A defect detection system, method, apparatus, electronic device, and storage medium

Abstract

This application provides a defect detection system, method, apparatus, electronic device, and storage medium, relating to the field of machine vision technology. In this system, a proximity sensor sends a sensing signal to a control device whenever a product object approaches. Upon receiving the sensing signal, the control device encodes the product object to obtain coded information and sends a trigger signal to a photodetector. The photodetector responds to the trigger signal by controlling a first camera and a second camera to take pictures, obtaining a first image and a second image. The control device receives each first image and each second image. For each coded information, according to a predetermined image selection method, it selects the first image and the second image containing the product object indicated by the coded information, and performs product object defect detection processing on the selected first and second images to determine the product object defect detection result indicated by the coded information. Therefore, the above system can improve the accuracy of product object defect detection results.

Description

Technical Field

[0001] This application relates to the field of machine vision technology, and in particular to a defect detection system, method, apparatus, electronic device, and storage medium. Background Technology

[0002] PVC (polyvinyl chloride) gloves are in high demand in the medical, industrial, and food sectors. While meeting market demand, the quality of PVC gloves is crucial for safe production in these industries. Therefore, PVC glove manufacturers are focusing on both efficient production methods and quality control. To achieve high-quality, high-efficiency production, manufacturers are increasingly opting for machine vision inspection solutions to minimize downtime, ensure consistently safe and high-quality products, and easily track product processes throughout the supply chain.

[0003] In related technologies, a single-sided visual inspection method is used to detect defects in PVC gloves. Specifically, PVC gloves hanging naturally on the production line pass through the front side of a background plate used to space multiple rows of PVC gloves under the action of a hinge, and enter the visual inspection area before passing through the background plate. The image acquisition mechanism acquires an image of one side (front or back) of a row of PVC gloves that has entered the visual inspection area, and reports the acquired glove image to the control device. The control device then detects the defects in the gloves based on the glove image.

[0004] However, since defects may exist on both sides of PVC gloves, and the relevant technology can only capture images of one side of a row of gloves on an assembly line, the image acquisition mechanism in the relevant technology has blind spots, resulting in incomplete defect detection results and ultimately low accuracy of the defect detection results.

[0005] It is evident that improving the accuracy of defect detection is an urgent problem to be solved for products on production lines that require multifaceted inspection, such as PVC gloves. Summary of the Invention

[0006] The purpose of this application is to provide a defect detection system, method, apparatus, electronic device, and storage medium to improve the accuracy of defect detection results for product objects. The specific technical solution is as follows:

[0007] In a first aspect, in order to achieve the above objectives, this application discloses a defect detection system, including: a control device, an optical detection device, and a proximity sensor; wherein, the optical detection device is provided with a first camera and a second camera arranged in a staggered manner, the first camera being a camera for taking pictures of one side of each product object to be inspected being conveyed on the production line, and the second camera being a camera for taking pictures of the other side of each product object.

[0008] The proximity sensor is installed on the path of the production line and is used to send a sensing signal to the control device whenever a product object approaches.

[0009] The control device is used to encode the currently approaching product object whenever it receives the sensing signal, obtain the encoding information of the currently approaching product object, and send a photo trigger signal to the optical detection device.

[0010] The optical detection device is used to respond to the received photo-taking trigger signal, control the first camera and the second camera to take pictures, obtain the first image and the second image, and report the first image and the second image to the control device.

[0011] The control device is further configured to receive each first image and each second image; and, for each encoded information, to select a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and to perform product object defect detection processing on the selected first image and second image, and to determine the defect detection result of the product object indicated by the encoded information based on the result obtained from the product object defect detection processing.

[0012] The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

[0013] Optionally, for each piece of encoded information, the control device selects a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, including:

[0014] For each piece of encoded information, a first image containing the product object indicated by the encoded information is determined based on the order corresponding to the encoded information, the order corresponding to each first image, and the first positional difference; and a second image containing the product object indicated by the encoded information is determined based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference.

[0015] The order of the encoded information is the order in which the encoded information is sorted according to the generation time; the order of each first image is the order in which the first image is sorted according to the shooting time; and the order of each second image is the order in which the second image is sorted according to the shooting time.

[0016] Optionally, the first position difference is: when a product object to be photographed enters the field of view of the first camera and there is a product object approaching the proximity sensor, there is an interval of N1 product objects between the product object entering the field of view of the first camera and the object approaching the proximity sensor.

[0017] The second positional difference is as follows: when a product object to be photographed enters the field of view of the second camera and there is a product object approaching the proximity sensor, there is an interval of N2 product objects between the product object entering the field of view of the second camera and the object approaching the proximity sensor; where N1 and N2 are different values.

[0018] The control device determines the first image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each first image, and the first position difference, including:

[0019] From each first image, select the first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1;

[0020] The control device determines a second image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference, including:

[0021] From each of the second images, select the second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1.

[0022] Optionally, the control device is further configured to: starting from the first first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoding value of each first image; and starting from the first second image captured by the second camera, encode each second image in sequence according to the shooting time to obtain the encoding value of each second image.

[0023] The control device selects a first image from each first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, including:

[0024] From each first image, determine the first image whose encoded value meets the first condition, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1; wherein, the first condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N1+1.

[0025] The control device selects a second image from each second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, including:

[0026] From each second image, determine the second image whose encoded value meets the second condition, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1; wherein, the second condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N2+1.

[0027] Optionally, the control device is further configured to: starting from the next image after the N1+1th first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoding value of each first image; and starting from the next image after the N2+1th second image captured by the second camera, encode each second image in sequence according to the shooting time to obtain the encoding value of each second image.

[0028] The control device selects a first image from each first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, including:

[0029] From each first image, determine the first image whose encoded value matches the order corresponding to the encoded information, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1.

[0030] The control device selects a second image from each second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, including:

[0031] From each second image, determine the second image whose encoded value matches the order corresponding to the encoded information, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1.

[0032] Optionally, the system further includes a rejection device, wherein the rejection device is positioned on the path of the production line after the optical inspection device;

[0033] The control device is also used to control the rejection device to remove the product object indicated by the coded information from the production line when the defect detection result of the product object indicated by any coded information is detected and meets the rejection conditions.

[0034] The rejection device is used to reject the object represented by the encoded information from the production line under the control of the control device.

[0035] Optionally, the positional difference between the proximity sensor and the rejection device is as follows: when a product object enters the operable area of ​​the rejection device and a product object approaches the proximity sensor, there is an interval of N3 product objects between the product object entering the operable area and the product object approaching the proximity sensor.

[0036] The control device controls the rejection device to reject the product object indicated by the coded information from the production line, including:

[0037] After receiving the specified sensing signal, when the sensing signal is received for the N3+1th time, an object rejection instruction is sent to the rejection device so that the rejection device rejects the product object indicated by the coded information from the production line.

[0038] The specified sensing signal is the sensing signal emitted when the product object represented by the encoded information approaches.

[0039] Optionally, each product object is associated with a product mold when it is transferred on the production line; the optical inspection device is also equipped with a third camera and a fourth camera arranged in a staggered manner, the third camera is a camera that takes pictures of one side of each product mold transferred on the production line, and the fourth camera is a camera that takes pictures of the other side of each product mold.

[0040] The optical detection device is also used to control the third camera and the fourth camera to take pictures in response to the received picture trigger signal, to obtain the third image and the fourth image, and to report the third image and the fourth image to the control device;

[0041] The control device is further configured to receive each third image and each fourth image; and, for each encoded information, determine a third image that matches the shooting time of the selected first image and a fourth image that matches the shooting time of the selected second image, perform product mold defect detection processing on the selected third image and fourth image, determine a mold detection result based on the result of the product mold defect detection processing, and associate the obtained mold detection result with the defect detection result of the product object indicated by the encoded information.

[0042] Optionally, the support for transferring product objects on the production line is also provided with a limiter.

[0043] Optionally, the control device includes: a main control device and a programmable logic controller (PLC);

[0044] The PLC is used to encode the currently approaching product object whenever it receives the sensing signal, send the encoded information to the main control device, and send a photo trigger signal to the optical detection device.

[0045] The main control device is used to receive each first image and each second image; and, for each encoded information, according to a predetermined image selection method, select a first image and a second image containing the product object indicated by the encoded information, and perform product object defect detection processing on the selected first image and second image, and determine the defect detection result of the product object indicated by the encoded information based on the result obtained from the product object defect detection processing.

[0046] The optical detection device is specifically used to report the first image and the second image to the main control device.

[0047] Secondly, in order to achieve the above objectives, embodiments of this application disclose a defect detection method based on a defect detection system, applied to a control device; the method includes:

[0048] Whenever a sensing signal is received from a proximity sensor, the currently approaching product object is encoded to obtain the encoded information of the currently approaching product object, and a photo-taking trigger signal is sent to the optical detection device, so that the optical detection device responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtains the first image and the second image, and reports the first image and the second image to the control device; wherein, the proximity sensor sends a sensing signal to the control device whenever a product object approaches;

[0049] Receive each first image and each second image;

[0050] For each encoded information, a first image and a second image containing the product object indicated by the encoded information are selected according to a predetermined image selection method. The selected first image and second image are then subjected to product object defect detection processing. Based on the result obtained from the product object defect detection processing, the defect detection result of the product object indicated by the encoded information is determined.

[0051] The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

[0052] Thirdly, in order to achieve the above objectives, embodiments of this application disclose a defect detection device based on a defect detection system, applied to a control device; the defect detection device includes:

[0053] The transmitting module is used to encode the currently approaching product object whenever it receives a sensing signal from the proximity sensor, obtain the encoding information of the currently approaching product object, and send a photo-taking trigger signal to the photodetector so that the photodetector responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtain a first image and a second image, and reports the first image and the second image to the control device; wherein, the proximity sensor sends a sensing signal to the control device whenever a product object approaches;

[0054] A receiving module is used to receive each first image and each second image;

[0055] The detection module is used to select a first image and a second image containing the product object indicated by the coded information according to a predetermined image selection method for each coded information, and to perform product object defect detection processing on the selected first image and second image. Based on the result of the product object defect detection processing, the module determines the defect detection result of the product object indicated by the coded information. The image selection method is a selection method set based on the first position difference between the proximity sensor and the first camera and the second position difference between the proximity sensor and the second camera.

[0056] This application also provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0057] Memory, used to store computer programs;

[0058] When a processor executes a program stored in memory, it implements the steps of any of the defect detection methods described above.

[0059] This application also provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of any of the defect detection methods described above.

[0060] Beneficial effects of the embodiments in this application:

[0061] This application provides a defect detection system, which includes a control device, an optical inspection device, and a proximity sensor. Whenever a product object approaches, the proximity sensor sends a sensing signal to the control device. Whenever the control device receives a sensing signal, it encodes the currently approaching product object to obtain encoded information, and then sends a photo-taking trigger signal to the optical inspection device. The optical inspection device responds to the trigger signal by controlling a first camera and a second camera to take pictures, obtaining a first image and a second image. The control device acquires each first image and each second image, and for each encoded information, selects the corresponding first image and second image of the product object under a predetermined image selection method. The control device then performs product object defect detection processing on the images.

[0062] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0063] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0064] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0065] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0067] Figure 1 This is a schematic diagram of the structure of a defect detection system provided in an embodiment of this application;

[0068] Figure 2 This is a schematic diagram of another defect detection system provided in an embodiment of this application;

[0069] Figure 3(a) is a schematic diagram of an optical inspection device in a defect detection system provided in an embodiment of this application;

[0070] Figures 3(b), 3(c), 3(d) and 3(e) are schematic diagrams of various views of an optical inspection device in a defect detection system provided in an embodiment of this application;

[0071] Figure 4 This is a schematic diagram of another defect detection system provided in an embodiment of this application;

[0072] Figure 5 This is a schematic diagram of another defect detection system provided in an embodiment of this application;

[0073] Figure 6 A front view of the overall equipment of a defect detection system provided in an embodiment of this application;

[0074] Figure 7 A top view of the overall equipment of a defect detection system provided in an embodiment of this application;

[0075] Figure 8 A schematic flowchart illustrating a defect detection method for a control device provided in an embodiment of this application;

[0076] Figure 9 A flowchart illustrating a defect detection process provided in an embodiment of this application;

[0077] Figure 10 This is a schematic diagram of a defect detection device provided in an embodiment of this application;

[0078] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0080] To facilitate understanding of the solution, the technical terms used in the embodiments of this application will be introduced first:

[0081] Machine vision online inspection system: A machine vision online inspection system is an inspection device integrated into the production process. It is an industrial application system that integrates optical, mechanical, electronic, computing, and software technologies. By detecting and sensing the spatiotemporal patterns of electromagnetic radiation, it can automatically acquire one or more images of the target object, process and analyze various features of the acquired images, make qualitative and quantitative interpretations based on the analysis results to obtain an understanding of the target object, and make corresponding decisions. It is a process control system that makes decisions and executes practical actions.

[0082] Industrial cameras: a key component in machine vision systems, whose most essential function is to convert light signals into ordered electrical signals.

[0083] Proximity sensors are sensors designed to replace contact-based detection methods such as limit switches, enabling detection without physical contact with the object. Proximity sensors detect the movement and presence of an object and convert this information into an electrical signal. Detection methods that convert this information into an electrical signal include those utilizing eddy currents generated in the metal of the object due to electromagnetic induction, detecting changes in capacitance caused by the proximity of a metal object, and methods using levers and guide switches.

[0084] Optical inspection and rejection machine: A vision system that integrates hand and glove image acquisition, defect detection and rejection mechanisms using optical imaging, software, electrical and other technologies.

[0085] Deep learning image segmentation: a technique and process based on deep learning convolutional neural network models to divide an image into several specific regions with unique properties and extract targets of interest. It is a key step from image processing to image analysis.

[0086] Deep learning object detection: This is based on deep learning convolutional neural network models to identify the category of objects in an image, or to predict the location of objects, or to determine the location and category of multiple objects.

[0087] With the development of market demand, defect detection systems are playing an increasingly important role in industrial production. Among related technologies, there are manual inspection methods and machine vision-based online inspection methods for detecting defects in products.

[0088] Taking PVC gloves as an example, in existing glove production lines, after the gloves are formed, they enter the demolding process on a hand mold. During this process, it is necessary to inspect the surface of the gloves for surface defects such as oil stains, defects, or pinholes.

[0089] Taking defect detection of PVC gloves as an example, the manual inspection methods provided by relevant technologies are as follows:

[0090] During the process of PVC gloves entering the demolding process on the production line, surface defects of the PVC gloves are detected by human eyes. Then, gloves with surface defects are peeled off from the mold manually. This not only requires a lot of manpower and is very labor-intensive, but also has a low detection accuracy. In addition, due to cost constraints, it is impossible to inspect all gloves produced, and there is a certain probability of omission.

[0091] Taking defect detection of PVC gloves as an example, the machine vision online inspection solution provided by related technologies is as follows:

[0092] In related technologies, a single-sided visual inspection method is used to detect defects in PVC gloves. Specifically, PVC gloves hanging naturally on the production line pass through the front side of a background plate used to space multiple rows of PVC gloves under the action of a hinge, and enter the visual inspection area before passing through the background plate. The image acquisition mechanism acquires an image of one side (front or back) of a row of PVC gloves that has entered the visual inspection area, and reports the acquired glove image to the control device. The control device then detects the defects in the gloves based on the glove image.

[0093] However, since PVC gloves may have defects on both sides, and the relevant technology can only capture one side of a row of gloves on the production line, the image acquisition mechanism in the relevant technology has blind spots, resulting in incomplete defect detection results and ultimately low accuracy of the defect detection results.

[0094] As discussed above, the single-sided visual inspection method used in related technologies for detecting defects in PVC gloves suffers from low accuracy. Therefore, improving the accuracy of defect detection for products requiring multi-sided inspection on production lines, such as PVC gloves, is a pressing issue that needs to be addressed.

[0095] To address the issue of low accuracy in defect detection results for products, embodiments of this application provide a defect detection system, method, apparatus, electronic device, and storage medium.

[0096] This application provides a defect detection system, which includes a control device, an optical inspection device, and a proximity sensor. The optical inspection device comprises a first camera and a second camera arranged in a staggered manner. The first camera is used to photograph one side of each product object to be inspected during the assembly line transport, and the second camera is used to photograph the other side of each product object. The system includes:

[0097] The proximity sensor is installed on the path of the production line and is used to send a sensing signal to the control device whenever a product object approaches.

[0098] The control device is used to encode the currently approaching product object whenever it receives the sensing signal, obtain the encoding information of the currently approaching product object, and send a photo trigger signal to the optical detection device.

[0099] The optical detection device is used to respond to the received photo-taking trigger signal, control the first camera and the second camera to take pictures, obtain the first image and the second image, and report the first image and the second image to the control device.

[0100] The control device is further configured to receive each first image and each second image; and, for each encoded information, to select a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and to perform product object defect detection processing on the selected first image and second image to obtain the defect detection result of the product object indicated by the encoded information.

[0101] The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

[0102] It should be noted that the control device can be a separate control unit that communicates with the optical inspection device and the proximity sensor. In this case, the control device can realize the control functions it possesses. For example, the control device can be a computer with visual inspection software deployed on it. Of course, the functions of the control device can also be realized by at least two devices working together. That is to say, the device form of the control device can be a combination of at least two devices, which is also reasonable. The visual inspection software can encode the sensing signal emitted by the proximity sensor, emit a photo-taking trigger signal, receive images sent by the optical inspection device, select images containing product objects, perform product object defect detection processing on the images, and make corresponding decisions.

[0103] The camera installed in the optical inspection device can be an industrial camera. In this application, the optical inspection device can utilize the essential function of an industrial camera: converting light signals into ordered electrical signals, so as to perform a picture-taking operation after receiving a picture-taking trigger signal, obtain an image, and then report the image to the control device.

[0104] Proximity sensors are sensors designed to replace contact-based detection methods such as limit switches, allowing for detection without physical contact with the object being detected.

[0105] In addition, the product object can be any product on the production line that requires image detection to obtain defect detection results. For example, the product object can be PVC gloves.

[0106] It should be noted that the control device, optical inspection device and proximity sensor in the embodiments of this application can jointly form a complete machine vision online inspection system to realize defect detection and processing of products in the production process and make corresponding decisions based on the defect detection results.

[0107] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0108] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0109] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0110] The following describes a defect detection system provided in an embodiment of this application, with reference to the accompanying drawings.

[0111] Figure 1 This is a schematic diagram of a defect detection system provided in an embodiment of this application. The system includes: a proximity sensor 110, a control device 120, and a photodetector 130. The proximity sensor 110 is connected to the control device 120, and the control device 120 is connected to the photodetector 130. The photodetector 130 is equipped with a first camera 1301 and a second camera 1302 arranged in a staggered manner. The first camera 1301 is a camera that takes pictures of one side of each product object to be inspected during the production line transport, and the second camera 1302 is a camera that takes pictures of the other side of each product object.

[0112] The proximity sensor 110 is disposed on the path of the production line and is used to send a sensing signal to the control device 120 whenever a product object approaches.

[0113] The control device 120 is used to encode the currently approaching product object whenever it receives the sensing signal, determine the encoding information of the currently approaching product object based on the result of the defect detection processing of the product object, and send a photo trigger signal to the optical detection device 130.

[0114] The optical detection device 130 is used to control the first camera 1301 and the second camera 1302 to take pictures in response to the received picture-taking trigger signal, to obtain a first image and a second image, and to report the first image and the second image to the control device 120.

[0115] The control device 120 is further configured to receive each first image and each second image; and, for each encoded information, select a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and perform product object defect detection processing on the selected first image and second image to obtain the defect detection result of the product object indicated by the encoded information.

[0116] The image selection method is a selection method set based on the first positional difference between the proximity sensor 110 and the first camera 1301 and the second positional difference between the proximity sensor 110 and the second camera 1302.

[0117] It should be noted that the defect detection system provided in this application is an inspection device integrated into the production process. After the production process, rows of products on the assembly line await entry into the inspection area of ​​the defect detection system. For example, if there are two rows of PVC gloves hanging naturally on the assembly line that need to be transferred, each row of PVC gloves will enter the inspection area of ​​the defect detection system under the action of a hinge. Furthermore, since at least one row of products can be transferred on the assembly line, the optical inspection device can be equipped with a first camera and a second camera arranged in a staggered manner for each row of products. That is, the first camera and the second camera can capture images of multiple sides of each product in the same row.

[0118] Furthermore, along the product transport direction of the production line, the optical inspection device 130 can be placed after the proximity sensor. That is, any product object first passes the proximity sensor, and then, after a period of transport, passes through the image acquisition area of ​​the optical inspection device 130. At this time, the product object captured by the camera of the optical inspection device 130 is different from the product object sensed by the proximity sensor 110. Alternatively, the optical inspection device 130 can be placed in a position that satisfies the following condition: when any product object approaches the proximity sensor 110, it simultaneously enters the image acquisition area of ​​the optical inspection device. In this case, the product object approaching the proximity sensor and the product object captured by the target camera in the optical inspection device belong to the same product object; where the target camera is the camera closer to the proximity sensor 110 among the first and second cameras. For ease of description, the following description uses the first camera as the camera closer to the proximity sensor 110.

[0119] Regarding the proximity sensor 110, since it is installed along the path on the production line, it ensures that whenever a product object approaches, the proximity sensor 110 can detect it and send a sensing signal to the control device 120, providing the control device 120 with the conditions to encode the product object. It is understood that the proximity sensor 110 can generate an alternating magnetic field internally. When a product object suspended on the hinge of the production line approaches as the production line moves, the hinge will undergo electromagnetic induction with the alternating magnetic field generated by the vibrator, producing an induced current. As the product object passes the proximity sensor 110, the proximity sensor 110 can convert the position and presence information of the product object into a sensing signal without contacting it, and then send the sensing signal to the control device 120. The type of proximity sensor 110 can be selected according to the needs of the production site; this application does not limit the specific type of proximity sensor.

[0120] For the control device 120, whenever the proximity sensor 110 sends a sensing signal, the control device 120 can receive the sensing signal and encode the currently approaching product object, thereby assigning identification information to the product object at each position that passes the proximity sensor. Through the encoding information, product objects at different positions on the production line can be distinguished, ultimately providing the conditions for the association of multi-face detection of the same product object.

[0121] Furthermore, since defect detection requires image analysis, the control device 120 can send a photo-taking trigger signal to the optical inspection device 130 whenever it receives a sensing signal, thereby triggering the first and second cameras in the optical inspection device to acquire images. In other words, the optical inspection device 130 is triggered to acquire images every time a product object approaches.

[0122] Regarding the optical inspection device 130, the optical inspection device can also be called an optical inspection machine; whenever the optical inspection device 130 receives a photo-taking trigger signal from the control device 120, it can control the first camera 1301 and the second camera 1302 to take pictures, obtain a first image and a second image, and upload the first image and the second image to the control device 120. It is understandable that, since the first camera 1301 and the second camera 1302 are arranged in a staggered manner, and the first camera 1301 and the second camera 1302 are simultaneously photographed under the same photographing trigger signal, the resulting images cannot correspond to the same product object. However, precisely because the first camera 1301 and the second camera 1302 are arranged in a staggered manner, and the first camera and the second camera are used to photograph different sides of the object in the same row, the optical detection device 130 can trigger the first camera 1301 to photograph one side of the product object that is not obstructed, which is the first image, under the action of the current photographing trigger signal. When another photographing trigger signal arrives, it can trigger the second camera 1302 to photograph the other side of the product object that is not obstructed, which is the second image. For example, if the first camera and the second camera collect images of adjacent product objects, then the other photographing trigger signal is the next photographing trigger signal after the current photographing trigger signal.

[0123] Understandably, the first image can be the front of the product being photographed, and correspondingly, the second image can be the back of the product being photographed; conversely, the first image can also be the back of the product being photographed, and correspondingly, the second image can be the front of the product being photographed. For example, in a glove production line, the target glove has entered the optical inspection device. In response to a first trigger signal, the optical inspection device controls the first and second cameras to take pictures. The first camera can capture the front of the target glove, and the second camera can capture an image that does not include the target glove. As the production line continues to run, the optical inspection device, in response to a second trigger signal, controls the first and second cameras to take pictures. The first camera can capture an image that does not include the target glove, and the second camera can capture the back of the target glove.

[0124] Furthermore, it is understood that in the production environment, the first positional difference between proximity sensor 110 and the first camera 1301, and the second positional difference between proximity sensor 110 and the second camera 1302, are regular characteristics. Based on these regular characteristics, each product object can be associated with a first image and a second image containing that product object. Specifically, the control device 120 can identify each product object that sequentially approaches proximity sensor 110 on the production line through encoded information. By combining the already determined first and second positional differences, it can select the first and second images corresponding to each product object. That is, according to a predetermined image selection method, it selects the first and second images of the product object indicated by each encoded information in the report. For example, the proximity sensor, the first camera, and the second camera are arranged closely together with no gaps between them, meaning there are no product objects present. The first, second, and third product objects approach the proximity sensor sequentially in chronological order. When the first product object approaches the proximity sensor, the first and second cameras capture images that do not include the first product object. When the second product object approaches the proximity sensor, the first camera captures a frontal image of the first product object, and the second camera captures an image that does not include any product objects. When the third product object approaches the proximity sensor, the first camera captures a frontal image of the second product object, and the second camera captures a rear image of the first product object. Thus, the control device can select the frontal and rear images of the first product object based on positional differences and encoded information. For example, a product object can enter the image acquisition area of ​​the first camera while approaching the proximity sensor. The first camera and the second camera photograph adjacent product objects. In this case, when the first object approaches the proximity sensor, the proximity sensor sends a signal that triggers the control device to issue a photo-taking trigger signal. The first camera can then photograph the first object that is approaching the proximity sensor, but the second camera cannot photograph it. When the next object approaches, the second camera can then photograph the image of the product object that previously passed the proximity sensor. At this time, the first image photographed by the first camera and the second image photographed by the second camera are used as the two images of the first object to be associated.

[0125] After selecting the first and second images of the corresponding product object, the control device 120 can perform product object defect detection processing on the images, thereby obtaining the defect detection result of the corresponding product object. For example, the control device can use a pre-trained detection model for product object defect detection, and then input the first and second images into the detection model to perform product object defect detection. The defect detection model can be established using deep learning segmentation and object detection techniques. For instance, the product object defect detection model can employ a scheme combining deep learning and traditional 2D algorithms, incorporating defect types as determined by the human eye into the network for learning, thereby training the model. This application does not limit the specific defect detection process.

[0126] There are multiple ways to determine the defect detection result of the product object indicated by the encoded information based on the result of the product object defect detection processing. For example, after performing product object defect detection processing on the first image and the second image, the control device 120 can obtain the defect detection results of the first image and the second image respectively. When both the first image and the second image indicate that the product object is defective, that is, under the AND-NOT logical relationship, the result that the corresponding product object is defective can be obtained, i.e., the defect detection result of the product object indicated by the encoded information is defective; or, when at least one of the defect detection results of the first image or the second image indicates that the product object is defective, that is, under the OR-NOT logical relationship, the result that the corresponding product object is defective can be obtained, i.e., the defect detection result of the product object indicated by the encoded information is defective; or, when both the first image and the second image indicate that the product object is not defective, that is, under the NOT-NOT logical relationship, the result that the corresponding product object is not defective can be obtained, i.e., the defect detection result of the product object indicated by the encoded information is not defective.

[0127] Furthermore, the specific content of the defect detection result of the first image and the specific content of the defect result of the second image can be text content indicating whether there is a defect, such as "defective" or "no defect". In this case, if the defect detection result is "defective", it indicates that one side of the product object is defective, and if the defect detection result is "no defect", it indicates that one side of the product object is not defective. Of course, the specific content of the defect detection result of the first image and the specific content of the defect result of the second image can be defect category information. The defect category information can include category information indicating no defect, and category information indicating the specific defect category to which the defect belongs when a defect exists. In this case, if the defect detection result is the category information of the specific defect category, it indicates that one side of the product object is defective, and if the defect detection result is the category information indicating no defect, it indicates that one side of the product object is not defective. It should be noted that when the specific content of the defect detection result of the first image and the specific content of the defect result of the second image can be defect category information, the defect detection result of the product object indicated by the encoded information can also represent the specific category information of the defect.

[0128] Optionally, the optical inspection device 130 can be controlled by a strobe light source, which increases the brightness range of the image, greatly improves the image quality, makes the image closer to human eye perception, and is compatible with high-speed production line cycles, while also increasing the service life of the hardware.

[0129] In addition, limiters can be installed on the supports used to transfer product objects on the production line.

[0130] For limit switches, whenever a product object approaches the proximity sensor, the limit switch can cause the production line to stop smoothly and briefly. At this time, the optical detection device can control the camera to take a picture and obtain a clear image.

[0131] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0132] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0133] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0134] Optionally, in another embodiment, such as Figure 2 As shown, in Figure 1 Based on the defect detection system shown, each product object is associated with a product mold when it is transferred on the production line; the optical inspection device 130 is also equipped with a third camera 1303 and a fourth camera 1304 arranged in a staggered manner. The third camera 1303 is a camera that takes pictures of one side of each product mold transferred on the production line, and the fourth camera 1304 is a camera that takes pictures of the other side of each product mold.

[0135] The optical detection device 130 is also used to control the third camera 1303 and the fourth camera 1304 to take pictures in response to the received picture-taking trigger signal, to obtain a third image and a fourth image, and to report the third image and the fourth image to the control device 120.

[0136] The control device 120 is further configured to receive each third image and each fourth image; and, for each encoded information, determine a third image that matches the shooting time of the selected first image and a fourth image that matches the shooting time of the selected second image, perform product mold defect detection processing on the selected third image and fourth image, determine a mold detection result based on the result obtained from the product mold defect detection processing, and associate the obtained mold detection result with the defect detection result of the product object indicated by the encoded information.

[0137] Since product molds are typically located above the product object (e.g., a glove is attached to the bottom of a hand mold), the third camera can be positioned above the first camera, and the fourth camera above the second camera. This allows the product mold of the passing product object to simultaneously enter the image acquisition area of ​​the third camera when the first camera takes a picture of it. Similarly, when the second camera takes a picture of the passing product object, the associated product mold can simultaneously enter the image acquisition area of ​​the fourth camera. Furthermore, since at least one row of product objects can be transported on the production line, at least one row of product molds can also exist. Therefore, the optical inspection device can have staggered first and second cameras for each row of product objects, and staggered third and fourth cameras for each row of product molds. In other words, the first and second cameras capture images of multiple faces of each product object in the same row, while the third and fourth cameras capture images of multiple faces of each product mold in the same row. For example, in a specific application, a schematic diagram of an optical inspection device including a first camera 1301, a second camera 1302, a third camera 1303, and a fourth camera 1304 can be shown in Figure 3(a); wherein, the area within the dashed box is the shooting range of the camera lens optical path, and the first camera 1301, the second camera 1302, the third camera 1303, and the fourth camera 1304 are mounted on the main body 1305 of the optical inspection device; to show the optical inspection device more clearly, Figures 3(b)-3(e) show engineering schematic diagrams of various views of the optical inspection device.

[0138] It should be noted that the functions of the third camera 1303 and the fourth camera 1304 in the optical inspection device 130 are basically the same as those of the first camera 1301 and the second camera 1302. The difference is that the objects photographed by the third camera 1303 and the fourth camera 1304 can be product molds associated with the product object. Correspondingly, the control device 120 can determine that the image whose shooting time matches the selected first image is the third image, and determine that the image whose shooting time matches the selected second image is the fourth image. For example, in a glove production line, if a specific glove is selected and the first image corresponding to the specific glove has been determined, then the third image captured by the third camera, which was captured simultaneously by the first camera that captured the first image, can be selected. This third image is the image of the specific hand mold associated with the specific glove. Similarly, a fourth image containing the specific hand mold associated with the specific glove can be selected.

[0139] It is understood that when the control device 120 performs defect detection processing on the determined third and fourth images, the defect detection processing method is the same as the defect detection processing method performed by the control device 120 on the determined first and second images, and will not be described in detail in this embodiment.

[0140] It should be noted that there can be multiple ways to determine the mold inspection result based on the results obtained from the product mold defect detection processing. Similarly, there can also be multiple ways to determine the defect detection result of the product object indicated by the encoded information based on the results obtained from the product object defect detection processing. It is understood that, in the case of AND, OR, and NOT logical relationships, the control device 120 can determine the product object defect detection result based on the defect detection results of the first and second images; the specific determination method has been described in the previous embodiment. Correspondingly, in the case of AND, OR, and NOT logical relationships, the control device 120 can also determine the product mold defect detection result based on the defect detection results of the third and fourth images. For example, after performing product mold defect detection processing on the third image and the fourth image, the control device 120 can obtain the defect detection results of the third image and the fourth image respectively. When the defect detection results of the third image and the fourth image both indicate that the product mold is defective, that is, under the logical relationship of AND, the result that the corresponding product mold is defective can be obtained, that is, the defect detection result of the product mold indicated by the coding information is defective; or, when at least one of the defect detection results of the third image or the fourth image indicates that the product mold is defective, that is, under the logical relationship of OR, the result that the corresponding product mold is defective can be obtained, that is, the defect detection result of the product mold indicated by the coding information is defective; or, when the defect detection results of the third image and the fourth image both indicate that the product mold is not defective, that is, under the logical relationship of NOT, the result that the corresponding product mold is not defective can be obtained, that is, the defect detection result of the product mold indicated by the coding information is not defective.

[0141] Furthermore, the specific content of the defect detection results in the third image and the fourth image can be textual content indicating whether a defect exists, such as "defective" or "no defect." In this case, if the defect detection result is "defective," it indicates that one side of the product mold is defective, while if the defect detection result is "no defect," it indicates that one side of the product mold is not defective. Alternatively, the specific content of the defect detection results in the third image and the fourth image can be defect category information. This defect category information can include category information indicating the absence of a defect, and category information indicating the specific defect category to which the defect belongs when it exists. In this case, if the defect detection result is the category information of a specific defect category, it indicates that one side of the product mold is defective, while if the defect detection result is the category information indicating the absence of a defect, it indicates that one side of the product mold is not defective.

[0142] Understandably, after obtaining the defect detection results of the product mold, the control device 120 can correlate the mold defect detection results with the defect detection results of the product object. For example, in a glove production line, if the defect detection results of the target hand mold show that the hand mold has a crack, and the defect detection results of the target glove hanging below the target hand mold show that the glove has a tear or crack, then the system can correlate the tear or crack defect of the glove with the crack defect of the hand mold. As another example, in a glove production line, if the defect detection results of the target hand mold show that the hand mold has a dirt defect, and the defect detection results of the target glove hanging below the target hand mold show that the glove has an oil stain or flow mark defect, then the system can correlate the oil stain or flow mark defect of the glove with the dirt defect of the hand mold.

[0143] In another implementation, the defect detection system can also classify defects in the product object. When the control device detects a defect that seriously affects the quality of the product object, it can issue a notification regarding the product mold defect associated with the product object defect, reminding relevant personnel to perform maintenance and replacement of the product mold. For example, in a glove production line, if the control device detects a tear in the glove, and the associated hand mold defect detection result indicates a crack in the hand mold, the system can send the information about the cracked hand mold defect to relevant personnel to remind them to repair the hand mold.

[0144] In another implementation, the control device can save the data on product object defects and product mold defects after detecting them, and perform statistical analysis. This allows for prediction of product object defects and product mold defects with the support of data.

[0145] In this embodiment of the application, the defect detection system can simultaneously detect defects in both the product object and the product mold, and correlate the defect detection results of the two, which can further provide basic data for the analysis of the causes of defects.

[0146] Optionally, in another embodiment, such as Figure 4 As shown, in Figure 1 Based on the defect detection system shown, the control device 120 includes: a main control device 410 and a programmable logic controller (PLC) 420;

[0147] The PLC 420 is used to encode the currently approaching product object whenever it receives the sensing signal, and send the encoded information to the main control device 410, and send a photo trigger signal to the optical detection device 130.

[0148] The main control device 410 is used to receive each first image and each second image; and, for each encoded information, according to a predetermined image selection method, select a first image and a second image containing the product object indicated by the encoded information, and perform product object defect detection processing on the selected first image and second image, and determine the defect detection result of the product object indicated by the encoded information based on the result obtained from the product object defect detection processing.

[0149] The optical detection device 130 is specifically used to report the first image and the second image to the main control device 410.

[0150] It is understood that in this embodiment, after the control device 120 can be subdivided into a main control device 410 and a programmable logic controller (PLC) 420, the functions are also divided accordingly. The PLC 420 can receive each sensing signal, and each sensing signal represents each approaching product object. Therefore, the PLC can encode the currently approaching product object, and then send the encoded information to the main control device 410, as well as send a photo-taking trigger signal to the optical detection device 130. The main control device 410 can receive each first image and each second image; and for each encoded information, according to a predetermined image selection method, select the first image and the second image containing the product object indicated by the encoded information, and perform product object defect detection processing on the selected first image and the second image to obtain the defect detection result of the product object indicated by the encoded information.

[0151] In this embodiment, the main control device 410 and the PLC 420 are separate devices. The main control device 410 can be any type of electronic device, and this application does not limit it.

[0152] The control device provided in this embodiment consists of a main control device 410 and a PLC 420. Thus, the functions of the control device can be realized through the cooperation of the main control device 410 and the PLC 420. This separation achieves functional decoupling, thereby improving the ease of system maintainability and reducing operating costs.

[0153] Optionally, in another embodiment of this application, the control device 120 selects a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method for each piece of encoded information, which may include step A1:

[0154] Step A1: For each piece of encoded information, based on the order corresponding to the encoded information, the order corresponding to each first image, and the first positional difference, determine a first image containing the product object indicated by the encoded information; and based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference, determine a second image containing the product object indicated by the encoded information.

[0155] The order of the encoded information is the order in which the encoded information is sorted according to the generation time; the order of each first image is the order in which the first image is sorted according to the shooting time; and the order of each second image is the order in which the second image is sorted according to the shooting time.

[0156] In this embodiment, the encoded information can be sorted according to the generation time order to obtain the order corresponding to each encoded information, that is, the arrangement position of each encoded information in the queue formed by the encoded information; and the first images can be sorted according to the shooting time to obtain the order corresponding to each first image, that is, the arrangement position of each first image in the queue formed by the first images; similarly, the first images can be sorted according to the shooting time to obtain the order corresponding to each first image, that is, the arrangement position of each first image in the queue formed by the first images.

[0157] Based on the determination of the above-mentioned arrangement positions, and considering the existence of the first position difference and the second position difference, the encoded information, the first image, and the second image of the same product object will have different positions in their respective queues, and there is a regularity to this. Therefore, for each piece of encoded information, based on the order corresponding to the encoded information, the order corresponding to each first image, and the first position difference, the first image containing the product object indicated by the encoded information can be determined; and based on the order corresponding to the encoded information, the order corresponding to each second image, and the second position difference, the second image containing the product object indicated by the encoded information can be determined. In this way, the first image and the second image of the same product object can be obtained quickly.

[0158] It should be noted that the order obtained by sorting the encoded information according to the generation time can be a queue with a sequential order. The time interval between each encoded information in the queue can depend on the running speed of the production line. Correspondingly, the time interval between each first image and each second image can be consistent with the time interval between encoded information.

[0159] Furthermore, the sorted order of the encoded information can form an encoded information queue, the sorted order of the first images can form a first image queue, and the sorted order of the second images can form a second image queue. The time intervals between elements in these three queues can be consistent, and the position of each element in the queue can be known. When a product object approaches the proximity sensor, the first camera and the second camera can take pictures. Correspondingly, the encoded information queue can have one more encoded information element about the product object, the first image queue can have one more image element captured by the first camera, and the second image queue can have one more image element captured by the second camera. When the positional differences between the proximity sensor, the first camera, and the second camera are known—that is, when the number of elements differing between the proximity sensor, the first camera, and the second camera—it is possible to determine the first image and the second image containing the product object indicated by the corresponding encoded information.

[0160] Optionally, in one implementation, the first position difference can be: when a product object to be photographed enters the field of view of the first camera and there is a product object approaching the proximity sensor, there is an interval of N1 product objects between the product object entering the field of view of the first camera and the object approaching the proximity sensor.

[0161] The second positional difference can be: when a product object to be photographed enters the field of view of the second camera and there is a product object approaching the proximity sensor, there is an interval of N2 product objects between the product object entering the field of view of the second camera and the object approaching the proximity sensor; where N1 and N2 are different values.

[0162] Accordingly, the control device determines the first image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each first image, and the first positional difference, including:

[0163] From each first image, select the first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1;

[0164] The control device determines a second image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference, including:

[0165] From each of the second images, select the second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1. In this implementation, when describing positional differences, the number of product objects between the product object to be photographed and the new product object approaching the proximity sensor can be used as the distance unit when there is a product object to be photographed in the camera's field of view and a new product object approaches the proximity sensor. For example, in a glove production line, if there are 5 gloves between the first camera and the proximity sensor, and glove number 0 enters the field of view of the first camera, while a new glove approaches the proximity sensor at the same time, the new glove can be recorded as glove number 6. Here, N1 and N2 can be non-zero natural numbers; in particular, when the proximity sensor and the first camera are closely connected, N1 can be zero. In this implementation, the first image can be selected from a queue of first images arranged in the order they were captured. It's understood that the first image in the queue has a lower order. Therefore, the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1 is selected. That is, the image captured by the first camera after N1 product objects have been processed on the production line from the proximity sensor, i.e., the N1+1th image. For example, if the proximity sensor and the first camera are separated by 5 product objects, for any product object, if the encoded information corresponding to that product object has an order of 1, the system will select the image with the order 6 in the first image queue to obtain the first image of that product object.

[0166] The selection rules for the second image are the same as those for the first image. The difference is that the interval N1 between the proximity sensor and the first camera is different from the interval N2 between the proximity sensor and the second camera.

[0167] Based on the above description of the first position difference and the second position difference, when the encoding methods of the control device 120 for the first image and the second image are different, the specific implementation methods of the above steps B1 and B2 are different.

[0168] For example, in one implementation, the control device 120 may also be used to: starting from the first first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoding value of each first image; and starting from the first second image captured by the second camera, encode each second image in sequence according to the shooting time to obtain the encoding value of each second image.

[0169] Step B1, the control device 120 selects a first image from each of the first images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, which may include step B11:

[0170] B11, from each of the first images, determine the first image whose encoded value meets the first condition, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1; wherein, the first condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N1+1.

[0171] Step B2, the control device 120 selects a second image from each of the second images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, which may include step B21:

[0172] B21, from each second image, determine the second image whose encoded value meets the second condition, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1; wherein, the second condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N2+1.

[0173] The encoded values ​​of the first and second images represent the order of representation.

[0174] It is understandable that, starting with the first image captured by the first camera, each first image is encoded sequentially according to the shooting time. In this case, some of the first images captured by the first camera may not contain the product object. It should be noted that the method for selecting the first and second images corresponding to the product object can be the same as steps B1 and B2.

[0175] For example, in a glove production line, there is no gap between the proximity sensor, the first camera, and the second camera. The production line speed remains constant, the frequency at which each glove approaches the proximity sensor remains constant, and the image capture frequency of the first and second cameras remains constant. The positions of each glove at each image capture are shown in Table 1. Gloves 1, 2, and 3 are the gloves that approach the proximity sensor in chronological order. The first images captured by the first camera can be encoded as image 1, image 2, and image 3 in chronological order. Similarly, the second images captured by the second camera can also be encoded as image 1, image 2, image 3, and image 4 in chronological order. Image 1 and image 2; when glove 1 approaches the proximity sensor, the first camera captures image 1 without the glove image, and the second camera captures image 1 without the glove image; when glove 2 approaches the proximity sensor, the first camera captures image 2 containing the glove image, and the second camera captures image 2 without the glove image; when glove 3 approaches the proximity sensor, the first camera captures image 3 containing the glove image, and the second camera captures image 3 containing the glove image; at this time, the control device can select image 2 captured by the first camera and image 3 captured by the second camera as the image of glove 1.

[0176] proximity sensor First Camera Second camera First time taking photos Glove No. 1 No gloves No gloves Second photo Size 2 gloves Glove No. 1 No gloves Third photo Glove No. 3 Size 2 gloves Glove No. 1

[0177] Table 1

[0178] For example, in another implementation, the control device 120 may also be used to: starting from the next image after the N1+1th first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoded value of each first image; and starting from the next image after the N2+1th second image captured by the second camera, encode each second image in sequence according to the shooting time to obtain the encoded value of each second image.

[0179] Step B1, the control device 120 selects a first image from each of the first images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, which may include step B12:

[0180] B12, from each first image, determine the first image whose encoded value matches the order corresponding to the encoded information, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1.

[0181] Step B2, where the control device 120 selects a second image from each of the second images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, may include step B22:

[0182] B22, from each second image, determine the second image whose encoded value matches the order corresponding to the encoded information, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1.

[0183] It is understandable that, starting from the next image after the (N1+1)th first image captured by the first camera, each first image is encoded sequentially according to the capture time. In this case, each first image and each second image captured by the first and second cameras can be an image containing a product object. It should be noted that the method for corresponding the first and second images of the product object can be consistent with steps B1 and B2.

[0184] For example, in a glove production line, there is a gap of one glove between the proximity sensor and the first camera; and a gap of two gloves between the proximity sensor and the second camera. The production line speed remains constant, the frequency at which each glove approaches the proximity sensor remains constant, and the image capture frequency of the first and second cameras remains constant. The positions of each glove at each image capture are shown in Table 2. Gloves 1, 2, 3, 4, and 5 are gloves that approach the proximity sensor in chronological order. Starting from the third image captured by the first camera, each first image captured by the first camera is encoded as image 3, image 4, and image 5 according to the capture time sequence. Similarly, starting from the fourth image captured by the first camera, the second camera... The captured second images are coded as image 4, image 5, and image 6 according to the shooting time sequence. When glove 3 is close to the proximity sensor, the first camera captures image 3, which includes the image of glove 2, and the second camera captures image 4, which includes the image of glove 1. When glove 4 is close to the proximity sensor, the first camera captures image 4, which includes the image of glove 3, and the second camera captures image 5, which includes the image of glove 2. When glove 5 is close to the proximity sensor, the first camera captures image 5, which includes the image of glove 4, and the second camera captures image 6, which includes the image of glove 3. At this time, the control device can select image 4 captured by the first camera and image 6 captured by the second camera as the image of glove 3.

[0185] proximity sensor First Camera Second camera First time taking photos Glove No. 3 Size 2 gloves Glove No. 1 Second photo Size 4 gloves Size 3 gloves Size 2 gloves Third photo Size 5 gloves Size 4 gloves Size 3 gloves

[0186] Table 2

[0187] In this embodiment, the control device can encode images that contain or do not contain product objects to obtain an image order. This allows the device to select images of the corresponding product objects based on the order of the product object encoding information, the order of each image, and the positional differences of each component in the system. This facilitates product object defect detection and further improves the accuracy of product object defect detection results.

[0188] Optionally, in another embodiment, such as Figure 5 As shown, in Figure 1 Based on the defect detection system shown, it also includes: a rejection device 510, wherein the rejection device is located after the optical inspection device 130 on the path of the production line;

[0189] The control device 120 is also used to control the rejection device 510 to remove the product object indicated by the coding information from the production line when the defect detection result of the product object indicated by any coding information is detected and meets the rejection conditions.

[0190] The rejection device 510 is used to reject the object represented by the encoded information from the production line under the control of the control device 120.

[0191] The rejection device, also known as a rejection machine, is used to reject non-compliant products.

[0192] Optionally, in one implementation, the positional difference between the proximity sensor 110 and the rejection device 510 is as follows: when a product object enters the operable area of ​​the rejection device and a product object approaches the proximity sensor, there is an interval of N3 product objects between the product object entering the operable area and the product object approaching the proximity sensor.

[0193] The control device controls the rejection device to reject the product object indicated by the coded information from the production line, which may include the following steps:

[0194] After receiving the specified sensing signal, when the sensing signal is received for the N3+1th time, an object rejection instruction is sent to the rejection device so that the rejection device rejects the product object indicated by the coded information from the production line.

[0195] The specified sensing signal is the sensing signal emitted when the product object represented by the encoded information approaches.

[0196] Understandably, when the control device 120 detects a defect in a product object and it meets the rejection criteria, it can send a rejection instruction to the rejection device 510. The rejection device 510 can receive the instruction from the control device and perform a rejection operation on the (N3+1)th product object. For example, if there are 6 product objects between the control device and the rejection device, and the control device detects that product object 0 needs to be rejected, and product object 7 is just approaching the proximity sensor, product object 0 enters the operating area of ​​the rejection device. Therefore, the control device can issue a rejection instruction upon receiving the sensing signal from product object 7.

[0197] The rejection criteria can be determined from the defect detection results of the product, including pre-defined rejection defects. For example, on a glove production line, when the control device detects defects in the glove under test such as broken fingers, holes, large tears, wrist scratches, black oil stains, grease stains, recycled materials, cracks, flow marks, or excess material, the control rejection device removes the glove under test from the production line. This application does not limit the specific rejection criteria.

[0198] It should be noted that the rejection device 510 can be applied in scenarios where the optical inspection device 130 includes a first camera, a second camera, a third camera, and a fourth camera. For example, in... Figure 2Based on the embodiment shown, the rejection device 510 can be connected after the optical inspection device to reject specified product objects from the production line.

[0199] For example, on a glove production line, the rejection device may include a main body of the equipment and a gripper hook. The gripper hook may be controlled by a controllable robotic arm mounted on the main body of the equipment. After the controllable robotic arm receives a rejection command sent by the control device, it can perform a rejection operation on the target glove by pulling the glove with the gripper hook.

[0200] In this embodiment, a rejection device is added to the defect detection system, and the control device is equipped with the corresponding rejection device function. This allows unqualified products to be removed from the production line in a timely manner, thereby improving the quality of the products.

[0201] To facilitate understanding of defect detection systems, the following will combine... Figure 6 A schematic diagram of the system is provided. For example, such as... Figure 6 As shown, the defect detection system provided in this embodiment is applied to a glove production line and may include a proximity sensor ( Figure 6 (Not shown in the image), limit switch 01, optical inspection device 130, rejection device 510, and cabinet 04 equipped with control devices. Correspondingly, Figure 7 for Figure 6 Top view.

[0202] For example, the optical inspection device 130 can acquire images of multiple sides of gloves on the production line, and the control device can determine the image corresponding to each glove based on a predetermined image selection method, that is, the first image and the second image of the glove indicated by any coded information, and then perform glove defect detection processing on the image of each glove to obtain the multi-sided defect detection result of the glove; the control device can determine the gloves that need to be rejected based on the glove defect detection result and send the rejection signal to the rejection device 510; the rejection device 510 can perform the rejection action to reject the gloves that need to be rejected.

[0203] For example, the optical inspection device 130 and the rejection device 510 can be combined to form an optical inspection rejection machine.

[0204] The defect detection system provided in this application belongs to a machine vision online inspection system, which consists of a proximity sensor, a limiter, a light inspection device, a rejection device, and a control device. This system can not only realize timely feedback of detection and analysis results to the equipment for execution, but also better adapt to different production lines and realize flexible splitting and linking.

[0205] Based on the aforementioned defect detection system, from the perspective of a control device, this application provides a defect detection method. For example... Figure 8As shown, this defect detection method is applied to the control device of a defect detection system and may include the following steps:

[0206] S801, whenever a sensing signal is received from the proximity sensor, the currently approaching product object is encoded to obtain the encoding information of the currently approaching product object, and a photo-taking trigger signal is sent to the optical detection device, so that the optical detection device responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtains the first image and the second image, and reports the first image and the second image to the control device; wherein, the proximity sensor sends a sensing signal to the control device whenever a product object approaches.

[0207] S802 receives each of the first images and each of the second images.

[0208] S803, for each piece of encoded information, according to a predetermined image selection method, select a first image and a second image containing the product object indicated by the encoded information, and perform product object defect detection processing on the selected first image and second image to obtain the defect detection result of the product object indicated by the encoded information;

[0209] The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

[0210] The specific implementation methods of S801-S803 have been described in the above embodiments, so the embodiments of this application will not be elaborated on in detail here.

[0211] For example, step S803, which involves selecting a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method for each piece of encoded information, may include step A1:

[0212] A1, for each piece of encoded information, based on the order corresponding to the encoded information, the order corresponding to each first image, and the first position difference, determine a first image containing the product object indicated by the encoded information; and, based on the order corresponding to the encoded information, the order corresponding to each second image, and the second position difference, determine a second image containing the product object indicated by the encoded information.

[0213] The order of the encoded information is the order in which the encoded information is sorted according to the generation time; the order of each first image is the order in which the first image is sorted according to the shooting time; and the order of each second image is the order in which the second image is sorted according to the shooting time.

[0214] Wherein, the first position difference is: when a product object to be photographed enters the field of view of the first camera and there is a product object approaching the proximity sensor, there is an interval of N1 product objects between the product object entering the field of view of the first camera and the object approaching the proximity sensor.

[0215] The second positional difference is as follows: when a product object to be photographed enters the field of view of the second camera and there is a product object approaching the proximity sensor, there is an interval of N2 product objects between the product object entering the field of view of the second camera and the object approaching the proximity sensor; where N1 and N2 are different values.

[0216] Step A1, determining the first image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each first image, and the first positional difference, may include step B1:

[0217] B1: Select the first image from each first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1.

[0218] Step A1, determining the second image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference, may include step B2:

[0219] B2: Select a second image from each second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1.

[0220] For example, in one embodiment, the method further includes: starting from the first first image captured by the first camera, encoding each first image in sequence according to the shooting time to obtain the encoding value of each first image; and starting from the first second image captured by the second camera, encoding each second image in sequence according to the shooting time to obtain the encoding value of each second image.

[0221] Step B1, selecting the first image from each of the first images whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1, may include step B11:

[0222] B11: From each of the first images, determine the first image whose encoded value meets the first condition, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1; wherein, the first condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N1+1.

[0223] Step B2, selecting a second image from each of the second images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, may include step B12:

[0224] B12: From each second image, determine the second image whose encoded value meets the second condition, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1; wherein, the second condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N2+1.

[0225] For example, in another embodiment, the method may further include: starting from the next image after the N1+1th first image captured by the first camera, encoding each first image in sequence according to the shooting time to obtain the encoding value of each first image; and starting from the next image after the N2+1th second image captured by the second camera, encoding each second image in sequence according to the shooting time to obtain the encoding value of each second image.

[0226] Step B1, selecting the first image from each of the first images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, may further include step B21:

[0227] B21, from each first image, determine the first image whose encoded value matches the order corresponding to the encoded information, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1.

[0228] Step B2, the step of selecting a second image from each of the second images whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, may further include step B22:

[0229] B22, from each second image, determine the second image whose encoded value matches the order corresponding to the encoded information, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1.

[0230] For example, in one embodiment, the defect detection method of the control device applied to the defect detection system may further include step C1:

[0231] When a defect detection result of a product object indicated by any coded information is detected and meets the rejection criteria, the rejection device is controlled to remove the product object indicated by the coded information from the production line.

[0232] Step C1, the step of controlling the rejection device to reject the product object indicated by the coded information from the production line, may further include step C11:

[0233] C11, after receiving the specified sensing signal, when the sensing signal is received for the N3+1th time, sends an object rejection instruction to the rejection device so that the rejection device rejects the product object indicated by the coded information from the production line.

[0234] The specified sensing signal is the sensing signal emitted when the product object represented by the encoded information approaches.

[0235] For example, in one embodiment, the defect detection method for a control device applied to a defect detection system may further include the following steps:

[0236] D1 receives each of the third and fourth images.

[0237] D2, for each coded information, determine the third image that matches the shooting time of the selected first image and the fourth image that matches the shooting time of the selected second image, perform product mold defect detection processing on the selected third and fourth images to obtain mold detection results, and associate the obtained mold detection results with the defect detection results of the product object indicated by the coded information.

[0238] The specific implementation of the above method steps has been described in other embodiments, and will not be repeated in this embodiment.

[0239] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0240] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0241] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0242] To further understand the working principle of a defect detection system, the following section describes the defect detection process by illustrating the steps involved in defect detection for each component in a production line. For example... Figure 9 As shown, from the perspective of the process flow of each component in the production line, the defect detection process can include the following steps:

[0243] S901: Whenever a product object approaches, the proximity sensor sends a sensing signal to the PLC.

[0244] S902: Whenever the PLC receives a sensing signal, it encodes the product object that is currently nearby and sends the encoded information to the main control device, as well as sending a photo trigger signal to the optical detection device.

[0245] Steps S903 and S906 are executed simultaneously after step S902.

[0246] S903, the optical detection device responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtains the first image and the second image, and reports the first image and the second image to the main control device.

[0247] S904, the main control device receives each first image and each second image.

[0248] S905, for each piece of encoded information, the main control device selects a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and performs product object defect detection processing on the selected first image and second image. Based on the result obtained from the product object defect detection processing, the defect detection result of the product object indicated by the encoded information is determined.

[0249] In step S906, the optical detection device, in response to the received image capture trigger signal, controls the third and fourth cameras to capture images, obtaining a third image and a fourth image, and reports the third image and the fourth image to the main control device. Step S907 is then executed after step S906.

[0250] S907, the main control device receives each third image and each fourth image.

[0251] S908, for each piece of encoded information, the main control device determines a third image that matches the shooting time of the selected first image and a fourth image that matches the shooting time of the selected second image, performs product mold defect detection processing on the selected third and fourth images, and determines the mold detection result based on the result obtained from the product mold defect detection processing.

[0252] Step S909 is executed after steps S905 and S908.

[0253] S909, the main control device associates the obtained mold inspection results with the defect inspection results of the product object indicated by the coded information.

[0254] S9010, the main control device determines whether the product object indicated by the encoded signal needs to be removed from the production line. If yes, proceed to step S9011; if no, proceed to step S9013.

[0255] For example, the main control device can determine whether a product needs to be removed according to a predetermined removal rule.

[0256] S9011, the main control device sends a rejection command to the rejection device.

[0257] For example, the main control device can send a rejection instruction to the rejection device after determining that the product needs to be rejected;

[0258] S9012, the rejection device receives the rejection command issued by the main control device and executes the rejection operation.

[0259] For example, the rejection device can receive rejection instructions from the master control device and perform rejection operations using predetermined rejection methods.

[0260] S9013, the main control device does not send rejection instructions to the rejection device.

[0261] The specific implementation methods of S901-S909 have been described in the above embodiments, so the embodiments of this application will not be elaborated on in detail here.

[0262] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0263] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0264] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0265] Based on the aforementioned defect detection system, from the perspective of a control device, this application provides a defect detection device. For example... Figure 10 As shown, this defect detection device is used in the control device of a defect detection system and may include the following modules:

[0266] The transmitting module 1010 is configured to encode the currently approaching product object whenever it receives a sensing signal from the proximity sensor, obtain the encoding information of the currently approaching product object, and send a photo-taking trigger signal to the photodetector so that the photodetector responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtain a first image and a second image, and reports the first image and the second image to the control device; wherein, the proximity sensor sends a sensing signal to the control device whenever a product object approaches;

[0267] The receiving module 1020 is used to receive each of the first images and each of the second images;

[0268] The detection module 1030 is used to select, for each piece of encoded information, a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and to perform product object defect detection processing on the selected first image and second image to obtain the defect detection result of the product object indicated by the encoded information; wherein, the image selection method is a selection method set based on the first position difference between the proximity sensor and the first camera and the second position difference between the proximity sensor and the second camera.

[0269] The specific implementation of the above-mentioned device has been described in other embodiments, and will not be repeated in this embodiment.

[0270] In summary, in this embodiment, the image of the product object on the production line acquired by the optical inspection device includes images of multiple sides of the product object. Furthermore, the control device can determine the image corresponding to each product object based on a predetermined image selection method, namely, a first image and a second image containing any encoded information indicating the product object. Then, product object defect detection processing is performed on the image of each product object. Based on the results of the product object defect detection processing, the defect detection results for both sides of the corresponding product object can be determined. Therefore, compared to the prior art of acquiring single-sided images for defect detection processing, the defect detection system of this solution determines the defect detection results of the product object based on multi-sided visual defect detection and correlation judgment of the defect detection results corresponding to each side. This eliminates blind spots in the field of view, greatly improves the completeness of the defect detection results, and thus improves the accuracy of the product object defect detection results.

[0271] Furthermore, in this solution, the optical inspection device is equipped with a first and second camera arranged in a staggered manner to acquire images of multiple sides of the product object. It is evident that this optical inspection device can achieve simultaneous image acquisition of multiple sides of the product object within a compact design space, while ensuring full online inspection. Compared to traditional visual inspection systems that inspect each side individually, this significantly improves inspection efficiency.

[0272] Furthermore, in this solution, the encoding of product objects based on proximity sensors can provide the conditions for the association of multi-faceted detection of the same product object.

[0273] This application also provides an electronic device, such as... Figure 11 As shown, it includes a processor 1101, a communication interface 1102, a memory 1103, and a communication bus 1104, wherein the processor 1101, the communication interface 1102, and the memory 1103 communicate with each other through the communication bus 1104.

[0274] Memory 1103 is used to store computer programs;

[0275] The processor 1101 is used to implement the above-mentioned defect detection method when executing the program stored in the memory 1103.

[0276] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0277] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0278] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0279] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0280] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described defect detection methods.

[0281] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the defect detection methods described above.

[0282] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state disk (SSD)).

[0283] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0284] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the method embodiments are basically similar to the system embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the system embodiments.

[0285] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A defect detection system, characterized in that, include: The device includes a control unit, an optical inspection unit, and a proximity sensor. The optical inspection unit is equipped with a first camera and a second camera arranged in a staggered manner. The first camera is used to photograph one side of each product object to be inspected as it is conveyed on the production line, and the second camera is used to photograph the other side of each product object. The proximity sensor is installed on the path of the production line and is used to send a sensing signal to the control device whenever a product object approaches. The control device is used to encode the currently approaching product object whenever it receives the sensing signal, obtain the encoding information of the currently approaching product object, and send a photo-taking trigger signal to the optical detection device; the encoding information is used to identify the product object. The optical detection device is used to respond to the received photo-taking trigger signal, control the first camera and the second camera to take pictures, obtain the first image and the second image, and report the first image and the second image to the control device. The control device is further configured to receive each first image and each second image; and, for each encoded information, to select a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, and to perform product object defect detection processing on the selected first image and the second image respectively, and to determine the defect detection result of the product object indicated by the encoded information based on the result obtained from the product object defect detection processing on the selected first image and the second image. The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

2. The system according to claim 1, characterized in that, For each piece of encoded information, the control device selects a first image and a second image containing the product object indicated by the encoded information according to a predetermined image selection method, including: For each piece of encoded information, a first image containing the product object indicated by the encoded information is determined based on the order corresponding to the encoded information, the order corresponding to each first image, and the first positional difference; and a second image containing the product object indicated by the encoded information is determined based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference. The order of the encoded information is the order in which the encoded information is sorted according to the generation time; the order of each first image is the order in which the first image is sorted according to the shooting time; and the order of each second image is the order in which the second image is sorted according to the shooting time.

3. The system according to claim 2, characterized in that, The first positional difference is as follows: when a product object to be photographed enters the field of view of the first camera and there is a product object approaching the proximity sensor, there is an interval of N1 product objects between the product object entering the field of view of the first camera and the object approaching the proximity sensor. The second positional difference is as follows: when a product object to be photographed enters the field of view of the second camera and there is a product object approaching the proximity sensor, there is an interval of N2 product objects between the product object entering the field of view of the second camera and the object approaching the proximity sensor; where N1 and N2 are different values. The control device determines the first image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each first image, and the first position difference, including: From each first image, select the first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1; The control device determines a second image containing the product object indicated by the encoded information based on the order corresponding to the encoded information, the order corresponding to each second image, and the second positional difference, including: From each of the second images, select the second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1.

4. The system according to claim 3, characterized in that, The control device is also used to: starting from the first first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoded value of each first image; Starting with the first second image captured by the second camera, each second image is encoded sequentially according to the shooting time to obtain the encoded value of each second image; The control device selects, from each of the first images, a first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, including: From each first image, determine the first image whose encoded value meets the first condition, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs by N1+1; wherein, the first condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N1+1. The control device selects a second image from each second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, including: From each second image, determine the second image whose encoded value meets the second condition, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1; wherein, the second condition is that the encoded value is not less than the order corresponding to the encoded information and differs by N2+1.

5. The system according to claim 3, characterized in that, The control device is also used to: starting from the next image after the N1+1th first image captured by the first camera, encode each first image in sequence according to the shooting time to obtain the encoded value of each first image; Starting from the next image after the N2+1th second image captured by the second camera, each second image is encoded sequentially according to the capture time to obtain the encoded value of each second image; The control device selects, from each of the first images, a first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1, including: From each first image, determine the first image whose encoded value matches the order corresponding to the encoded information, and obtain the first image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N1+1. The control device selects a second image from each second image whose corresponding order is not less than the order corresponding to the encoded information and differs from it by N2+1, including: From each second image, determine the second image whose encoded value matches the order corresponding to the encoded information, and obtain the second image whose corresponding order is not less than the order corresponding to the encoded information and differs by N2+1.

6. The system according to any one of claims 1-5, characterized in that, The system further includes a rejection device, wherein the rejection device is positioned after the optical inspection device on the path of the production line; The control device is also used to control the rejection device to remove the product object indicated by the coded information from the production line when the defect detection result of the product object indicated by any coded information is detected and meets the rejection conditions. The rejection device is used to reject the object represented by the encoded information from the production line under the control of the control device.

7. The system according to claim 6, characterized in that, The positional difference between the proximity sensor and the rejection device is as follows: when a product object enters the operable area of ​​the rejection device and a product object approaches the proximity sensor, there is an interval of N3 product objects between the product object entering the operable area and the product object approaching the proximity sensor. The control device controls the rejection device to reject the product object indicated by the coded information from the production line, including: After receiving the specified sensing signal, when the sensing signal is received for the N3+1th time, an object rejection instruction is sent to the rejection device so that the rejection device rejects the product object indicated by the coded information from the production line. The specified sensing signal is the sensing signal emitted when the product object represented by the encoded information approaches.

8. The system according to any one of claims 1-5, characterized in that, Each product object is associated with a product mold when it is transferred on the production line; the optical inspection device is also equipped with a third camera and a fourth camera arranged in a staggered manner. The third camera is a camera that takes pictures of one side of each product mold transferred on the production line, and the fourth camera is a camera that takes pictures of the other side of each product mold. The optical detection device is also used to control the third camera and the fourth camera to take pictures in response to the received picture trigger signal, to obtain the third image and the fourth image, and to report the third image and the fourth image to the control device; The control device is further configured to receive each third image and each fourth image; and, for each encoded information, determine a third image that matches the shooting time of the selected first image and a fourth image that matches the shooting time of the selected second image, perform product mold defect detection processing on the selected third image and fourth image, determine a mold detection result based on the result of the product mold defect detection processing, and associate the obtained mold detection result with the defect detection result of the product object indicated by the encoded information.

9. The system according to any one of claims 1-5, characterized in that, The assembly line is also equipped with limiters on the support for transferring product objects.

10. The system according to any one of claims 1-5, characterized in that, The control device includes: a main control device and a programmable logic controller (PLC); The PLC is used to encode the currently approaching product object whenever it receives the sensing signal, send the encoded information to the main control device, and send a photo trigger signal to the optical detection device. The main control device is used to receive each first image and each second image; and, for each encoded information, according to a predetermined image selection method, select a first image and a second image containing the product object indicated by the encoded information, and perform product object defect detection processing on the selected first image and second image, and determine the defect detection result of the product object indicated by the encoded information based on the result obtained from the product object defect detection processing. The optical detection device is specifically used to report the first image and the second image to the main control device.

11. A defect detection method based on the defect detection system according to any one of claims 1-10, characterized in that, Applied to a control device; the method includes: Whenever a proximity sensor sends a sensing signal, the system encodes the currently approaching product object to obtain its encoding information, and sends a photo-taking trigger signal to the optical detection device. This causes the optical detection device to respond to the received photo-taking trigger signal, controlling the first and second cameras to take photos, obtaining a first image and a second image, and then reporting the first and second images to the control device. The proximity sensor sends a sensing signal to the control device whenever a product object approaches; the encoding information is used to identify the product object. Receive each first image and each second image; For each encoded information, a first image and a second image containing the product object indicated by the encoded information are selected according to a predetermined image selection method. Product object defect detection processing is performed on the selected first image and the second image respectively. Based on the results obtained from the product object defect detection processing of the selected first image and the second image, the defect detection result of the product object indicated by the encoded information is determined. The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

12. A defect detection device based on the defect detection system according to any one of claims 1-10, characterized in that, Used in control devices; The defect detection device includes: The transmitting module is used to encode the currently approaching product object whenever it receives a sensing signal from the proximity sensor, obtain the encoding information of the currently approaching product object, and send a photo-taking trigger signal to the photodetector, so that the photodetector responds to the received photo-taking trigger signal, controls the first camera and the second camera to take pictures, obtain a first image and a second image, and reports the first image and the second image to the control device; wherein, the proximity sensor sends a sensing signal to the control device whenever a product object approaches; the encoding information is used to identify the product object; The receiving module is used to receive each first image and each second image; The detection module is used to select, according to a predetermined image selection method, a first image and a second image containing the product object indicated by the encoded information for each encoded information, and to perform product object defect detection processing on the selected first image and the second image respectively. Based on the results obtained from the product object defect detection processing on the selected first image and the second image, the module determines the defect detection result of the product object indicated by the encoded information. The image selection method is a selection method set based on the first positional difference between the proximity sensor and the first camera and the second positional difference between the proximity sensor and the second camera.

13. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method of claim 11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method of claim 11.

Images