CN115343293A - Negative plate detection control system and production line based on machine vision - Google Patents
Negative plate detection control system and production line based on machine vision Download PDFInfo
- Publication number
- CN115343293A CN115343293A CN202210956044.6A CN202210956044A CN115343293A CN 115343293 A CN115343293 A CN 115343293A CN 202210956044 A CN202210956044 A CN 202210956044A CN 115343293 A CN115343293 A CN 115343293A
- Authority
- CN
- China
- Prior art keywords
- cathode plate
- production line
- image
- grade
- robot arm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 122
- 238000001514 detection method Methods 0.000 title claims abstract description 107
- 230000007547 defect Effects 0.000 claims abstract description 138
- 238000007689 inspection Methods 0.000 claims description 28
- 238000005286 illumination Methods 0.000 claims description 9
- PXHVJJICTQNCMI-UHFFFAOYSA-N Nickel Chemical compound [Ni] PXHVJJICTQNCMI-UHFFFAOYSA-N 0.000 description 58
- 239000000047 product Substances 0.000 description 57
- 229910052759 nickel Inorganic materials 0.000 description 29
- 238000000034 method Methods 0.000 description 24
- 230000000875 corresponding effect Effects 0.000 description 19
- 238000012545 processing Methods 0.000 description 12
- 230000002950 deficient Effects 0.000 description 9
- 238000003708 edge detection Methods 0.000 description 9
- 239000013598 vector Substances 0.000 description 9
- 230000001276 controlling effect Effects 0.000 description 8
- 238000005868 electrolysis reaction Methods 0.000 description 8
- 238000001914 filtration Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 239000013589 supplement Substances 0.000 description 5
- 230000001788 irregular Effects 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000011148 porous material Substances 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009713 electroplating Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3422—Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Processing (AREA)
Abstract
The invention provides a cathode plate detection control system based on machine vision and a production line, and belongs to the technical field of cathode plate detection. The system comprises: the edge control server is used for determining defect classification and/or defect classification of the cathode plate image; the industrial camera is connected with the edge control server and used for acquiring a cathode plate image of a cathode plate on an operation station and transmitting the cathode plate image to the edge control server; and the production line controller is connected with the edge control server and is also connected with the production line robot arm and used for executing the overturning and/or moving of the cathode plate through the production line robot arm based on the information and/or the control instruction of the detection result transmitted by the edge control server. The invention is used for automatically and rapidly realizing the classification and the classification of the defects of the cathode plate.
Description
Technical Field
The invention relates to the technical field of cathode plate detection, in particular to a cathode plate detection control system based on machine vision and a production line.
Background
Under a new era, the development of new energy industry has high-quality requirements, and the creation of products with hard quality and high-efficiency production capacity gradually become key indexes for acquiring rich domestic market resources and opening overseas market spaces. In the battery component manufacturing industry, the demand of market terminals for components has increased year by year, and a high-quality component foundation is also required to realize a high-quality terminal product.
The common battery cathode plate is usually manufactured by an electrolytic method, such as a nickel electrolytic cathode plate (nickel plate), and irregular defects such as air holes and/or nodules occur due to the fluctuation of an electrolytic process in the manufacturing process of the cathode plate, and the quality of the cathode plate product is seriously affected due to the irregular defects such as a large number of air holes and/or nodules. The occurrence of unevenness defects is difficult to completely avoid, and the unevenness defects are important factors for determining whether manufactured cathode plate products are qualified and for grade assessment.
At present, the detection of the cathode plate products is to detect whether the products are qualified or not and finish grade marking based on experience of production line workers, for example, grade classification or grade marking is carried out between unqualified cathode plate products and between qualified cathode plate products, and the efficiency of the cathode plate production line is poor.
Disclosure of Invention
The invention aims to provide a cathode plate detection control system and a cathode plate detection production line based on machine vision, which avoid the fluctuation of the quality of cathode plate products caused by manual detection of cathode plates, further realize a fully-automatic cathode plate detection production line, and improve the quality, cost and inspection efficiency of cathode plate products.
In order to achieve the above object, an embodiment of the present invention provides a cathode plate detection control system based on machine vision, including:
the industrial camera is used for acquiring a cathode plate image of the cathode plate on the operation station;
the edge control server is connected with the industrial camera and used for determining the defect classification and/or defect classification of the cathode plate image based on the cathode plate image transmitted by the industrial camera;
and the production line controller is connected with the edge control server and is also connected with the production line robot arm and used for executing the overturning and/or moving of the cathode plate through the production line robot arm based on the information and/or the control instruction of the detection result transmitted by the edge control server.
Specifically, the production line controller is used for controlling the production line robot arm to move the cathode plate to a specified sorting channel based on the information of the detection result.
Specifically, the information of the detection result is represented by a numerical value of the defect classification, and the sorting channels comprise a sorting channel of a qualified cathode plate and a sorting channel of an unqualified cathode plate.
Specifically, the production line controller is specifically configured to:
when the numerical value is determined to be unqualified information, the negative plate is moved to a sorting channel of the unqualified negative plate by controlling the production line robot arm;
and when the numerical value is determined to be qualified information, the production line robot arm is controlled to move the cathode plate to the sorting channel of the qualified cathode plate.
Specifically, the information of the detection result is represented by a grade fraction, which is calculated by classifying the defects under each defect classification, and the sorting channels comprise a good-grade cathode plate, a sorting channel of a qualified cathode plate and a sorting channel of a unqualified cathode plate.
Specifically, the production line controller is specifically configured to:
when the grade is determined to be higher than the marking grade, the cathode plate is moved to a sorting channel of the good-grade cathode plate by controlling the production line robot arm;
when the grade is determined to be lower than the score grade and higher than the qualified grade, moving the cathode plate to a sorting channel of the qualified grade cathode plate by controlling the production line robot arm;
and when the grade is determined to be lower than the qualified grade, the cathode plate is moved to the sorting channel of the unqualified grade cathode plate by controlling the production line robot arm.
Specifically, the edge control server is specifically configured to receive a first side image and a second side image of the cathode plate; wherein the work station comprises two station location areas; the first surface image and the second surface image are respectively acquired by the industrial cameras on the two station position areas, and one is a cathode plate image of the cathode plate after being turned by the production line robot arm;
the production line controller is further used for receiving a control instruction transmitted by the edge control server, and the control instruction is used for controlling the production line robot arm to overturn the cathode plate and move the cathode plate to one of the two station position areas.
Specifically, the edge control server is specifically configured to receive a first side image and a second side image of the cathode plate; wherein both the first side image and the second side image are acquired by an industrial camera on a station position area of the operation station, and one is a cathode plate image of the cathode plate after being turned over by the production line robot arm;
the production line controller is further used for receiving a control instruction transmitted by the edge control server, and the control instruction is used for controlling the production line robot arm to overturn the cathode plate.
Specifically, this negative plate detection control system still includes:
an illumination system disposed at the work station.
The embodiment of the invention provides a production line which comprises the cathode plate detection control system based on machine vision.
The invention realizes an automatic and rapid cathode plate detection control system, provides defect detection and product grading scribing for a large-scale and batched cathode plate production line (assembly line/production line), realizes automatic sorting based on defect grading/grade scribing, and improves the efficiency and cost of the cathode plate detection process.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 is a schematic diagram of an exemplary cathode plate inspection control system according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating a detection flow of an exemplary edge control server according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an exemplary cathode plate inspection control system according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of the functional overview and data processing flow of an exemplary cathode plate detection control system according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an exemplary first cathode plate defect detection scenario according to an embodiment of the invention;
fig. 6 is a schematic diagram of an exemplary second cathode plate defect detection scenario according to an embodiment of the invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
Example 1
The embodiment of the invention provides a cathode plate detection control system based on machine vision. The cathode plate detection control system can be an intelligent and automatic system for checking the appearance quality of the nickel plate. As shown in fig. 1, the cathode plate detection control system may include:
the industrial camera is used for acquiring a cathode plate image of the cathode plate on the operation station;
the edge control server is connected with the industrial camera and used for determining the defect classification and/or defect classification of the cathode plate image based on the cathode plate image transmitted by the industrial camera;
and the production line controller is connected with the edge control server and is also connected with the production line robot arm and used for executing the overturning and/or moving of the cathode plate based on the information and/or the control instruction of the detection result transmitted by the edge control server through the production line robot arm. And the detection result comprises defect classification and/or defect classification determined by the edge control server based on the cathode plate image acquired and transmitted by the industrial camera, and both the defect classification and the defect classification can be represented by numerical values.
In the embodiment of the present invention, the edge control server may be a virtual server or a physical server, the server may be configured with a container, and the container may be loaded with a script file, an image file/an installed program; the image file/program may include a file of a pre-trained convolutional neural network model and a computational program; the image file/program may be used to perform defect detection of the cathode plate, and the image file/program may receive the aforementioned cathode plate image and output information of defect classification and/or defect classification. In other applications, the server may be configured with an executable script file or a compiled program, the script file/compiled program may have the same functions as the aforementioned model and calculation program and may be used to perform defect detection of the cathode plate, and the script file/compiled program may receive the aforementioned cathode plate image and output information of defect classification and/or defect classification.
The aforementioned model is used to output a defect classification based on the received cathode plate image. The foregoing calculation procedure may be used for defect classification of the cathode plate image under each defect classification, and the calculation procedure may be configured with one or more area ratio thresholds, and the defect classification of the cathode plate image is determined based on the calculated area ratio, relative to the area ratio thresholds. The calculating the area ratio may specifically include any one of the following:
calculating the area ratio according to the area of the defect area in the cathode plate image and the design area of the cathode plate;
calculating an area ratio by the area of the defect region in the cathode plate image and the area calculated by the edge size of the cathode plate image in the cathode plate image;
after the cathode plate image is rasterized, the area ratio is calculated based on the pixel points occupied by the hole image and/or the nodule image in the defect region in the cathode plate image and the pixel points occupied by the cathode plate image in the cathode plate image. In the embodiment of the invention, the position point of the image is the coordinate point of the rasterized image, the pixel is the minimum cell described by the position point of the image, the minimum cell has a color numerical value, and the pixel point is the minimum cell.
The cathode plate detection control system can also comprise: the lighting system is arranged at the operation station and used for improving the exposure or shooting of the industrial camera to the cathode plate; and the quality inspection terminal is used for presenting the detection results of the cathode plate image and the cathode plate image. The lighting system may comprise a light source and a light source controller. The industrial camera may include a visible light camera and a lens. In some high-efficiency production scenarios, the first side of the cathode plate to be detected may correspond to a first illumination system and a first industrial camera, and after the first side has completed the detection of the cathode plate, the second side of the cathode plate is detected after the cathode plate is turned over and moved, at which time the second side may correspond to a second illumination system and a second industrial camera, so that the first illumination system and the first industrial camera may be used to detect the first side of other undetected cathode plates at the beginning/during the detection of the second side of the cathode plate.
In the embodiment of the invention, the cathode plate detection control system is used as an independent unit of the nickel electrolysis plate arrangement inspection unit to perform appearance inspection on the nickel electrolysis cathode plate, the analysis and detection result is stored in a local database, the input conditions of the cathode plate detection control system are that an industrial camera shoots and corresponds to an illumination system starting signal and a cathode plate current station position signal, and the output result is a digital quantity signal representing qualification and ineligibility, namely the appearance quality of the cathode plate is qualified and unqualified.
The industrial camera may have a function of capturing a photograph and a function of video photographing, for example, a gigabit industrial GigE camera is selected. The production line controller can select a PLC controller. The power supply of the lighting and industrial cameras is provided by a PLC control system, and the signals of the lighting system and the states of the signals are controlled by the PLC. In order to facilitate the capacity expansion and the upgrade of the later-stage system, the GigE industrial camera is connected with the edge control server through a ten-million card rail type network switch. The industrial camera is connected with the gigabit card rail type network switch sequentially through a signal controller and a distributed IO (or a connector/converter with the same function); and the light source controller is connected with the gigabit card rail type network switch through distributed IO. The quality inspection terminal is connected with the gigabit card rail type network switch and can comprise a display, a workstation server and a human-computer interaction device.
In one example of the embodiment of the present invention, the information of the detection result may be represented by a numerical value of defect classification, and the designated sorting passage includes a sorting passage of a qualified cathode plate and a sorting passage of a unqualified cathode plate. And when the production line controller determines that the numerical value is unqualified information, the production line controller controls a production line robot arm to move the cathode plate to the sorting channel of the unqualified cathode plate. And when the production line controller determines that the numerical value is qualified information, the production line controller controls the production line robot arm to move the cathode plate to the sorting channel of the qualified cathode plate.
In another example of the embodiment of the present invention, the information of the detection result may be represented by a grade, which is calculated from the statistics of defect classification under each defect classification, and the sorting channels include a good-grade cathode plate, a sorting channel of a good-grade cathode plate, and a sorting channel of a defective cathode plate. When the production line controller determines that the grade is higher than the marking grade, the production line controller controls the robot arm of the production line to move the cathode plate to a sorting channel of the cathode plate with a good grade; when the production line controller determines that the grade is lower than the marking grade and higher than the qualified grade, the production line controller controls a robot arm of a production line to move the cathode plate to a sorting channel of the qualified grade cathode plate; and when the production line controller determines that the grade is lower than the qualified grade, the production line controller controls the production line robot arm to move the cathode plate to the sorting channel of the unqualified grade cathode plate. In the above examples, lower may mean less than, higher may mean greater than or equal to; the accept fraction is less than the scribe fraction and the reject fraction is less than the accept fraction.
As an illustrative example of the present disclosure, as shown in fig. 2, the edge control server may be specifically configured to: carrying out edge detection on the acquired image by utilizing automatic edge detection, and delimiting a detected ROI; performing light supplement processing on the image by using an image light supplement algorithm, inputting a nickel plate LOGO sample, and identifying the nickel plate LOGO in the defined detection ROI by using a template matching method; the method comprises the steps of utilizing an automatic threshold segmentation method to segment image data at a position of a filtering LOGO to obtain a defective area, extracting a feature vector of the part, wherein the feature vector comprises geometric features, color features, texture features and designated features, loading a pre-trained neural network model, sending the features into the model, predicting by a classifier of the model, determining defect classification, realizing defect classification through pixel statistics of defects, taking the defect classification and the defect classification as detection results (or taking the grade of a lineation/the grade score of a calculation as the detection results), outputting the detection results to a channel corresponding to a PLC (programmable logic controller) of a nickel electrolytic plating inspection unit through a serial port RS485 port, and controlling a robot arm to execute corresponding actions based on the defect classification by the PLC, such as moving a nickel plate to a processing channel corresponding to each defect classification and defect classification or turning over the nickel plate, so as to detect the other surface of the nickel plate. The edge control server can avoid the product identification to cause interference to the defect detection of the cathode plate.
In the embodiment of the invention, the edge control server inspects the appearance surface of the nickel plate, namely, the industrial camera acquires the appearance image of the nickel plate, the edge control server (optionally) performs image analysis and processing, database management, data information storage and external release in a non-real-time manner, and the final result information of the image processing is transmitted to a real-time system (a PLC (programmable logic controller) and a robot arm) of a production line. The detection result can be stored in a local database configured in the edge control server, and the appearance quality inspection control condition of the nickel plate is information such as an industrial camera snapshot and a starting signal of a corresponding lighting system, the current position state of the nickel plate and the like. Therefore, as shown in fig. 3, the cathode plate detection control system realizes the acquisition and light source control of the cathode plate image, the image preprocessing (i.e., edge detection and segmentation), then the detection, identification and rating of the defect detection defect area are realized, and finally, the detection result can be stored, managed and output and presented in a report form.
Referring to fig. 4, the edge control server and the quality inspection terminal may form a machine vision inspection system, which sends light source control signals and industrial camera control signals to front-end hardware (light source and industrial camera) and receives image data (at least one frame of cathode plate image) sent back by the industrial camera. The edge control server is used for: the system collects real-time data, preprocesses, extracts and detects defects, identifies and classifies, grades the defects, distributes results, monitors the system service state and the state of relevant hardware, and the like. The quality inspection terminal is used for: the method comprises the following steps of displaying/storing a detection result, confirming the detection result, carrying out statistical analysis, generating a report, applying an expert level and the like. The edge control server can send the detection result and the cathode plate image (which can be an original cathode plate image) on the operation station to the quality inspection terminal.
In an exemplary embodiment of the disclosure, as shown in fig. 5, a production line of a cathode plate detection control system is used, the production line has a plate arrangement inspection unit (which may be regarded as a cathode plate detection channel, that is, in practice, there may also be multiple plate arrangement inspection units, that is, multiple cathode plate detection channels, and a production line controller may control at least one cathode plate detection channel), the plate arrangement inspection unit has a plate conveying structure, and the plate arrangement inspection unit may be divided into multiple functional areas, for example, a plate arrangement area 100, an operation station (including two station position areas, a first station position area 101 and a second station position area 102), and a grade marking area 103. The production line may also include sorting lanes 104, 105, 106, the configuration of which may be designed according to specific product quality requirements and production line characteristics. For example, only the sorting channels of the conforming cathode plates and the nonconforming cathode plates may be configured with the sorting channel 104 as the channel for the conforming products (conforming cathode plates), and the sorting channel 105 as the channel for the nonconforming products (nonconforming cathode plates), and for example, the sorting channel for sorting by grade may be configured with the sorting channel 104 as the channel for the good-grade products (good-grade cathode plates), with the sorting channel 105 as the channel for the normal-grade products (conforming cathode plates), and with the sorting channel 106 as the channel for the nonconforming products (nonconforming cathode plates). The ranking area 100 and the grade marking area 103, as well as the sorting lanes 104, 105, 106, are configured with a transport speed V (speed values can be configured independently/the same). This arrange board inspection unit can also include a plurality of production line robotic arms that are controlled by production line controller, produces line robotic arm and can realize or realize etc. through free mobile formula multiaxis industrial robot through guide rail type industrial robot. The production line robot arm 107 may be used for movement operations of the cathode plates of the work stations, and the production line robot arm 108 may be used for sorting operations of the cathode plates. In some application scenarios, the in-line robotic arm 107 may also be used for the operation of sorting rejected cathode plates into sorting lanes 106.
The state information (for example, attitude data, task data being executed, and the like) of each production line robot arm can be stored into a designated storage space (or unit address) by the production line controller, the edge control server can access the unit addresses through a serial port RS485 port to obtain the state information of each production line robot arm, and a designated control instruction can be generated conveniently to realize the operation of the cathode plate, for example, if the attitude of the production line robot arm 108 does not return to the designated attitude, the control instruction can include a reset instruction and an instruction of sorting operation after resetting is needed. For the work station, the first station position area 101 has a first station position identifier 01, and the second station position area 102 has a second station position identifier 00. The production line is also provided with industrial cameras 109 and 111 and lighting systems 110 and 112, the industrial cameras 109 and 111 and the lighting systems 110 and 112 receive control instructions (such as shooting instructions and light brightness adjusting instructions) of the edge control server or start power switch signals through the controller, and the industrial cameras 109 and 111 transmit cathode plate images to the edge control server through the controller.
The industrial camera 109 and the lighting system 110 are optically aligned with the first station position area 101 and can capture the first station position identifier 01 and the cathode plate image, so that the first station position identifier 01 can be used for associating the corresponding industrial camera 109, the lighting system 110 and the photographed cathode plate image, and an edge control server can conveniently issue a control instruction to a production line controller and perform image data processing. Similarly, second station location identifier 00 can be used to associate the corresponding industrial camera 111, lighting system 112, and captured cathode plate image. In some energy-saving application scenes, any lighting system comprises a plurality of light-emitting channel lamp groups, and each light-emitting channel lamp group can be independently controlled to be turned on or off, so that illumination with adaptive brightness can be applied according to the light reflection degree of the surface of a product, and electric energy can be saved; the power supplies of the lighting system and the industrial camera corresponding to the same station position identification can be synchronously started and closed, so that electric energy is saved when the cathode plate image is not acquired.
In a first cathode plate defect detection scenario, cathode plate 113 and cathode plate 114 (defect 115 with a very small area) are transported into sorting tunnel 104 substantially without defects, cathode plate 116 is about to be sorted by production line robot arm 108 into sorting tunnel 105 due to defect 117 with a larger (area ratio above area ratio threshold) area, and cathode plate 118 with defect 119 with a larger area already exists in sorting tunnel 105. The cathode plate has product marks, which are represented by irregular areas (filled with oblique lines) in fig. 5, such as product marks 127, and for convenience of description, the side of the cathode plate with the product marks is referred to as a front side, and the side without the product marks is referred to as a back side. The grade marking area 103 is provided with a cathode plate 120 with a detected defect 121 and a cathode plate 122 without a detected defect; the second station position area 102 bears a cathode plate 123 which is acquired by the industrial camera 109 and is acquiring a front image and is acquiring a back image by the industrial camera 111, the first station position area 101 bears a cathode plate 124 which is acquiring a front image and is not acquiring a back image by the industrial camera 109, the edge control server can simultaneously turn on the industrial camera 109, the lighting system 110 and the industrial cameras 111 and 112, the edge control server can receive the front image of the cathode plate 124 and the back image of the cathode plate 123 and complete the defect detection of the cathode plate, can respectively record the images of the corresponding surfaces of the cathode plates, and can temporarily turn off the industrial camera 109, the lighting system 110 and the industrial cameras 111 and 112; the panel arrangement area 100 has arranged, ordered cathode panels 125, 126, 128 to be tested. The production line controller may have a sequencing record for the cathode plates in each functional area, so as to control the operation of the production line robot arm, for example, the cathode plates 116 in the first order in the grade marking area 103 are in situ, the production line controller controls the production line robot arm to designate and take out the cathode plates 116 in the first order and sort the cathode plates to the sorting passage 105 based on the information (for example, unqualified information) of the detection result of the cathode plates 116 transmitted by the edge controller, and the production line robot arm takes out the cathode plates in the first order in the current grade marking area 103. The sequential marking designation is only an example manner of automatic control, in other application scenarios, the panel arrangement inspection unit may set a corresponding unique identifier for each cathode panel that is supported, and the unique identifier does not change, the production line robot arm has a function of identifying the unique identifier, for example, the unique identifier is a barcode, a two-dimensional code, or the like set in the product identifier, and the production line robot arm may have a camera, a scanner, or the like.
In the second cathode plate defect detection scene, as shown in fig. 6, the occurrence timing of the second cathode plate defect detection scene is slightly later (e.g., several seconds) than the occurrence timing of the first cathode plate defect detection scene. At this point, the cathode plate 120 on the grade marking area 103 has been removed by the in-line robot arm 108 and is about to be sorted to the sorting lane 105. After the edge control server determines that the double-sided cathode plate defect detection of the cathode plate 123 is completed, the edge control server may transmit a control command for flipping the cathode plate 123 (to make the front side with the product identifier face in a uniform direction) and moving the cathode plate 123 from the second station position area 102 to the grade marking area 103 to the production line controller (the production line robot arm 107 has completed executing the control command in fig. 6) based on the status information of the production line robot arm 107 transmitted by the production line controller. If the cathode plate 123 has a serious defect, the edge control server transmits information of the detection result of the cathode plate 123 to the production line controller based on the status information of the production line robot arm 107 transmitted by the production line controller, and the production line controller can control the production line robot arm 107 to take out and sort the cathode plate 123 to the sorting channel 106. Cathode plate 125 has been moved (by in-line robotic arm 107 in this scenario inertial transport or in other scenarios) to first station location area 101. After the edge control server determines the defect classification or defect classification of the front image (i.e., the first side image) of the cathode plate 124 in the first cathode plate defect detection scenario, control instructions for flipping the cathode plate 124 by 180 ° and moving the cathode plate 124 from the first station position area 101 to the second station position area 102 are transmitted to the production line controller based on the status information of the production line robot arm 107 transmitted by the production line controller, whereas in other application scenarios, the working station may have only one station position area, and only the control instruction for flipping is needed. After the cathode plate 124 is moved to the second station location area 102, the edge control server can again turn on the industrial camera 109, the illumination system 110, and the industrial cameras 111, 112 simultaneously, the edge control server can receive the reverse image (i.e., the second side image) of the cathode plate 124 and the front image of the cathode plate 125 and complete the performance of the defect detection of the cathode plate, and can record the images and detection results in a database relative to the unique identification of the cathode plate 124 and the cathode plate 125, and the edge control server can again briefly turn off the industrial camera 109, the illumination system 110, and the industrial cameras 111, 112. In fig. 5 and 6, the dashed short line frame is a moving target position area or a home position area, and the dotted dot frame is a protruding mark of a defect.
The embodiment of the invention also provides a production line, which can be an automatic production line and is equipment for the cathode plate detection process. The production line can comprise the cathode plate detection control system, the plate arrangement inspection unit and the production line robot arm, and the sorting channel. The embodiment of the invention can realize edge detection of the nickel electrolysis cathode plate, LOGO detection of the nickel electrolysis cathode plate and defect detection of pores or nodules of the nickel electrolysis cathode plate; and meanwhile, estimating the area corresponding to each type of defect and the percentage of the total area by the number of the pixels occupied by each defect, and judging the integral quality grade of the nickel plate. In the prior art, the grade marking and grade classification of the cathode plate products can only be performed by manual operation of production line workers, workers judge the quality of the nickel plates through experience, the quality standards of the nickel plates are difficult to unify, different workers have fluctuation on the inspection results of the nickel plates, the labor intensity of the workers is high, the production operation efficiency is low, the cost is high, meanwhile, when the number of the nickel plates to be inspected is small at a turning plate frame in an operation station, the turning plates are easy to be unstable, the nickel plates have the risk of being thrown out of the turning plate frame, and potential safety hazards exist.
Example 2
The trained convolutional neural network model can be adopted to identify the image of the defect on the metal surface, however, because the product identification is also part of the appearance of the cathode plate, the types of the identification are various, the cathode plate identifications with different specifications are different or change along with requirements, the product identification easily causes the rough, uneven or local irregular contour edge of the smooth surface of the cathode plate, the image of the product identification is difficult to identify as the image of the defect by the convolutional neural network model, manual detection intervention is difficult to avoid at the moment, or a special sample of the training identification is collected to easily cause overfitting or under-fitting of the model to the defect identification, the cost for adjusting and optimizing the model is difficult to receive, the accuracy and precision of the detection result are fluctuant, and the requirements on the accuracy, precision and detection stability of online use are difficult to meet.
The embodiment of the invention belongs to the same inventive concept as the embodiment 1, and provides a cathode plate detection method which can be applied to an edge control server in the embodiment 1 in the forms of a model file and a calculation program, and when the edge control server executes a machine instruction corresponding to the model file and the calculation program, the edge control server executes the cathode plate detection method, so that the interference of product identification on the defect detection of the cathode plate is avoided. The cathode plate detection method can comprise the following steps:
s1) performing edge detection on the collected cathode plate image, and determining cathode plate edge information in the cathode plate image;
s2) based on the cathode plate edge information, defining an interested area in the cathode plate image;
s3) identifying the position area of the product identifier in the interested area, and filtering the position area of the product identifier in the cathode plate image;
s4) segmenting the filtered cathode plate image to obtain an area image of a defective area;
s5) extracting the feature vector of the region image, and determining the defect classification of the defect region through the feature vector.
In the embodiment of the present invention, the cathode plate may be located at a designated operation station, the operation station may have a configured station position identifier, the operation station has a configured industrial camera, and the industrial camera performs image acquisition on the cathode plate and transmits an image (or an image) of the cathode plate corresponding to the station position identifier to the aforementioned server or device.
As an illustrative example of the present disclosure, any one or more of the Sobel, prewitt, roberts, canny, marr-Hildreth, etc. edge detectors may be called, and in some advantageous implementations, the Canny edge detector may be preferred. After edge detection, cathode plate edge information of the cathode plate image may be obtained, which may include location points/pixel points on the contour, to determine the contour edge of the cathode plate image on the cathode plate image.
The cathode plate edge information shows that the area image in the outline edge of the cathode plate image in the cathode plate image is a key local area image, and the rest area image in the cathode plate image can be a background image on an operation station and a non-key local area image of operation machinery and the like. The Region image in the outline edge can be defined as a Region of interest (ROI) in the cathode plate image, so that the influence of the image characteristics in the non-critical Region image, such as a work station background, a roller or a support frame, on defect detection can be avoided through the ROI. The setting of the ROI may be set in combination with the shape of a specific cathode plate, the arrangement orientation of product marks, and the like, such as an angular region and a region of a specified size. It can be understood that the ROI of the embodiment of the present invention is not strictly limited to include only the cathode plate image, and the ROI of the embodiment of the present invention may be divided into a combined region of the cathode plate image region and the remaining regions in order to adapt to the detection scene of the required product.
In order to avoid exposure nonuniformity and fluctuation of random image effect, the negative plate image can be subjected to light supplementing processing through a light supplementing algorithm. The fill-in processing may include adjusting the current cathode plate image based on the average of various parameters such as brightness, contrast, and saturation of the normally exposed and not excessively dark/excessively bright region, and may be adjusted only for the region image within the ROI.
In an exemplary embodiment disclosed in the embodiments of the present invention, the cathode plate may be a nickel plate, the product identifier may be a commercial identifier or a LOGO (LOGO) of the cathode plate, the nickel plate LOGO image sample is called, and the position region of the nickel plate LOGO is identified by using a template matching method in the defined detection ROI, thereby implementing a simple and fast LOGO position determination step. In some advantageous embodiments, nickel plate LOGO image sample sets can be configured in the database, each sample set can be bound with the station position identification of the designated work station, and even if the cathode plate placed by the robot arm on the production line has various and different product identifications, the rapid matching identification can be determined.
The position area of the product mark in the image of the filter cathode plate can adopt a filling and overlapping method. The method specifically comprises the following steps:
filling a position area of the product identification;
and superposing the filled area image and the cathode plate image, and taking the superposed image as the filtered cathode plate image.
The filling may include initializing a mask corresponding to the location area of the product identifier, assigning a value to the mask based on pixels of the image of the defect-free cathode plate, where the assigned mask is a filled area image, and the filled area image may be on a layer of the cathode plate image and may be opaque.
Because the cathode plate defect can cause the cathode plate image on the cathode plate image to have the abnormal image effect different from the cathode plate with a smooth surface, such as shadow or highlight, the image segmentation method can be adopted to determine the area image containing the defect area. In some advantageous embodiments, an automatic threshold segmentation function/method (Otsu) may be invoked to simply and quickly segment the region image of the defective region.
In the manufacturing process of cathode plates, electrolysis methods are often used, and the defects occurring in cathode plates manufactured by electrolysis methods are characterized by a limited data scale. In view of this, image samples of defective areas of the cathode plate occurring in the electrolytic process and image samples of areas that are substantially regarded as being free of defects may be collected in advance, and a neural network model may be trained using the image samples, resulting in a classifier that can be used on-line instantly. The region image of the defective region may be input into a neural network model, from which a defect classification of the defective region is determined by its classifier. In the neural network model, the extracted feature vectors may include at least one of geometric features, color features, and texture features, which is not limited to embodiments and may allow for the design and selection of other feature vectors.
If the classifier determines that the area image of the defect area is regarded as the non-defective area image, the subsequent production line process can be executed, and the defect detection of the cathode plate on the operation station is fed back to the database/visualization equipment to be normal, namely the defect detection result is a qualified product. If the defect classification of the defect area is determined (the defect detection abnormality of the cathode plate on the operation station can be fed back to the database/visualization equipment, namely the defect detection result is unqualified), the defect classification of the defect area can be determined according to the relative size of the area image of the defect area and the cathode plate image. Specifically, after determining that the result output by the classifier indicates that the defect in the defect region is classified as a pore defect or a nodule defect or a specified type of defect, the method may further include:
s6) determining the number of pixels of the area image of the defect area in the cathode plate image;
s7) calculating the image area corresponding to the image with the pixel number;
s8) calculating the area ratio of the image area to the total image area of the cathode plate image;
s9) determining the defect classification corresponding to the area ratio under the defect classification.
As an exemplary example of the present disclosure, the total image area may be calculated based on edge information of the cathode plate image in the cathode plate image, for example, the edge information may include values of the length, width, and the like of the cathode plate image. The defect classification may include general defects and critical defects. If the calculated area ratio is below the configured general defect area ratio threshold, it may be determined that the defect within the current defect region is a general defect, and if the calculated area ratio is above the configured general defect area ratio threshold, it may be determined that the defect within the current defect region is a severe defect. Wherein there are a plurality of defect classifications for the cathode plate, coefficients may be assigned to each type of defect by summing the coefficients with the area ratios of the respective defect classifications, the result of the summation being the area ratio of the cathode plate and compared to a threshold value. It is understood that this is not an implementation defined by the embodiments of the present invention, but rather, the defect classification can be selected and designed to conform to the product line and the product.
In some efficient production scenarios, after the defect classification is completed, the cathode plate detection method may further include:
and carrying out grade marking on the cathode plate on the production line.
The marking method comprises the steps of carrying out numerical statistics on defect grades of all defect classifications of the two sides of each negative plate corresponding to a production batch of a production line (for example, carrying out weighted summation on area ratios, wherein the summation result is taken as a grade), and carrying out grade marking, wherein the marking can be based on the comparison between the grade score of each negative plate and a configured grade threshold, if the grade score is higher than the configured grade score, the marking is a good grade or a specified grade, and if the grade score is lower than the configured grade score, the marking is a general grade or other specified grade.
The embodiment of the invention can determine the edge of the cathode plate and the interested area of the cathode plate by carrying out edge detection on the cathode plate image, realizes product identification in the concerned key local area through the interested area so as to be convenient for automatically filtering and avoiding the interference of images of a non-cathode plate, and does not need to manually participate in a defect detection procedure (such as manually shielding product identification), thereby avoiding the influence of the product identification on the prediction result of the area image after the image of the product identification is filtered. The embodiment of the invention further realizes product grading through the image area and the total image area corresponding to the number of the pixels in the defect area. The embodiment of the invention further realizes the adjustment of parameters such as brightness, contrast and the like of the cathode plate image through light supplement processing, improves the precision of subsequent segmentation and feature extraction, and avoids the influence of image effect change caused by random factors on the defect prediction result and classification. The embodiment of the invention further adopts a matching method to directly determine whether the product identification exists in the region of interest, thereby improving the complexity of defect detection and the realization cost. The embodiment of the invention preferably performs filling and superposition to realize the filtration of the product identification, thereby improving the response speed of defect detection. The embodiment of the invention further improves the complexity and the realization cost of defect detection by adopting an automatic threshold segmentation function.
Example 3
The embodiment of the invention and the embodiments 1 and 2 belong to the same inventive concept, and the embodiment of the invention provides a cathode plate detection device, which can comprise:
the edge detection module is used for performing edge detection on the acquired cathode plate image and determining cathode plate edge information in the cathode plate image;
the area dividing module is used for dividing an interested area in the cathode plate image based on the cathode plate edge information;
the filtering identification module is used for identifying the position area of the product identification in the interested area and filtering the position area of the product identification in the cathode plate image;
the region segmentation module is used for segmenting the filtered cathode plate image to obtain a region image of a defective region;
and the defect prediction module is used for extracting the feature vector of the area image and determining the defect classification of the defect area through the feature vector.
In embodiments of the present invention, the cathode plate detection apparatus (or any one of the modules thereof) may be implemented based on hardware such as one or more controllers and/or an electronic device with a processor, and in some cases, the cathode plate detection apparatus may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Array (FPGA), application Specific Integrated Circuit (ASIC), application Specific Standard Product (ASSP), system on a chip (SoC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof.
Specifically, this negative plate detection device can also include:
a defect classification module to: determining the number of pixels of an area image of the defect area in the cathode plate image; calculating an image area corresponding to the image having the number of pixels; calculating the area ratio of the image area to the total image area of the cathode plate image; determining a defect classification corresponding to the area ratio under the defect classification.
Specifically, this negative plate detection device can also include:
the light supplement processing module is used for: and performing light supplement processing on the cathode plate image.
Specifically, identifying the location area of the product identifier in the area of interest includes:
and carrying out template matching on the image sample of the product identifier and each position area in the region of interest to determine the position area of the product identifier.
Specifically, filtering the position area of the product identifier in the cathode plate image includes:
filling a location area of the product identifier;
and overlapping the filled area image and the cathode plate image, and taking the overlapped image as the filtered cathode plate image.
Specifically, the feature vector includes at least one of a geometric feature, a color feature and a texture feature; the cathode plate is a nickel electrolysis cathode plate.
Specifically, the cathode plate detection device can also turn over the cathode plate through a production line controller and a robot arm; this negative plate detection device can also carry out the grade to producing the negative plate on the line and rule.
Example 4
The embodiment of the invention and the embodiments 1 to 3 belong to the same inventive concept, and the embodiment of the invention provides an edge control server.
Edge control servers are intended to mean devices with instruction processing and computing capabilities in various forms, such as computers, industrial computers, servers, etc., the processor and the memory of which can be implemented in the form of a system-on-chip (SoC or MCU) or assembled directly using a circuit board with a connection interface. The memory stores instructions executable by the at least one processor, and the at least one processor implements the method of embodiment 2 by executing the instructions stored by the memory.
Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solutions of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications all belong to the protection scope of the embodiments of the present invention.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, the embodiments of the present invention do not describe every possible combination.
Those skilled in the art will understand that all or part of the steps in the method according to the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present application. While the foregoing storage media may be non-transitory, the storage media may include: a U-disk, a hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a magnetic disk, or an optical disk.
In addition, any combination of various different implementation manners of the embodiments of the present invention is also possible, and the embodiments of the present invention should be considered as disclosed in the embodiments of the present invention as long as the combination does not depart from the spirit of the embodiments of the present invention.
Claims (10)
1. A negative plate detection control system based on machine vision is characterized by comprising:
the industrial camera is used for acquiring a cathode plate image of the cathode plate on the operation station;
the edge control server is connected with the industrial camera and used for determining the defect classification and/or defect classification of the cathode plate image based on the cathode plate image transmitted by the industrial camera;
and the production line controller is connected with the edge control server and is also connected with the production line robot arm and used for executing the overturning and/or moving of the cathode plate through the production line robot arm based on the information and/or the control instruction of the detection result transmitted by the edge control server.
2. The machine vision based cathode plate inspection control system of claim 1,
the production line controller is used for controlling the production line robot arm to move the cathode plate to an appointed sorting channel based on the information of the detection result.
3. The machine-vision based cathode plate inspection control system of claim 2, wherein,
the information of the detection result is represented by the numerical value of the defect classification, and the sorting channel comprises a sorting channel of a qualified cathode plate and a sorting channel of an unqualified cathode plate.
4. The machine-vision-based cathode plate detection control system of claim 3, wherein the in-line controller is specifically configured to:
when the numerical value is determined to be unqualified information, the negative plate is moved to a sorting channel of the unqualified negative plate by controlling a robot arm of the production line;
and when the numerical value is determined to be qualified information, the production line robot arm is controlled to move the cathode plate to the sorting channel of the qualified cathode plate.
5. The machine-vision based cathode plate inspection control system of claim 2, wherein,
the information of the detection result is represented by a grade of the product, which is obtained by calculating the classification of the defects under each defect classification, and the sorting channels comprise a good-grade cathode plate, a sorting channel of a qualified cathode plate and a sorting channel of an unqualified cathode plate.
6. The machine-vision-based cathode plate detection control system of claim 5, wherein the in-line controller is specifically configured to:
when the grade is determined to be higher than the marking grade, the cathode plate is moved to a sorting channel of the good-grade cathode plate by controlling the production line robot arm;
when the grade is determined to be lower than the score grade and higher than the qualified grade, moving the cathode plate to a sorting channel of the qualified grade cathode plate by controlling the production line robot arm;
and when the grade is determined to be lower than the qualified grade, the cathode plate is moved to the sorting channel of the unqualified grade cathode plate by controlling the production line robot arm.
7. The machine vision based cathode plate inspection control system of claim 1,
the edge control server is specifically configured to receive a first side image and a second side image of the cathode plate; wherein the work station comprises two station location areas; the first surface image and the second surface image are respectively collected by the industrial cameras on the two station position areas, and one is a cathode plate image of the cathode plate after being turned by the production line robot arm;
the production line controller is further used for receiving a control command transmitted by the edge control server, and the control command is used for controlling the production line robot arm to overturn the cathode plate and move the cathode plate to one of the two station position areas.
8. The machine vision based cathode plate inspection control system of claim 1,
the edge control server is specifically configured to receive a first side image and a second side image of the cathode plate; wherein both the first side image and the second side image are acquired by an industrial camera on a station position area of the operation station, and one is a cathode plate image of the cathode plate after being turned over by the production line robot arm;
the production line controller is further used for receiving a control instruction transmitted by the edge control server, and the control instruction is used for controlling the production line robot arm to overturn the cathode plate.
9. A machine vision based cathode plate inspection control system according to any one of claims 1 to 8, characterized in that it further comprises:
an illumination system disposed at the work station.
10. A production line characterized in that it comprises a machine vision based cathode plate inspection control system according to claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210956044.6A CN115343293A (en) | 2022-08-10 | 2022-08-10 | Negative plate detection control system and production line based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210956044.6A CN115343293A (en) | 2022-08-10 | 2022-08-10 | Negative plate detection control system and production line based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115343293A true CN115343293A (en) | 2022-11-15 |
Family
ID=83952576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210956044.6A Pending CN115343293A (en) | 2022-08-10 | 2022-08-10 | Negative plate detection control system and production line based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115343293A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000020951A (en) * | 1998-07-03 | 2000-01-21 | Unique Technol Internatl Pte Ltd | Flexible processing system for substrate for hard disk |
| WO2008110061A1 (en) * | 2007-03-12 | 2008-09-18 | 3I Systems Corp | Plane substrate auto-test system and the method thereof |
| CN101566583A (en) * | 2008-04-23 | 2009-10-28 | 财团法人工业技术研究院 | Defect detection system of panel component |
| CN114619428A (en) * | 2022-03-28 | 2022-06-14 | 之视科技(上海)有限公司 | Postweld detection device based on welding robot vision and application algorithm thereof |
| CN114638797A (en) * | 2022-03-09 | 2022-06-17 | 三门三友科技股份有限公司 | Method and device for detecting copper surface defects based on linear array camera |
| CN114723668A (en) * | 2022-03-09 | 2022-07-08 | 三门三友科技股份有限公司 | Cathode copper quality detection method and system based on area-array camera |
-
2022
- 2022-08-10 CN CN202210956044.6A patent/CN115343293A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000020951A (en) * | 1998-07-03 | 2000-01-21 | Unique Technol Internatl Pte Ltd | Flexible processing system for substrate for hard disk |
| WO2008110061A1 (en) * | 2007-03-12 | 2008-09-18 | 3I Systems Corp | Plane substrate auto-test system and the method thereof |
| CN101566583A (en) * | 2008-04-23 | 2009-10-28 | 财团法人工业技术研究院 | Defect detection system of panel component |
| CN114638797A (en) * | 2022-03-09 | 2022-06-17 | 三门三友科技股份有限公司 | Method and device for detecting copper surface defects based on linear array camera |
| CN114723668A (en) * | 2022-03-09 | 2022-07-08 | 三门三友科技股份有限公司 | Cathode copper quality detection method and system based on area-array camera |
| CN114619428A (en) * | 2022-03-28 | 2022-06-14 | 之视科技(上海)有限公司 | Postweld detection device based on welding robot vision and application algorithm thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11488294B2 (en) | Method for detecting display screen quality, apparatus, electronic device and storage medium | |
| CN110659660B (en) | Automatic optical detection classification equipment using deep learning system and training equipment thereof | |
| CN104749184B (en) | Automatic optical detection method and system | |
| CN112053318A (en) | Two-dimensional PCB defect real-time automatic detection and classification device based on deep learning | |
| US11657599B2 (en) | Method for detecting appearance of six sides of chip multi-layer ceramic capacitor based on artificial intelligence | |
| CN109840900A (en) | A kind of line detection system for failure and detection method applied to intelligence manufacture workshop | |
| CN111289538A (en) | PCB element detection system and detection method based on machine vision | |
| CN108257171A (en) | Car radar assembling aperture detection method based on light vision | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| CN102901735B (en) | System for carrying out automatic detections upon workpiece defect, cracking, and deformation by using computer | |
| CN116593479B (en) | Method, device, equipment and storage medium for detecting appearance quality of battery cover plate | |
| WO2023134286A1 (en) | Online automatic quality testing and classification method for cathode copper | |
| CN116385430A (en) | Machine vision flaw detection method, device, medium and equipment | |
| CN117392042A (en) | Defect detection method, defect detection apparatus, and storage medium | |
| CN114627122A (en) | Defect detection method and device | |
| CN111178445A (en) | Image processing method and device | |
| CN114518526A (en) | Automatic testing machine control system suitable for PCB board ICT | |
| CN110596118A (en) | Print pattern detection method and print pattern detection device | |
| CN117589770A (en) | PCB patch board detection method, device, equipment and medium | |
| CN116681677A (en) | Lithium battery defect detection method, device and system | |
| CN115343293A (en) | Negative plate detection control system and production line based on machine vision | |
| CN116958035A (en) | Product part defect detection method, device, medium and electronic equipment | |
| CN115330713A (en) | Cathode plate detection method based on machine vision, server and readable storage medium | |
| CN113267506A (en) | Wood board AI visual defect detection device, method, equipment and medium | |
| CN105548210A (en) | Detection method for American-standard power source extension wire based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |