurality of two-dimensional images including the at least one identification feature image, and then to find the front of the article. the image processing program used to enable the first robot arm to next clamp the article and move it onto the second conveyor belt by confirming the relative position and the relative three-dimensional coordinates; a second mechanism control program used to control the second robot arm to move the second video camera in multiple steps so that the multiple surfaces of the article come within the second image capture range of the second video camera in multiple steps; a second video camera control program used to control the second video camera to capture images of the article and obtain the N article images during the process of the second robot arm moving the second video camera in multiple steps; and a defect recognition program used to perform the feature extraction process on the N article images to obtain N feature images, and then perform the feature matching process between each of the feature images and a feature template.

[0017] In one possible embodiment, electronic equipment teeth ,before The first conveyor belt and the second conveyor belt are information-coupled to a machine control device for controlling the first conveyor belt and the second conveyor belt.

[0018] In another possible embodiment, the electronic equipment is integrated into a machine controller for controlling the first conveyor belt and the second conveyor belt, and the machine controller is included in an automated production line. Effect of the Invention

[0019] According to the present invention, the relative position and relative three-dimensional coordinates of the article are established by using the method of identifying the laser label, and then, by controlling the carrying and moving of one video camera by one robot arm, the video camera can be accurately faced with the inspected surface of the article, realizing automatic target image capture of the article, and ensuring the defect inspection accuracy. Moreover, when the defect inspection system of the present invention is applied to inspect other articles for defects, the technician does not need to manually adjust the installation height and installation angle of the video camera, and does not need to tune the light source direction and intensity. [Brief description of the drawings]

[0020] [Figure 1] FIG. 1 is a perspective view showing a typical thread defect inspection system according to a conventional technology. [Figure 2A] FIG. 1 is a first perspective view of an automatic target image capture and calibration system adapted for inspection according to the present invention. [Figure 2B] FIG. 2 is a second perspective view of the automatic target image capture and calibration system of the present invention. [Diagram 3] 2C is a block diagram of the first robot arm, the first video camera, the second robot arm, the second video camera, and the electronic device shown in FIG. 2B. [Figure 4] FIG. 4 is a third perspective view of the automatic target image capture and calibration system applied for inspection of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] In order to more clearly describe the automatic target image capture and calibration system applied for the inspection proposed by the present invention, a preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Please refer to Figs. 2A and 2B, which respectively show a first and a second perspective view of an automatic target image capture and calibration system 1 adapted for inspection according to the present invention. As shown in FIG. 2A and FIG. 2B, the automatic target image capture and calibration system 1 of the present invention includes a first robot arm 11, a first video camera 12, at least one second robot arm 13, at least one second video camera 14, electronic equipment According to the design of the present invention, the first robot arm 11 is installed to be adjacent to the first conveyor belt 1F and the second conveyor belt 1P at the same time, among which, the first conveyor belt 1F conveys a plurality of articles 2, and at least one surface of the articles 2 is provided with an identification feature 21. In one possible embodiment, the identification feature 21 is a label, for example a 2D barcode tag, a QR code tag, a matrix barcode tag, a tag containing graphic patterns, a tag containing text messages, a tag containing alphabets or a tag containing numeric figures, whereby the label may be a laser label made on one surface of the article 2 using laser engraving technology, or a label sticker.

[0023] In another possible embodiment, the identification feature 21 is one of the recognizable features on the article 2. Specifically, if the identification feature 21 is a laser label, the multiple pieces of basic information of the article 2 that can be simultaneously mounted on the laser label (i.e., the identification feature 21) include a vendor identification number (Vendor ID, VID), a product identification number (Product ID, PID), a lot number, and a serial number. Moreover, in the technology of the present invention, the identification feature 21 is used to simultaneously recognize the object angle of the article 2.

[0024] 2A and 2B, the first video camera 12 is disposed adjacent to the first robot arm 11. More specifically, the at least one second robot arm 13 is disposed adjacent to the second conveyor belt 1P, and each second robot arm 13 holds a second video camera 14. 3 shows the first robot arm 11, the first video camera 12, the second robot arm 13, the second video camera 14, and electronic equipment FIG. 15 is a block diagram of the 15. As shown in FIGS. 2A, 2B and 3, electronic equipment 15 is electrically connected to the first robot arm 11, the first video camera 12, the at least one second robot arm 13, and the at least one second video camera 14, and includes a microprocessor 15P and a memory 15M. According to the design of the present invention, the memory 15M stores an application program compiled using a programming language, and also contains a database 157. It should be understood that during normal operation of the automatic target image capture and calibration system 1 of the present invention, the microprocessor 15P executes the application program through access to the memory 15M. More specifically, the application program includes a plurality of sub-programs, and the plurality of sub-programs include a first mechanism control program 151, a first video camera control program 152, a label recognition program 153, an image processing program 150, a second mechanism control program 154, a second video camera control program 155, and a defect recognition program 156.

[0025] During normal operation, the microprocessor 15P executes the first mechanism control program 151 to control the first robot arm 11 to pick up one item 2 from the first conveyor belt 1F, hold the item 2 and move it within the first image capture range of the first video camera 12, and then flip the item 2 over in a flipping order. In the process of flipping the object 2 according to the flip order, the microprocessor 15P simultaneously executes the first video camera control program 152 to control the first video camera 12 to sequentially capture a plurality of 2D images of the object 2. Continuously, the microprocessor 15P executes the label recognition program 153 to perform label recognition processing on the plurality of 2D images to find at least one of the 2D images including an identification feature image. In other words, after the object 2 is fed onto the first conveyor belt 1F, the automatic target image capture and calibration system 1 of the present invention immediately flips the object 2 to find the identification feature 21, thereby confirming the object angle of the object 2.

[0026] As shown in Figures 2A, 2B and 3, after the identification feature 21 is found by flipping the item 2, the object angle and spatial position of the item 2 are immediately confirmed. Next, the microprocessor 15P executes the image processing program 150 to reconstruct a 1D image or a 3D image of the item 2 according to the plurality of 2D images including at least one image of the identification feature, and then confirms the relative position and the relative 3D coordinates of the item 2. In one embodiment, the image processing program 150 includes an image processing algorithm, which may be, for example, but is not limited to, a Delaunay triangulation algorithm, a Crust triangulation algorithm, or a Triangulated Irregular Network (TIN) algorithm.

[0027] After grasping the relative position and relative three-dimensional coordinates of the article 2, the first robot arm 11 moves and places the article 2 on the second conveyor belt 1P under the control of the microprocessor 15P. Thereafter, as the microprocessor 15P executes the second mechanism control program 154, the second robot arm 13 controls the second video camera 14 to move in multiple steps so that the multiple surfaces of the article 2 are within the second image capture range of the second video camera 14 in multiple steps. In other words, the second robot arm 13 is controlled to move the second video camera 14 in multiple steps so that the second video camera 14 faces the multiple surfaces of the article 2 in multiple steps.

[0028] In the process of the second robot arm 13 moving the second video camera 14 in multiple steps, the microprocessor 15P executes the second video camera control program 155 to control the second video camera 14 to capture N to N object images of the multiple surfaces of the object 2, where N is a predetermined positive integer. In general, an industrial product (i.e., the object 2) has six surfaces including an upper surface, a lower surface, a front surface, a rear surface, a left surface, and a right surface. When performing an appearance inspection, the most detailed appearance inspection is performed on some surfaces of the six surfaces of the industrial product according to the relevant inspection regulations. Therefore, the industrial product can be said to have N test surfaces. Thus, it can be understood that after the object 2 is placed on the second conveyor belt 1P, the microprocessor 15P immediately controls the second robot arm 13 to move the second video camera 14 in multiple steps, thereby facing the N test surfaces of the object 2 in multiple steps, and correspondingly acquiring N object images from the N test surfaces.

[0029] Finally, the microprocessor 15P executes the defect recognition program 156 to perform feature extraction processing on the article image to obtain corresponding feature images, and then performs a feature matching process between each of the feature images and a feature template to determine whether each of the article images contains at least one defect. It should be understood that the defect recognition program 156 includes at least one pre-trained article defect recognition model, which completes the feature matching between the feature image and the feature template and identifies the type of defect contained in the feature image.

[0030] FIG. 4 is a third perspective view of the automatic target image capture and calibration system 1 applied for inspection of the present invention. As shown in FIG. 4, in a possible embodiment, the automatic target image capture and calibration system 1 of the present invention may further include a first light source 16 and at least one second light source 17. electronic equipment 15 and is disposed adjacent to the first video camera 12, electronic equipment In response to control from the second light source 15, the first inspection light is supplied to brightly illuminate the first image capture range. electronic equipment 15 and is positioned adjacent to the second video camera 14, electronic equipment In response to control from 15, a second inspection light is supplied to brightly illuminate the second image capture range.

[0031] To further explain, the automatic target image capture and calibration system 1 of the present invention electronic equipment 15 is ,before The first conveyor belt 1F and the second conveyor belt 1P are connected to a machine control device for controlling the first conveyor belt 1F and the second conveyor belt 1P, and the machine control device is included in an automated production line. electronic equipment 15 can also be directly integrated into the machine control device for controlling the first conveyor belt 1F and the second conveyor belt 1P.

[0032] The above is a complete and clear description of the automatic target image capture and calibration system applied for inspection according to the present invention. However, it should be emphasized that the above detailed description is a concrete description of the workable embodiments of the present invention, but the scope of the present invention is not limited to these embodiments, and the equivalent implementations or modifications thereof are still within the scope of the claims of the present application without departing from the technical spirit of the present invention. [Explanation of symbols]

[0033] <Conventional Technology> 1a Thread defect inspection system 11a Transport platform 12a Filming equipment 13a Image processing device 14a Adjustable support mechanism <The present invention> 1. Automatic target image acquisition and calibration system 11 First Robot Arm 1F 1st conveyor belt 1P 2nd conveyor belt 12 No. 1 Video Camera 13 Second Robot Arm 14 2nd Video Camera 15 electronic equipment 15P Microprocessor 15M memory 150 Video Processing Program 151 First mechanism control program 152 First video camera control program 153 Label Recognition Program 154 Second mechanism control program 155 Second video camera control program 156 Defect Awareness Program 157 Database 16 1st light source 17 Second light source 2 Goods 21 Identifying Features

Claims

1. 1. An automatic target image capture and calibration system adapted for inspection, comprising a first robotic arm, a first video camera located proximate to the first robotic arm, at least one second robotic arm, and electronics, the first robot arm is disposed so as to approach the first conveyor belt and the second conveyor belt simultaneously, the first conveyor belt conveys a plurality of articles, and an identification feature is disposed on at least one surface of the articles; the at least one second robot arm is disposed adjacent to the second conveyor belt, and each second robot arm holds a second video camera; The electronic device is electrically connected to the first robot arm, the first video camera, the at least one second robot arm, and the at least one second video camera, and includes a microprocessor and a memory, and an application program edited using a programming language is stored in the memory, whereby the microprocessor executes the application program through access to the memory, thereby starting the following functions: a function of controlling the first robot arm to pick up one of the articles from the first conveyor belt, hold the article, move it within a first image capture range of the first video camera, and then turn the article over in a turn-over order; In the process of flipping the object in the flipping order, the first video camera is controlled to sequentially capture a plurality of two-dimensional images of the object; A function of performing a label recognition process on the plurality of said two-dimensional images to find at least one said two-dimensional image including an identification feature image; a function of confirming a relative position and a relative three-dimensional coordinate of the object according to a plurality of the two-dimensional images including at least one of the identification feature images, and controlling the first robot arm to clamp the object and move it onto the second conveyor belt; the second robot arm controls the second video camera to move in a plurality of times so that the plurality of surfaces of the article are within a second image capturing range of the second video camera in a plurality of times; and controlling the second video camera to capture N object images from N object images (N is a predetermined positive integer) among a plurality of surfaces of the object during a process in which the second robot arm moves the second video camera in a plurality of times. and dynamically performing a function of: performing a feature extraction process on the article image to obtain a corresponding feature image; and then performing a feature matching process between each of the feature images and a feature template to determine whether each of the article images contains at least one defect. Automatic target image acquisition and calibration system.

2. 2. The automatic target image capture and calibration system of claim 1, further comprising: a first light source electrically connected to the electronics and positioned adjacent to the first video camera to provide a first inspection light in response to control from the electronics to illuminate the first image capture range; and at least one second light source electrically connected to the electronics and positioned adjacent to the second video camera to provide a second inspection light in response to control from the electronics to illuminate the second image capture range.

3. 2. The automatic target image capture and calibration system of claim 1, wherein the identification feature is a label, and the label is any one selected from the group consisting of a 2D barcode tag, a QR code tag, a matrix barcode tag, a tag containing graphic patterns, a tag containing text messages, a tag containing alphabets, and a tag containing numeric figures.

4. 2. The automatic target image capture and calibration system of claim 1, wherein the identifying feature is one of the recognizable features on the article.

5. 4. The automatic target image capture and calibration system of claim 3, wherein the plurality of basic information of the item mounted on the label includes a vendor ID (VID), a product ID (PID), a lot number, and a serial number.

6. The automatic target image capture and calibration system of claim 1, characterized in that the electronic device reconstructs a one-dimensional image or a three-dimensional image of the object according to a plurality of the two-dimensional images including at least one of the identification feature images according to an image processing program, and then confirms the relative position and the relative three-dimensional coordinates of the object.

7. 7. The automatic target image acquisition and calibration system according to claim 6, wherein the image processing program includes one image processing algorithm selected from the group consisting of a Delaunay triangulation algorithm, a Crust triangulation algorithm, and a Triangulated Irregular Network (TIN) algorithm.

8. The application program includes a plurality of subprograms, the plurality of subprograms being: a first mechanism control program used to control the first robot arm to pick up one of the articles from the first conveyor belt, to pinch and move the article within the first image capturing range of the first video camera, and then to turn over the article according to the turnover sequence; a first video camera control program used to control the first video camera to capture images of the article and obtain the plurality of two-dimensional images when turning over the article according to the turnover sequence; a label recognition program used to execute the label recognition process on the plurality of two-dimensional images to find at least one of the two-dimensional images including the identification feature image; and a label recognition program used to reconstruct the three-dimensional image of the article according to the plurality of two-dimensional images including the at least one identification feature image, and then to confirm the relative position and the relative three-dimensional coordinates of the article.

7. The automatic target image capture and calibration system of claim 6, further comprising: an image processing program used to enable the first robot arm to clamp the article and move it onto the second conveyor belt; a second mechanism control program used to control the second robot arm to move the second video camera in multiple steps so that the multiple surfaces of the article are within the second image capture range of the second video camera in multiple steps; a second video camera control program used to control the second video camera to capture images of the article and obtain the N article images during the process of the second robot arm moving the second video camera in multiple steps; and a defect recognition program used to perform the feature extraction process on the N article images to obtain N feature images, and then perform the feature matching process between each of the feature images and a feature template.

9. 9. The automatic target image capture and calibration system of claim 8, wherein the electronic device is information-coupled to a machine controller for controlling the first and second conveyor belts.

10. 9. The automatic target image capture and calibration system of claim 8, wherein the electronics is integrated into a machine control device for controlling the first transport belt and the second transport belt, and the machine control device is included in an automated production line.

Citation Information

Patent Citations

  • Device for detecting quality of keycaps

    CN105414042A

  • Inspection method, inspection apparatus, processing apparatus, program, and recording medium

    JP2017015396A

  • Exterior appearance inspection method

    JP2018059830A

  • Method of controlling robot system, and robot system

    JP2019119027A

  • Image Acquisition System And Image Acquisition Method

    US20200080944A1