Edge equipment, center equipment, defect detection method, equipment and medium

Abstract

The invention provides edge equipment, center equipment, a defect detection method, equipment and a medium. According to one example of the application, the edge device may include an edge detection unit and an edge management unit; wherein the edge detection unit is connected with the edge management unit and is used for acquiring detection data of a to-be-detected product based on configuration data of the to-be-detected product and generating a detection result of the detection data by utilizing a defect detection model; and the edge management unit is used for sending the detection result to the central equipment, so that the central equipment judges that the to-be-detected product has the defect problem.

Description

Technical Field

[0001] This application relates to the field of detection technology, and in particular to an edge device, a central device, a defect detection method, an apparatus, and a medium. Background Technology

[0002] In the industrial product manufacturing sector, the stability and consistency of product quality are the cornerstones of a manufacturer's sustainable development. Due to unavoidable technological limitations, various defects are inevitable during the manufacturing process, such as appearance defects, abnormal properties, substandard functional indicators, or even complete damage. These defects not only affect the performance of the product but may also pose a threat to consumer safety. Therefore, implementing effective quality inspection during the product manufacturing process is of paramount importance.

[0003] Current defect detection systems typically employ a centralized private cloud model to identify and detect product defects. However, with the continuous expansion of product production capacity, the centralized private cloud model faces problems such as increased system pressure, increased detection latency, and excessive network bandwidth consumption, resulting in high system resource pressure and low detection efficiency. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this application provides an edge device, a central device, a defect detection method, an apparatus, and a medium.

[0005] According to a first aspect of any embodiment of this application, an edge device is provided, comprising: an edge detection unit and an edge management unit; wherein...

[0006] The edge detection unit is connected to the edge management unit and is used to obtain the detection data of the product to be detected based on the configuration data of the product to be detected, and to generate the detection result of the detection data using a defect detection model.

[0007] The edge management unit is used to send the detection results to the central device so that the central device can determine that the product under test has a defect.

[0008] According to a second aspect of any embodiment of this application, a central device is provided, comprising: a central detection unit and a central management unit; wherein...

[0009] The central management unit is connected to the central detection unit and is used to receive the detection results of the detection data sent by the edge device. The detection results are generated based on the configuration data of the product to be detected and the defect detection model.

[0010] The central testing unit is used to determine, based on the testing results, that the product under test has a defect.

[0011] According to a third aspect of the embodiments of this application, a defect detection method is provided, the method being applied to an edge device, the edge device comprising: an edge detection unit and an edge management unit;

[0012] The method includes:

[0013] The edge detection unit obtains the detection data of the product under test based on the configuration data of the product under test, and generates the detection result of the detection data using the defect detection model;

[0014] The edge management unit sends the detection results to the central device so that the central device can determine that the product under test has a defect.

[0015] According to a fourth aspect of the embodiments of this application, a defect detection method is provided, the method being applied to a central device, the central device comprising: a central detection unit and a central management unit;

[0016] The method includes:

[0017] The central management unit receives the detection results from the detection data sent by the edge devices. The detection results are generated based on the configuration data of the product to be detected and the defect detection model.

[0018] Based on the test results, the central testing unit determines that the product under test has a defect.

[0019] According to a fifth aspect of any embodiment of this application, an electronic device is provided, comprising:

[0020] processor;

[0021] Memory used to store processor-executable instructions;

[0022] The processor executes the executable instructions to implement the method performed by the edge device or the method performed by the central device as described in any of the embodiments of this application above.

[0023] According to a sixth aspect of any embodiment of this application, a computer-readable storage medium is provided having computer instructions stored thereon that, when executed by a processor, implement the method performed by an edge device or by a central device as described in any of the embodiments of this application above.

[0024] The technical solution provided in this application may include the following beneficial effects:

[0025] As can be seen from the above embodiments, the edge detection unit obtains the detection data of the product under test based on the configuration data of the product under test, and generates the detection results of the detection data using the defect detection model. The edge management unit sends the detection results to the central device so that the central device can determine that the product under test has a defect problem. The edge device can use its own resources to process the detection data and generate the detection results in real time. The central device does not need to perform model detection but only needs to perform simple judgment, which reduces the amount of data that the central device needs to process. This enables the central device and the edge device to work together, ensuring the reasonable allocation and utilization of computing resources. While improving the detection efficiency of the product under test, it reduces the resource pressure on the central device.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form part of this application, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0028] Figure 1 This is a schematic diagram of the structure of a defect detection system according to an exemplary embodiment of this application;

[0029] Figure 2 This is a schematic diagram of the structure of an edge device according to an exemplary embodiment of this application;

[0030] Figure 3 This is a schematic diagram of the structure of a central device according to an exemplary embodiment of this application;

[0031] Figure 4 This is a flowchart illustrating a defect detection method according to an exemplary embodiment of this application;

[0032] Figure 5 This is a flowchart illustrating another defect detection method according to an exemplary embodiment of this application;

[0033] Figure 6 This is an interactive diagram illustrating a registration method for an edge device according to an exemplary embodiment of this application;

[0034] Figure 7 This is a schematic diagram illustrating an online site and an offline site according to an exemplary embodiment of this application;

[0035] Figure 8 This is an interactive diagram illustrating an offline site defect detection method according to an exemplary embodiment of this application;

[0036] Figure 9This is an interactive diagram illustrating a defect detection method for a first online site according to an exemplary embodiment of this application;

[0037] Figure 10 This is a flowchart illustrating a defect detection method for a second online site according to an exemplary embodiment of this application;

[0038] Figure 11 This is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of this application. Detailed Implementation

[0039] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0040] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0041] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0042] Current defect detection systems, due to their centralized private cloud model, face problems such as increased system pressure, increased detection latency, and excessive network bandwidth consumption as product production capacity continues to expand, resulting in high system resource pressure and low detection efficiency.

[0043] To address the aforementioned problems, this application proposes an edge device, a central device, a defect detection method, and a system. The following embodiments are provided to further illustrate this application:

[0044] Please see Figure 1 , Figure 1This is a schematic diagram illustrating the structure of a defect detection system according to an exemplary embodiment of this application. The defect detection system is used to detect whether a product has defects. The defect detection system may include: an edge device 10, a central device 11, a station 12, a Manufacturing Execution System (MES) 13, a product inspection device 14, and a Defective File System (DFS) 15.

[0045] Edge device 10 is a computing device deployed close to site 12, responsible for initial data processing and inspection, and can be an edge cloud. Edge device 10 can communicate with central device 11, manufacturing execution system 13, product inspection equipment 14, and defect file system 15.

[0046] Central device 11 is the core processing unit of the defect detection system, responsible for receiving, analyzing, and storing detection results from edge devices, as well as performing higher-level data processing and decision-making; it can be a central cloud. Central device 11 communicates with manufacturing execution system 13 and defect file system 15.

[0047] Site 12 is a location for preliminary product testing or storage, and can be an offline or online site. An offline site is a process site that acquires test data of the product to be tested offline, while an online site is a process site that acquires test data online.

[0048] Manufacturing Execution System 13 is an information system used in manufacturing to monitor and guide the production process. Manufacturing Execution System 13 can communicate with edge devices 10 and central devices 11.

[0049] Product testing equipment 14, located at site 12, is used to perform preliminary testing on products using physical or chemical methods, and to generate product testing documents. Product testing equipment 14 can be automated optical inspection (AOI) equipment, scanners, etc.

[0050] Product testing equipment 14 can store the generated test files in the storage directory of the site's local server, or upload them to the defect file system 15 for storage.

[0051] The defect file system 15 is a database or file system that stores online testing files for the products to be tested.

[0052] Please see Figure 2 , Figure 2 This is a schematic diagram illustrating the structure of an edge device according to an exemplary embodiment of this application. The edge device 10 may include an edge detection unit 103 and an edge management unit 101.

[0053] The edge detection unit 103, connected to the edge management unit 101, is used to acquire detection data of the product under test based on its configuration data. It then uses a defect detection model to generate detection results from the detection data. The edge detection unit 103 sends the detection results to the edge management unit 101.

[0054] The edge management unit 101 is used to send the detection results of the detection data to the central device 11 so that the central device 11 can determine that the product to be detected has a defect.

[0055] The product to be tested is a product that requires quality inspection during the manufacturing process. It can be a product such as a glass substrate, electronic component, or optical component in the manufacturing process of a liquid crystal display (LCD), or various products from other industries.

[0056] Configuration data consists of relevant parameters and rule sets for sites and devices during edge device registration and defect detection. It can be Recipe data. Configuration data may include: the mapping between edge devices and their associated sites, and the mapping between offline sites and the storage directories of offline detection files, etc.

[0057] The test data is obtained by parsing the test files generated after the initial test of the product. It is used to reflect the size, shape and other attribute information of the product, as well as possible defects.

[0058] Taking a glass substrate as an example, the inspection data can include: defect data and product images. The defect data contains information about the locations of defects on the glass substrate, such as defect location, defect type, defect size, and defect quantity. The product image is an actual image of the glass substrate.

[0059] Defect detection models are built using techniques such as machine learning or deep learning. They are used to process and analyze inspection data to identify defects in the product being inspected. These models are trained and optimized using a large number of training samples, resulting in high detection accuracy and generalization ability.

[0060] The inspection results are obtained by processing and analyzing the inspection data using a defect detection model. The results may include information such as whether the product has defects, the location, size, and type of the defects.

[0061] Defects are quality problems caused by material issues, process issues, equipment issues, etc. in the manufacturing process of the product under test. For example, the product's size does not meet the requirements, the shape is deformed, the color is uneven, the texture is abnormal, etc.

[0062] The edge device 10 may further include an edge model service unit 104 and a detection scheduling unit 102. The edge model service unit 104 is used to start and manage the defect detection model in the edge device 10. The detection scheduling unit 102 is used to perform load balancing scheduling of defect detection tasks in the edge device 10 based on the load of multiple server nodes in the same site, thereby achieving load balancing of the edge device 10.

[0063] Please see Figure 3 , Figure 3 This is a schematic diagram of a central device according to an exemplary embodiment of this application. The central device 11 may include a central management unit 110 and a central detection unit 111.

[0064] The central management unit 110, connected to the central detection unit 111, is used to receive the detection results of the detection data sent by the edge device 10. The central management unit 110 then sends the detection results to the central detection unit 111.

[0065] The central inspection unit 111 is used to determine whether the product under inspection has defects based on the inspection results.

[0066] The central device 11 may also include a central model service unit 112. The central model service unit 112 is used to provide graphical and automated model training functions, as well as to manage the defect detection models in the central device 11.

[0067] Please see Figure 4 , Figure 4 This is a flowchart illustrating a defect detection method according to an exemplary embodiment of this application. The defect detection method can be applied to an edge device 10. The method may include the following steps:

[0068] Step 402: The edge detection unit obtains the detection data of the product under test based on the configuration data of the product under test, and generates the detection results of the detection data using the defect detection model.

[0069] In this step, edge device 10 starts a listener and listens for the product to be detected in online or offline sites based on the configuration data of the product to be detected. When it detects an inbound product to be detected in an online or offline site bound to edge device 10, it notifies edge detection unit 103 to create an edge detection task.

[0070] The edge detection unit 103 creates an edge detection task and acquires the inspection file of the product to be inspected generated by the product inspection equipment 14, either online or offline. The inspection file is preprocessed and parsed to obtain inspection data such as defect point data and product images.

[0071] The detection data is input into the defect detection model, which processes the data and performs defect detection to generate detection results. Edge devices do not need to transmit detection data to the central device for processing; they can process the data and generate results instantly, thereby improving defect detection efficiency.

[0072] In one embodiment, the configuration data may include warning conditions. After acquiring the detection data of the product to be detected, the edge detection unit 103 can compare the detection data with the warning conditions in the configuration data to determine whether the detection data meets the warning conditions. If it is determined that the detection data meets the warning conditions, a warning message is generated so that the product to be detected can be detected more quickly or assigned to a higher-level inspection personnel.

[0073] The warning conditions can be set based on specific thresholds, such as the number, size, and location of defects, or they can be set based on the importance of the product to be inspected, such as the product type of a key process or a key product line.

[0074] For example, the warning condition is that the number of defect points in the product to be inspected exceeds a preset number. When the number of defect points in the detection data exceeds the preset number, the edge detection unit 103 generates a warning message.

[0075] Warning information is used to prioritize the inspection of products to be inspected. Warning information may include: details such as the identification of the product to be inspected, the number, location, and type of defects, as well as the possible risk level or recommended actions.

[0076] As described above, the edge detection unit generates early warning information for the product to be inspected when the detection data meets the early warning conditions, ensuring that important or high-risk products to be inspected are given priority and improving the timeliness and response speed of defect detection.

[0077] Step 404: The edge management unit sends the test results to the central device so that the central device can determine that the product under test has a defect.

[0078] In this step, the edge management unit 101 sends edge task information, detection data, and detection results of the edge detection task to the central device 11. If the detection result is a negative result such as a detection anomaly, or if there are defects such as abnormal product size, abnormal shape, or abnormal structure, the central device 11 can determine that the product to be detected has a defect.

[0079] If the test result is normal, and there are no defects such as abnormal product size, abnormal shape, or abnormal structure, the central equipment 11 can determine that the product to be tested has no defects.

[0080] By using edge devices to perform defect detection on the edge side of the detection data, the central device does not need to perform model detection but only needs to make simple judgments, which can reduce the amount of data that the central device needs to process and reduce the resource pressure on the central device.

[0081] In one embodiment, before the edge device 10 sends the test results to the central device 11, the inspector can re-inspect the product to be tested on the edge device 10 based on the test data and test results, generating a manual re-inspection result. The edge management unit 101 can send the generated manual re-inspection result to the central device 11 as a new test result. The central device 11 determines whether the product to be tested has any defects based on the new test result.

[0082] The defect detection method in this embodiment obtains the detection data of the product under test based on the configuration data of the product under test by the edge detection unit, and generates the detection results of the detection data using the defect detection model. The edge management unit sends the detection results to the central device so that the central device can determine that the product under test has a defect problem. The edge device can process the detection data and generate the detection results in real time using its own resources. The central device does not need to perform model detection but only needs to perform simple judgment, which reduces the amount of data that the central device needs to process. It realizes the collaborative work of the central device and the edge device, ensures the rational allocation and utilization of computing resources, improves the detection efficiency of the product under test, and reduces the resource pressure on the central device.

[0083] Furthermore, the products to be inspected can be those from the bottleneck process segment in the manufacturing process. By converging defect detection to the edge equipment side, priority inspection of products from the bottleneck process segment can be ensured, thus avoiding impact on the production efficiency and capacity of products from the bottleneck process segment.

[0084] Please see Figure 5 , Figure 5 This is a flowchart illustrating another defect detection method according to an exemplary embodiment of this application. The defect detection method can be applied to central device 11. The method may include the following steps:

[0085] Step 502: The central management unit receives the detection results of the detection data sent by the edge device. The detection results are generated based on the configuration data of the product to be detected and the defect detection model.

[0086] In this step, edge device 10 generates detection results based on the configuration data of the product to be tested and the defect detection model in edge device 10. Edge device 10 then sends the detection results to central device 11.

[0087] The central management unit 110 receives the detection results from the detection data sent by the edge device 10.

[0088] Step 504: Based on the test results, the central testing unit determines that the product under test has defects.

[0089] In this step, if the test result is negative, indicating a defect, the central testing unit 111 can determine that the product under test has a defect. If the test result is normal, indicating no defect, the central testing unit 111 can determine that the product under test has no defect.

[0090] In one embodiment, after receiving the detection data and detection results sent by the edge management unit, the central detection unit 111 stores the detection data and detection results and generates central task information for the product to be tested. The central task information is used to record and store relevant information such as the detection data and detection results of the product to be tested.

[0091] The testing personnel can re-inspect the product to be tested based on the testing data and test results in the central task information on the central equipment 11 side, and generate the manual re-inspection result of the product to be tested. The central testing unit 111 stores the manual re-inspection result.

[0092] The central inspection unit 111 obtains the results of manual re-inspection of the product to be inspected based on the central task information. If the manual re-inspection result is negative, and it is determined that there is a defect in the manual re-inspection result, the product to be inspected is also determined to have a defect.

[0093] The defect information may include detailed information about the defect locations of the product under inspection. This detailed information reflects the location, shape and characteristics, severity, and cause of the defect, which helps in quickly locating and repairing the defect.

[0094] As mentioned above, storing the test data and results generates central task information for the product to be tested, which helps with subsequent analysis and traceability. Based on the central task information, the results of manual re-inspection of the product to be tested are obtained. If defects are found in the manual re-inspection results, the product to be tested is confirmed to have defects. Combining machine testing with manual judgment can significantly improve the accuracy of testing and reduce false alarms.

[0095] In one embodiment, after determining that the product under inspection has a defect, the central management unit 110 can immediately send a repair request to the manufacturing execution system 13 corresponding to the product under inspection. The manufacturing execution system 13 receives the repair request sent by the central equipment 11, learns detailed information about the defect, and then performs subsequent repairs and production adjustments.

[0096] The repair message is used to reflect detailed information about the defect. The repair message may include information such as defect type, location, and severity.

[0097] As described above, by sending repair information to the manufacturing execution system corresponding to the product under inspection through the central management unit, relevant personnel can quickly learn the detailed information of the product under inspection that needs repair, thereby reducing the product waiting time for repair.

[0098] The defect detection method in this embodiment receives the detection results of the detection data sent by the edge devices through the central management unit. Based on the detection results, the central detection unit determines that the product to be detected has defects. The edge devices can use their own resources to process the detection data and generate detection results in real time. The central devices do not need to perform model detection, but only need to perform simple judgment, which reduces the amount of data that the central devices need to process. This enables the central devices and edge devices to work together, ensuring the reasonable allocation and utilization of computing resources. While improving the detection efficiency of the product to be detected, it also reduces the resource pressure on the central devices.

[0099] The foregoing embodiments described how the collaborative work of edge devices and central devices reduces resource pressure on the defect detection system and improves detection efficiency. The following embodiments will provide a detailed description of the registration process prior to defect detection, and this method can be applied to any of the embodiments described above.

[0100] In one embodiment, please refer to Figure 6 , Figure 6 An interactive diagram of a registration method for an edge device is shown. This registration method may include the following steps:

[0101] Step 602: The central device creates nodes for the edge devices.

[0102] In this step, the central device 11 creates a corresponding node for the edge device 10 that will be connected, for management and communication purposes. A new record is added to the database, containing basic information such as the device ID of the edge device 10.

[0103] Step 604: Configure site identifiers on edge devices.

[0104] In this step, edge device 10 is configured with a site identifier and bound to the site 12 corresponding to the site identifier. At the same time, central device 11 automatically obtains the system fingerprint of the server where edge device 10 is located, displays it on the central device 11's page, and records the system fingerprint.

[0105] Among them, the site identifier is the unique identifier of the site, which can be the site ID, site name, etc.

[0106] Step 606: Configure the address of the central device on the edge device.

[0107] In this step, edge device 10 configures the address of central device 11 and stores the address of central device 11 in the database.

[0108] Step 608: The edge device sends a registration request to the central device.

[0109] In this step, edge device 10 stores the configured address of central device 11 in the database, automatically connects to central device 11, and establishes a connection with central device 11. After successfully connecting to central device 11, edge management unit 101 sends a registration request to central device 11. The registration request may carry the site identifier corresponding to edge device 10, and may also carry the system fingerprint of the server where edge device 10 resides.

[0110] Step 610: The central device registers the edge device.

[0111] In this step, the central management unit 110 receives the registration request sent by the edge device 10 and registers the edge device 10 based on the site identifier and other information carried in the registration request.

[0112] Step 612: The central device creates or updates configuration data.

[0113] In this step, the central device 11 creates or updates the configuration data of the site 12.

[0114] Step 614: The central equipment synchronizes the configuration data and defect detection model.

[0115] In this step, the central model service unit 112 trains the initial model to obtain the trained defect detection model.

[0116] The central management unit 110 searches for the configuration data of the site 12 based on the site identifier corresponding to the edge device 10, and synchronously sends the configuration data and defect detection model to the edge device 10.

[0117] When the configuration data is updated, the central device 11 can synchronize the configuration data with the edge devices bound to the sites associated with the configuration data in real time to ensure that the configuration data on the edge devices is updated in real time.

[0118] Step 616: Edge devices update configuration data and defect detection models.

[0119] In this step, the edge management unit 101 in the edge device 10 receives the configuration data and the trained defect detection model sent by the central device 11, updates the defect detection model in the edge device 10, and replaces the configuration data in the database with the received configuration data.

[0120] Step 618: The edge device periodically synchronizes configuration data.

[0121] In this step, the edge device 10 periodically requests the configuration data corresponding to the synchronization site identifier from the central device 11.

[0122] As described above, the initial model is trained by the central model service unit to obtain the trained defect detection model. The central model service unit receives the registration request sent by the edge device, registers the edge device based on the registration request, determines the configuration data according to the site identifier, and sends the configuration data and defect detection model to the edge device. The central device is responsible for complex model training and data analysis tasks, while the edge device is responsible for real-time data detection and response, thereby improving the overall system's resource utilization efficiency.

[0123] The foregoing embodiments described the registration process for edge devices and central devices in the early stages of defect detection. The following embodiments will provide a more detailed description of the defect detection process for online and offline sites, and can be applied to any of the embodiments described above.

[0124] To further introduce online and offline sites, Figure 7 A schematic diagram of an online site and an offline site is shown. (For example...) Figure 7 As shown, the defect detection system includes: a central device 11, a manufacturing execution system 13, a first online station a, a second online station b, an offline station c, an edge device 10A, an edge device 10C, inspection personnel on the central device 11 side, and inspection personnel on the edge device 10C side.

[0125] The central device 11 can communicate with various devices in the system. Edge device 10A is bound to the first online site a, edge device 10C is bound to the offline site c, and the second online site b has no bound edge devices.

[0126] In one embodiment, please refer to Figure 8 , Figure 8 An interactive diagram of an offline site defect detection method is shown. (Based on the above...) Figure 7 Taking the offline site c shown as an example, the defect detection method may include the following steps:

[0127] Step 802: The edge device monitors the storage directory of the offline detection files.

[0128] In this step, the edge device 10C creates a listener based on the correspondence between the offline site c and the storage directory of the offline detection files in the configuration data, and listens to the storage directory of the offline detection files in the offline site c.

[0129] Step 804: The edge device determines whether there are any newly added offline detection files.

[0130] In this step, the edge device 10C determines whether there are any newly added offline detection files in the storage directory.

[0131] If a new offline detection file exists in the storage directory, proceed to step 806;

[0132] If the newly added offline detection file does not exist in the storage directory, proceed to step 802.

[0133] Step 806: Edge device notification creates edge detection task.

[0134] In this step, after detecting a new offline detection file in offline site c, the listener sends a message to the edge detection unit in edge device 10C to notify the edge detection unit to create an edge detection task.

[0135] Step 808: The edge device parses the offline detection file to obtain detection data.

[0136] In this step, the edge detection unit, upon detecting a newly added offline detection file in offline site c, creates an edge detection task, generates edge task information, preprocesses and parses the offline detection file to obtain detection data, and records the detection data in the edge task information. The offline detection file is the detection file stored in the local storage directory of offline site c.

[0137] Step 810: The edge device uses the defect detection model to generate detection results.

[0138] In this step, the edge detection unit inputs the product image from the detection data into the defect detection model in the edge device 10C. The defect detection model performs defect detection on the product image and generates the detection results of the detection data.

[0139] The edge management unit in edge device 10C sends detection data and detection results to central device 11. Central device 11 stores the detection data and detection results in the database, creates a central detection task, and generates central task information for the central detection task. The central task information of central device 11 is consistent with the detection data, detection results, and other data in the edge task information of edge device 10C.

[0140] Step 812: The edge device uses the manual re-inspection result as the new inspection result.

[0141] In this step, the inspector performs a re-inspection on the platform provided by the edge inspection unit of the edge device 10C, generating a manual re-inspection result on the edge device side. The edge device 10C uses the manual re-inspection result on the edge device side as the new inspection result and sends the edge task information and the new inspection result to the central device 11.

[0142] Step 814: The central device monitors edge task information and inspection results.

[0143] In this step, the central device 11 listens to the edge task information and inspection results, stores the new inspection results in the database and records them in the central task information.

[0144] Step 816: The central equipment determines whether the defect meets the conditions for sending it for repair.

[0145] In this step, the central equipment 11 obtains new inspection results based on the central task information. If it is determined that there are defects in the new inspection results, it determines that the product to be inspected has defects and whether the defects meet the repair conditions in the configuration information.

[0146] If the defect meets the conditions for sending for repair, then proceed to step 818;

[0147] If the defect does not meet the conditions for repair, proceed to step 820.

[0148] Step 818: The central equipment sends a repair request to the manufacturing execution system.

[0149] In this step, if the defect meets the repair conditions in the configuration information, a repair operation is triggered, and the central management unit 110 sends the repair information to the manufacturing execution system 13.

[0150] Step 820: The central equipment does not need to send a repair request.

[0151] In this step, if the defect does not meet the conditions for sending for repair, the repair operation will not be triggered, and the central management unit 110 does not need to send repair information.

[0152] As described above, the edge detection unit responds to the newly added offline detection files in the offline site by parsing the offline detection files to obtain detection data, thereby realizing defect detection of the products to be tested in the offline site. There is no need to upload the offline detection files to the network, which can ensure timely detection of the products to be tested in the offline site, significantly reduce data transmission latency, and improve the detection efficiency of the products to be tested.

[0153] Furthermore, even if the site experiences network problems that prevent the test files of the products to be tested from being uploaded to the defect file system, offline test files can be obtained for defect testing, thereby avoiding impacting the defect testing of products in bottleneck process segments.

[0154] In one embodiment, please refer to Figure 9 , Figure 9 An interactive diagram of a defect detection method for a first online site is shown. (Based on the above...) Figure 7 Taking the first online site a as an example, the defect detection method may include the following steps:

[0155] Step 902: The edge device listens for the first arrival message.

[0156] In this step, the edge device 10A, based on the binding relationship between the edge device 10A and the first online site a in the configuration data, only listens for the first arrival message about the first online site a pushed by the manufacturing execution system 13.

[0157] The first arrival message is the arrival message of the product to be inspected arriving at the first online site a for defect inspection. It is used to reflect that there are uninspected products at the first online site a. The arrival message can reflect that a single product to be inspected has not been inspected, or it can reflect that multiple products to be inspected in a certain production batch have not been inspected.

[0158] Step 904: The edge device determines whether it has received the first arrival message.

[0159] In this step, edge device 10A determines whether it has detected the first arrival message.

[0160] If the first arrival message is detected, proceed to step 906;

[0161] If the first arrival message is not detected, proceed to step 902.

[0162] Step 906: Edge device notification creates edge detection task.

[0163] In this step, after listening to the first arrival message, the listener sends a message to the edge detection unit in the edge device 10A to notify the edge detection unit to create an edge detection task.

[0164] Step 908: Parse the online detection file to obtain detection data.

[0165] In this step, the edge detection unit, upon receiving the first arrival message, creates an edge detection task, generates edge task information, accesses the defect file system 15, and obtains the online detection file corresponding to the first arrival message. The online detection file is preprocessed and parsed to obtain detection data, which is then recorded in the edge task information.

[0166] Among them, the online inspection file is the inspection file stored in the defect file system 15. The online inspection file is the inspection file generated by the product inspection equipment 14 and uploaded to the defect file system 15 by the first online site a.

[0167] Step 910: The edge device uses the defect detection model to generate the detection results of the detection data.

[0168] In this step, the edge detection unit inputs the product image from the detection data into the defect detection model in the central device 11. The defect detection model performs defect detection on the product image and generates the detection results. The edge management unit in the edge device 10A sends edge task information and detection results to the central device 11.

[0169] Step 912: The central device monitors edge task information and inspection results.

[0170] In this step, the central device 11 listens to the detection data and inspection results in the edge task information, stores the detection data and results in the database, creates a central detection task, and generates the central task information for the central detection task. The central task information of the central device 11 is consistent with the detection data, detection results, and other data in the edge task information of the edge device 10A.

[0171] Step 914: The central equipment obtains the results of the manual re-inspection.

[0172] In this step, the inspectors perform a re-inspection on the platform provided by the central inspection unit 111 of the central equipment 11, generating a manual re-inspection result on the central equipment side. The central equipment 11 stores the manual re-inspection result on the central equipment side into the database and records it in the central task information. Based on the central task information, the central equipment 11 obtains the manual re-inspection result.

[0173] Step 916: The central equipment determines whether the defect meets the conditions for sending it for repair.

[0174] In this step, if the central equipment 11 determines that there is a defect in the manual re-inspection results, it determines that the product to be tested has a defect and judges whether the defect meets the repair conditions in the configuration information.

[0175] If the defect meets the conditions for sending for repair, then proceed to step 918;

[0176] If the defect does not meet the conditions for repair, proceed to step 920.

[0177] Step 918: The central equipment sends a repair request to the manufacturing execution system.

[0178] In this step, if the defect meets the repair conditions in the configuration information, a repair operation is triggered, and the central management unit 110 sends the repair information to the manufacturing execution system 13.

[0179] Step 920: The central equipment does not need to send a repair request.

[0180] In this step, if the defect does not meet the conditions for sending for repair, the repair operation will not be triggered, and the central management unit 110 does not need to send repair information.

[0181] As described above, the edge detection unit responds to the first arrival message from the first online site by parsing the online detection file corresponding to the first arrival message to obtain detection data. It can obtain the latest product arrival messages and online detection files pushed online in real time, ensuring the real-time performance and accuracy of the detection data.

[0182] In one embodiment, please refer to Figure 10 , Figure 10 A flowchart of a defect detection method for a second online site is shown. (Based on the above...) Figure 7 Taking the second online site b as an example, the defect detection method may include the following steps:

[0183] Step 1002: Receive the second arrival message from the second online site.

[0184] In this step, the central management unit 110 receives the second arrival message from the second online site. This second arrival message indicates that the product to be inspected has arrived at the second online site b for defect inspection, reflecting the presence of uninspected products at the second online site b.

[0185] Step 1004: Determine whether the second online site has a bound edge device.

[0186] In this step, the central management unit 110 checks whether the second online site b has registration information for edge devices and determines whether the second online site b has bound edge devices.

[0187] If it is determined that the second online site b does not have registration information for an edge device, then the second online site b does not have a bound edge device, and proceed to step 1006;

[0188] If it is determined that the second online site b has registration information for an edge device, then the second online site b has a bound edge device, and step 1010 is executed.

[0189] Step 1006: Parse the online detection file to obtain detection data.

[0190] In this step, the central detection unit 111, upon determining that there are no bound edge devices at the second online site, creates a central detection task, accesses the defect file system 15, obtains the online detection file corresponding to the second arrival message, parses the online detection file corresponding to the second arrival message, and obtains the detection data.

[0191] Step 1008: Use the defect detection model to generate the detection results of the detection data.

[0192] In this step, the central inspection unit 111 inputs the product image from the inspection data into the defect detection model in the central device 11. The defect detection model performs defect detection on the product image and generates the inspection results. Based on the inspection results, the central inspection unit 111 determines whether the product to be inspected has any defects.

[0193] Step 1010: Do not process the second arrival message.

[0194] In this step, if the central management unit 110 determines that there is a bound edge device at the second online site, it will not process the second arrival message, and the edge device bound to the second online site will perform defect detection on the product to be tested.

[0195] As described above, the central management unit receives the second arrival message from the second online site. When the central detection unit determines that there is no bound edge device at the second online site, it parses the online detection file corresponding to the second arrival message to obtain detection data. Using the defect detection model, it generates the detection results of the detection data, enabling the central device to handle the detection needs from online sites that are not bound to edge devices. This expands the coverage of the defect detection system and ensures that products from all sites can be detected in a timely manner.

[0196] Figure 11 This is a schematic diagram illustrating the structure of an electronic device according to an exemplary embodiment of this application. The electronic device may be, for example, a cloud device, mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, personal digital assistant, server, smart home appliance, in-vehicle system, etc. (Reference) Figure 11At the hardware level, the electronic device includes a processor 1102, an internal bus 1104, a network interface 1106, memory 1108, and non-volatile memory 1110, and may also include other hardware required for business operations. The processor 1102 reads the corresponding computer program from the non-volatile memory 1110 into the memory 1108 and then runs it, forming an edge device or a central device at the logical level. Of course, in addition to software implementation, this application does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0197] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions, such as a memory including instructions, is also provided, which can be executed by a processor to implement a method performed by an edge device or a method performed by a central device as in any of the above embodiments.

[0198] The non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk and optical data storage device, etc., and this application does not limit it.

[0199] In an exemplary embodiment, a computer program product including a computer program / instructions is also provided, which can be executed by a processor to implement a method executed by an edge device or a method executed by a central device as in any of the above embodiments.

[0200] The foregoing has described specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0201] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention filed herein. This application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and alterations can be made without departing from its scope. The scope of this application is limited only by the appended claims.

[0202] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An edge device, characterized in that, include: Edge detection unit and edge management unit; wherein, The edge detection unit is connected to the edge management unit and is used to obtain the detection data of the product to be detected based on the configuration data of the product to be detected, and to generate the detection result of the detection data using a defect detection model. The edge management unit is used to send the detection results to the central device so that the central device can determine that the product under test has a defect.

2. The edge device according to claim 1, characterized in that, The edge management unit is further configured to: Send a registration request to the central device, the registration request carrying the site identifier corresponding to the edge device; receive the configuration data and the trained defect detection model sent by the central device.

3. The edge device according to claim 1, characterized in that, The edge device is bound to the offline site; The edge detection unit is further configured to: in response to detecting a newly added offline detection file in the offline site, parse the offline detection file to obtain the detection data.

4. The edge device according to claim 1, characterized in that, The edge device is bound to the first online site; The edge detection unit is further configured to: in response to receiving a first arrival message from the first online site, parse the online detection file corresponding to the first arrival message to obtain the detection data; wherein the first arrival message is used to reflect the existence of undetected products to be detected in the first online site.

5. The edge device according to claim 1, characterized in that, The configuration data includes: early warning conditions; The edge detection unit is further configured to: generate warning information for the product to be detected when it is determined that the detection data meets the warning conditions, wherein the warning information is used to increase the priority of detecting the product to be detected.

6. A central device, characterized in that, include: Central testing unit and central management unit; among which, The central management unit is connected to the central detection unit and is used to receive the detection results of the detection data sent by the edge device. The detection results are generated based on the configuration data of the product to be detected and the defect detection model. The central testing unit is used to determine, based on the testing results, that the product under test has a defect.

7. The central equipment according to claim 6, characterized in that, Also includes: A central model service unit; the central model service unit is used to: train an initial model to obtain the trained defect detection model; The central management unit is further configured to: receive a registration request sent by the edge device, the registration request carrying a site identifier corresponding to the edge device; and register the edge device based on the registration request. Based on the site identifier, the configuration data is determined; the configuration data and the defect detection model are sent to the edge device.

8. The central equipment according to claim 6, characterized in that, The central management unit is further configured to: receive a second arrival message from the second online site, the second arrival message being used to reflect the existence of undetected products to be detected in the second online site; The central detection unit is further configured to: parse the online detection file corresponding to the second arrival message to obtain the detection data when it is determined that there is no bound edge device at the second online site; and generate the detection result of the detection data using the defect detection model in the central device.

9. The central equipment according to claim 6, characterized in that, The central detection unit is also used for: The detection data and the detection results are stored to generate the central task information of the product to be tested; Based on the central task information, the manual re-inspection results of the product to be tested are obtained. The manual re-inspection results are generated based on the test data and the test results. If the defect is found in the manual re-inspection results, the product to be tested is determined to have the defect; wherein the defect includes detailed information on the defect location of the product to be tested.

10. The central equipment according to claim 6, characterized in that, The central management unit is also used for: A repair request is sent to the manufacturing execution system corresponding to the product to be inspected. The repair request message is used to reflect detailed information about the defect.

11. A defect detection method, characterized in that, The method is applied to an edge device, the edge device including: an edge detection unit and an edge management unit; The method includes: The edge detection unit obtains the detection data of the product under test based on the configuration data of the product under test, and generates the detection result of the detection data using the defect detection model; The edge management unit sends the detection results to the central device so that the central device can determine that the product under test has a defect.

12. A defect detection method, characterized in that, The method is applied to a central device, which includes a central detection unit and a central management unit. The method includes: The central management unit receives the detection results from the detection data sent by the edge devices. The detection results are generated based on the configuration data of the product to be detected and the defect detection model. Based on the test results, the central testing unit determines that the product under test has a defect.

13. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor executes the executable instructions to implement the method as described in claim 11 or claim 12.

14. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the method as described in claim 11 or claim 12.

Images