An industrial manufacturing intelligent perception system based on light-sensing feature recognition

Abstract

The present invention belongs to the technical field of photosensitive feature recognition, and specifically is an industrial manufacturing intelligent perception system based on photosensitive feature recognition. A photosensitive recognition module is used to collect image information of products through monitoring equipment on a periodic basis, and an image processing module is used to make a preliminary judgment on the products based on the image information of the products in combination with product standard parameters. When it is preliminarily judged that the unqualified rate UZ is ≥ a preset quasi-unqualified rate threshold YL, an adjustment module is used to further analyze the quasi-unqualified products, and the weight compliance rate DL of the quasi-unqualified products is obtained in combination with the standard weight GB of the quasi-unqualified products. The weight compliance rate DL of the quasi-unqualified products is compared and analyzed with the preset weight compliance rate threshold GYL to obtain a maintenance signal. The system can accurately determine the type of equipment failure, and avoid the inability to timely identify and discover the abnormal appearance size of the product caused by the failure of the monitoring equipment. The staff will overhaul and adjust the normally operating processing equipment, which will lead to quality problems in subsequent products.

Description

Technical Field

[0001] The present invention belongs to the technical field of light-sensitive feature recognition, and specifically is an industrial manufacturing intelligent perception system based on light-sensitive feature recognition. Background Art

[0002] Light-sensitive feature recognition has a wide range of applications, such as image processing, machine vision, and security monitoring. By extracting light-sensitive features in images, such as edges, corners, and textures, automatic recognition and positioning of targets can be achieved. At the same time, light-sensitive feature recognition can also be used in the field of security monitoring. By analyzing the light-sensitive features in the surveillance video, abnormal events can be detected and alarmed. In the industrial manufacturing intelligent perception system based on light-sensitive feature recognition, it is possible to determine whether the product meets the processing requirements by performing light-sensitive feature recognition on the product.

[0003] A patent application with publication number CN113427731A discloses a machine vision-based injection molding machine mold detection system and method, which identifies and analyzes the collected mold image information through a detection model; reduces the cost of manual mold detection, and promptly handles exceptions when the mold is damaged or in poor condition, thereby greatly improving the product qualification rate and mold life; the mold monitoring module is used to monitor the working status of the mold in combination with the heartbeat signal and unqualified products; determines whether the mold is abnormal; and assigns corresponding operators to inspect or replace the mold after receiving the mold abnormality signal, thereby improving maintenance efficiency.

[0004] In the above-mentioned prior art, only the appearance dimensions of the product can be identified and monitored through machine vision. When the appearance dimensions of the product are abnormal, it is judged that there is a problem with the processing equipment. However, in the actual production process, the abnormal appearance dimensions of the product caused by the failure of the monitoring equipment cannot be identified and discovered in time, which will cause the staff to inspect and adjust the normally working processing equipment, which will cause quality problems in subsequent products.

[0005] To this end, the present invention provides an industrial manufacturing intelligent perception system based on light-sensitive feature recognition. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0007] The technical solution adopted by the present invention to solve the technical problem is: the system described in the present invention comprises: a data acquisition module: the data acquisition module is used to collect standard product parameters; the standard product parameters include the standard length dimension LB of the product and the standard weight GB of the product;

[0008] Light sensing recognition module: The light sensing recognition module is used to periodically collect image information of products;

[0009] Image processing module: The image processing module obtains the actual length dimension LS of each product based on the product image information, and makes a preliminary judgment on the product in combination with the product standard parameters: the number of unqualified products appearing in the cycle is summarized to obtain the total number of unqualified products in the initial inspection BH. The total number of products inspected in the cycle is ZZ. Through the formula Obtain the quasi-unqualified rate UZ, compare the quasi-unqualified rate UZ with the preset quasi-unqualified rate threshold YL, if the quasi-unqualified rate UZ ≥ the preset quasi-unqualified rate threshold YL, generate an adjustment signal and send the adjustment signal to the adjustment module; if the quasi-unqualified rate UZ < the preset quasi-unqualified rate threshold YL, it indicates that all devices are working normally;

[0010] After receiving the adjustment signal, the adjustment module obtains the weight compliance rate DL of the quasi-unqualified product in combination with the product standard weight GB, compares and analyzes it with the preset weight compliance rate threshold GYL, obtains a maintenance signal, and sends the maintenance signal to the maintenance module;

[0011] The maintenance module dispatches maintenance personnel to perform maintenance on the corresponding faulty equipment according to the type of the maintenance signal.

[0012] Furthermore, the image processing module performs a preliminary judgment on the product, specifically:

[0013] When (LB-M)≤product length dimension LS≤(LB+M), where M is the allowable error of product length dimension and is obtained based on actual experience, the product is marked as a qualified product;

[0014] When the product length dimension LS is less than (LB-M) or the product length dimension LS is greater than (LB+M), the product is marked as a quasi-unqualified product.

[0015] Furthermore, the light sensing recognition module collects image information of the product through the monitoring device in a periodic manner, specifically:

[0016] The monitoring equipment includes an industrial CCD camera, which is installed directly above the product conveyor belt through a bracket. When the product conveyor belt is working normally to transport products, the industrial CCD camera takes high-definition images of each product passing by.

[0017] Furthermore, the image processing module obtains the size parameters of the product according to the image information of the product, specifically:

[0018] The product image information collected by the light sensing recognition detection module is obtained, the product image information is processed, and the length dimension LS of the product is obtained from the image information. The product image information is processed, specifically including filtering, sharpening, mathematical morphological transformation, binarization, edge extraction, and contour extraction commonly used in the prior art to obtain the length dimension LS of the product on the product conveyor belt.

[0019] Furthermore, the adjustment module obtains the weight compliance rate DL of the quasi-unqualified products in combination with the product standard weight GB, specifically:

[0020] After receiving the adjustment signal, the adjustment module measures the weight of the quasi-unqualified product through the gravity sensor and compares it with the product standard weight GB. When (GB-W)≤weight of quasi-unqualified product≤(GB+W), where W is the allowable error value and is obtained based on actual experience, the quasi-unqualified product is marked as a product that meets the weight standard, and the total number of products that meet the weight standard is marked as GD. The weight compliance rate DL is obtained.

[0021] Further, the maintenance signal includes a monitoring equipment maintenance signal and a processing equipment maintenance signal;

[0022] Furthermore, the specific process of acquiring the maintenance signal of the monitoring equipment is as follows:

[0023] If the weight compliance rate DL ≥ the preset weight compliance rate threshold GYL, a monitoring equipment maintenance signal is generated;

[0024] Furthermore, the specific process of obtaining the processing equipment maintenance signal is as follows:

[0025] If the weight compliance rate DL is less than the preset weight compliance rate threshold GYL, a processing equipment maintenance signal is generated; the weight compliance rate DL is the weight compliance rate of products with unqualified length dimensions after preliminary judgment of the product. Products such as steel strips are basically transported and stored in rolls after processing. When different lengths are required for secondary processing of the steel strips, the material, thickness and width dimensions are basically the same, and the factor affecting the weight can only be the length dimension. Therefore, within the allowable error range of the standard length dimension, its weight also has a standard error range. If the product length dimension is unqualified but the weight is within the allowable error range of the product standard weight, and the weight compliance rate DL of the unqualified product is ≥ the preset weight compliance rate threshold GYL, it indicates that the monitoring equipment has a fault, and the processing equipment has no problem, and a monitoring equipment maintenance signal is generated; and when the product length dimension is unqualified but the weight is also within the allowable error range of the product standard weight, and the weight compliance rate DL is less than the preset weight compliance rate threshold GYL, it indicates that the monitoring equipment has no problem, and the processing equipment has a fault, and a processing equipment maintenance signal is generated.

[0026] Further, the maintenance module dispatches maintenance personnel to perform maintenance on the corresponding faulty equipment, specifically as follows:

[0027] According to the maintenance signal, collect the information of maintenance personnel for the corresponding equipment type, summarize the currently on-duty maintenance personnel, and retrieve the maintenance records of the maintenance personnel. The maintenance records include the maintenance time and specific maintenance content.

[0028] Obtain the time difference between the last maintenance completion time of the maintenance personnel and the current time.

[0029] Sort the maintenance personnel in descending order according to the time difference.

[0030] Select the maintenance personnel ranked first for equipment maintenance; by obtaining the information of the currently on-duty maintenance personnel and the time difference information between the last maintenance completion time of the corresponding maintenance personnel and the current time, and selecting the maintenance personnel with the largest time difference for maintenance, the maintenance personnel can be better allocated and have reasonable rest, and maintain a better working state.

[0031] Further, after the maintenance module receives the maintenance signal of the monitoring equipment, obtain the product deviation coefficient PC, and adjust the preset quasi-unqualified rate threshold YL according to the product deviation coefficient PC, specifically as follows:

[0032] Through the formula Obtain the product deviation coefficient PC, where a1, a2, and a3 are coefficient factors.

[0033] Adjust the preset quasi-unqualified rate threshold YL according to the product deviation coefficient PC, specifically as follows:

[0034] Set several adjustment levels and label them as Sb, b = 1, 2,..., 5; and S1 < S2 <... < S5, where b represents the adjustment level serial number; different product deviation coefficients correspond to corresponding adjustment levels.

[0035] Set each adjustment level Sb to correspond to a product deviation coefficient range, specifically: the preset product deviation coefficient range corresponding to S1 is (0, PC1], the preset product deviation coefficient range corresponding to S2 is (PC1, PC2],..., the preset product deviation coefficient range corresponding to S5 is (PC4, PC5], where 0 < PC1 < PC2 <... < PC5, and PC1, PC2, PC3, PC4, PC5 are preset product deviation coefficient thresholds.

[0036] When PC ∈ (PC b-1 , PC b], the adjustment level corresponding to the product deviation coefficient is Sb; the preset quasi-failure rate threshold YL is adjusted according to the obtained adjustment level Sb; the re-planned preset quasi-failure rate threshold XYL is obtained through the formula XYL=YL+Sb.

[0037] The beneficial effects of the present invention are as follows:

[0038] 1. The industrial manufacturing intelligent perception system based on light-sensitive feature recognition described in the present invention collects standard product parameters through a data acquisition module, uses a light-sensitive recognition module to periodically collect image information of products through monitoring equipment, and uses an image processing module to make a preliminary judgment on the products based on the image information of the products combined with the standard parameters of the products: monitors the products according to the set period, reduces the power consumption of the system, and can control the quality of the products at the same time, and performs intelligent perception of the production line based on the detection and monitoring results to detect problems in a timely manner.

[0039] 2. The industrial manufacturing intelligent perception system based on light-sensing feature recognition described in the present invention, after preliminarily judging that the unqualified rate UZ ≥ the preset quasi-unqualified rate threshold YL, further analyzes the quasi-unqualified products through the adjustment module, obtains the weight compliance rate DL of the quasi-unqualified products in combination with the standard weight GB of the quasi-unqualified products, and compares and analyzes with the preset weight compliance rate threshold GYL to obtain a maintenance signal, which can accurately identify the type of equipment failure, and avoid the failure to timely identify and discover the abnormal appearance size of the product caused by the failure of the monitoring equipment. The staff will overhaul and adjust the processing equipment that is working normally, which will lead to quality problems in subsequent products. After the monitoring equipment fails, the maintenance personnel can uniformly adjust the preset quasi-unqualified rate threshold YL according to the adjustment level Sb, and can re-plan the adjustment signal sending conditions to ensure the normal operation of the production line while ensuring the product quality. There is no need to adjust the position of the detection equipment or repair components. The maintenance and adjustment are convenient, and there is no need to stop the production line, thereby improving efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 It is a principle block diagram of an industrial manufacturing intelligent perception system based on light-sensing feature recognition of the present invention. DETAILED DESCRIPTION

[0042] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0043] In the prior art, only the appearance dimensions of the product can be identified and monitored through photosensitive feature recognition. When the appearance dimensions of the product are abnormal, it is judged that there is a problem with the processing equipment. However, in the actual production process, the abnormal appearance dimensions of the product caused by the failure of the monitoring equipment cannot be identified and discovered in time. Products such as steel strips are basically transported and stored in rolls after processing. When different lengths are required for secondary processing of the steel strips, the material, thickness and width dimensions are basically the same. The abnormality of the processing equipment can only be judged by the appearance length dimension, and it cannot be combined with the product weight and other factors. This will cause the staff to inspect and adjust the normally working processing equipment, which will cause quality problems in subsequent products.

[0044] Embodiment 1

[0045] like Figure 1 As shown, an industrial manufacturing intelligent perception system based on light-sensing feature recognition according to an embodiment of the present invention includes:

[0046] The data acquisition module is used to collect standard product parameters and transmit the collected standard product parameters to the image processing module; the standard product parameters include the standard length dimension LB and the standard weight GB of the product;

[0047] The light sensing recognition module is used to collect image information of the product through the monitoring equipment on a periodic basis, and send the image information of the product to the image processing module; specifically, in this embodiment, the period can be set according to the actual situation. For example, with a period of 2 hours, images of 100 products that continuously pass through the monitoring equipment are collected every 2 hours. This can reduce the power consumption of the system while controlling the quality of the product. The production line can be intelligently sensed based on the detection and monitoring results to detect problems in a timely manner.

[0048] The image processing module obtains the length dimension LS of each product based on the product image information, and makes a preliminary judgment on the product in combination with the product standard parameters: the number of unqualified products appearing in the cycle is summarized to obtain the total number of unqualified products in the initial inspection BH, the total number of products inspected in the cycle is ZZ, and the total number of unqualified products in the cycle is ZZ. Obtain the quasi-unqualified rate UZ, compare the quasi-unqualified rate UZ with the preset quasi-unqualified rate threshold YL, if the quasi-unqualified rate UZ ≥ the preset quasi-unqualified rate threshold YL, generate an adjustment signal and send the adjustment signal to the adjustment module; if the quasi-unqualified rate UZ < the preset quasi-unqualified rate threshold YL, it indicates that all devices are working normally;

[0049] After receiving the adjustment signal, the adjustment module obtains the weight compliance rate DL of the quasi-unqualified product in combination with the product standard weight GB, compares and analyzes it with the preset weight compliance rate threshold GYL, obtains a maintenance signal, and sends the maintenance signal to the maintenance module;

[0050] The maintenance module dispatches maintenance personnel to perform maintenance on the corresponding faulty equipment according to the type of the maintenance signal.

[0051] The image processing module in this embodiment performs a preliminary judgment on the product, specifically:

[0052] When (LB-M)≤product length dimension LS≤(LB+M), where M is the allowable error of product length dimension and is obtained based on actual experience, the product is marked as a qualified product;

[0053] When the product length dimension LS is less than (LB-M) or the product length dimension LS is greater than (LB+M), the product is marked as a quasi-unqualified product.

[0054] In this embodiment, the light sensing recognition module collects image information of the product through the monitoring device in a periodic manner, specifically:

[0055] The monitoring equipment includes an industrial CCD camera, which is installed directly above the product conveyor belt through a bracket. When the product conveyor belt is working normally to transport products, the industrial CCD camera takes high-definition images of each product passing by (it is supplemented here that the products are actually completed products, transported on the conveyor belt, and tested).

[0056] In this embodiment, the image processing module obtains the size parameters of the product according to the image information of the product, specifically:

[0057] The product image information collected by the light sensing recognition detection module is obtained, the product image information is processed, and the length dimension LS of the product is obtained from the image information.

[0058] Specifically, the product image information is processed, including filtering, sharpening, mathematical morphological transformation, binarization, edge extraction, and contour extraction commonly used in the prior art to obtain the length dimension LS of the product on the product conveyor belt.

[0059] In this embodiment, the adjustment module obtains the weight compliance rate DL of the quasi-unqualified products in combination with the product standard weight GB, specifically:

[0060] After receiving the adjustment signal, the adjustment module measures the weight of the quasi-unqualified product through the gravity sensor and compares it with the product standard weight GB. When (GB-W)≤weight of quasi-unqualified product≤(GB+W), where W is the allowable error value, the quasi-unqualified product is marked as a product that meets the weight standard, and the total number of products that meet the weight standard is marked as GD. The weight compliance rate DL is obtained.

[0061] Specifically, products such as steel strips are basically transported and stored in coils after processing. When different lengths are required for secondary processing of the steel strips, the material, thickness and width dimensions are basically the same, and the factor that affects the weight can only be the length dimension. Therefore, within the allowable error range of the standard length dimension, its weight also has a standard error range. Products that are initially judged to be unqualified in length dimension are weighed and compared, and the total number of products that meet the weight standard is recorded and marked as GD. Combined with the total number of unqualified products in the initial inspection BH, the weight compliance rate DL is calculated.

[0062] In this embodiment, the maintenance signal includes a monitoring equipment maintenance signal and a processing equipment maintenance signal;

[0063] The specific process of obtaining the maintenance signal of the monitoring equipment is as follows:

[0064] If the weight compliance rate DL ≥ the preset weight compliance rate threshold GYL, a monitoring equipment maintenance signal is generated;

[0065] The specific process of obtaining the processing equipment maintenance signal is as follows:

[0066] If the weight compliance rate DL is less than the preset weight compliance rate threshold GYL, a processing equipment maintenance signal is generated.

[0067] It can be explained that in this embodiment, the weight compliance rate DL is the weight compliance rate of the products with unqualified length dimensions after the preliminary judgment of the products. Products such as steel strips are basically transported and stored in rolls after processing. When different lengths are required for secondary processing of the steel strips, the material, thickness and width dimensions are basically the same, and the factor affecting the weight can only be the length dimension. Therefore, within the allowable error range of the standard length dimension, its weight also has a standard error range. If the length dimension of the product is unqualified but the weight is within the allowable error range of the standard weight of the product, and the weight compliance rate DL of the unqualified product is ≥ the preset weight compliance rate threshold GYL, it indicates that the monitoring equipment has a fault, and the processing equipment has no problem, and a monitoring equipment maintenance signal is generated; and when the length dimension of the product is unqualified but the weight is also within the allowable error range of the standard weight of the product, and the weight compliance rate DL is < the preset weight compliance rate threshold GYL, it indicates that the monitoring equipment has no problem, and the processing equipment has a fault, and a processing equipment maintenance signal is generated, which can accurately determine the type of equipment failure, avoid the failure of the monitoring equipment to cause the product appearance size abnormality to be identified and discovered in time, and the staff will overhaul and adjust the processing equipment that is working normally, which will lead to quality problems in subsequent products.

[0068] After receiving the maintenance signal of the monitoring device, the maintenance module obtains the product deviation coefficient PC and adjusts the preset quasi-failure rate threshold YL according to the product deviation coefficient PC, specifically:

[0069] By formula Obtain the product deviation coefficient PC, where a1, a2, and a3 are coefficient factors and are obtained based on actual experience;

[0070] Adjust the preset quasi-unqualified rate threshold YL according to the product deviation coefficient PC. Specifically:

[0071] Set several adjustment levels and label them as Sb, where b = 1, 2,..., 5; and S1 < S2 <... < S5, where b represents the adjustment level serial number; different product deviation coefficients correspond to corresponding adjustment levels;

[0072] Set that each adjustment level Sb corresponds to a product deviation coefficient range. Specifically: the preset product deviation coefficient range corresponding to S1 is (0, PC1], the preset product deviation coefficient range corresponding to S2 is (PC1, PC2],..., the preset product deviation coefficient range corresponding to S5 is (PC4, PC5], where 0 < PC1 < PC2 <... < PC5, and PC1, PC2, PC3, PC4, and PC5 are preset product deviation coefficient thresholds;

[0073] When PC ∈ (PC b-1 , PC b , then the adjustment level corresponding to the product deviation coefficient is Sb; adjust the preset quasi-unqualified rate threshold YL through the obtained adjustment level Sb to obtain the re-planned preset quasi-unqualified rate threshold XYL; obtain the re-planned preset quasi-unqualified rate threshold XYL through the formula XYL = YL + Sb.

[0074] It can be understood that after the maintenance module receives the monitoring device maintenance signal, it is determined that the monitoring device has a fault. According to actual experience, when the detection device has a fault, it is mostly that there is a problem with the judgment base point during image processing, resulting in the subsequent product length detection data being judged through this incorrect base point. Therefore, all subsequent products have problems. Such problems can be combined with the quasi-unqualified rate UZ, weight compliance rate DL and the corresponding preset quasi-unqualified rate threshold YL, preset weight compliance rate threshold GYL to calculate the corresponding product deviation coefficient PC. And each adjustment level Sb corresponds to a product deviation coefficient range. According to the obtained product deviation coefficient PC, judge the corresponding adjustment level Sb. In actual production, each product length data adjustment level Sb can correspond to a preset quasi-unqualified rate threshold YL adjustment method to obtain the re-planned preset quasi-unqualified rate threshold XYL. The specific value of the adjustment level Sb can be set according to historical experience.

[0075] In one embodiment: Assume that the currently set preset quasi-unqualified rate threshold value YL is 5%, and the currently measured quasi-unqualified rate UZ is 8% due to a failure of the detection equipment, but the quasi-unqualified rate of the actual product should be less than 5%, and according to the system, the quasi-unqualified rate is greater than the preset quasi-unqualified rate threshold value at this time, so an adjustment signal will be issued, and then the weight compliance rate DL and the preset weight compliance rate threshold value GYL are obtained according to the quasi-unqualified product, and the corresponding product deviation coefficient PC and the corresponding adjustment level Sb are obtained in combination with the above parameters, and at this time, Sb is preset to be 3%, and the corresponding preset quasi-unqualified rate threshold value is The adjustment method is (original YL+Sb) to obtain a re-planned preset quasi-failure rate threshold XYL of 8%. At this time, the product meets the processing requirements. When the subsequent measured quasi-failure rate UZ is greater than 5% or less than 8%, no adjustment signal will be issued; after the monitoring equipment fails, the maintenance personnel can make a unified adjustment to the preset quasi-failure rate threshold YL according to the adjustment level Sb, and re-plan the adjustment signal sending conditions to ensure the normal operation of the production line while ensuring product quality. There is no need to adjust the position of the detection equipment or repair components. The maintenance and adjustment are convenient, and there is no need to stop the production line, thereby improving efficiency.

[0076] Embodiment 2

[0077] Comparative Example 1, wherein another embodiment of the present invention is:

[0078] The maintenance module dispatches maintenance personnel to perform maintenance on the corresponding faulty equipment, specifically:

[0079] According to the maintenance signal, the maintenance personnel information of the corresponding equipment type is collected, the maintenance personnel currently on duty are summarized, and the maintenance records of the maintenance personnel are retrieved, wherein the maintenance records include the maintenance time and specific maintenance content;

[0080] Get the time difference between the last time the maintenance personnel completed the maintenance and the current time;

[0081] Sort the maintenance personnel by time difference from largest to smallest;

[0082] The maintenance personnel ranked first will be selected to perform equipment maintenance.

[0083] Specifically, by obtaining the information of the maintenance personnel currently on duty and the time difference between the last time the corresponding maintenance personnel completed maintenance and the current time, the maintenance personnel with the largest time difference are selected for maintenance, so that the maintenance personnel can be better allocated and have a reasonable rest, and maintain a better working state.

[0084] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An industrial manufacturing intelligent perception system based on light-sensing feature recognition, characterized by: The system comprises: a data acquisition module: the data acquisition module is used to collect standard product parameters; the standard product parameters include the product standard length dimension LB and the product standard weight GB; Light sensing recognition module: The light sensing recognition module is used to periodically collect image information of products; Image processing module: The image processing module obtains the actual length dimension LS of each product based on the product image information, and makes a preliminary judgment on the product in combination with the product standard parameters: the number of unqualified products appearing in the cycle is summarized to obtain the total number of unqualified products in the initial inspection BH. The total number of products inspected in the cycle is ZZ. Through the formula Obtain the quasi-unqualified rate UZ, compare the quasi-unqualified rate UZ with the preset quasi-unqualified rate threshold YL, if the quasi-unqualified rate UZ ≥ the preset quasi-unqualified rate threshold YL, generate an adjustment signal and send the adjustment signal to the adjustment module; if the quasi-unqualified rate UZ < the preset quasi-unqualified rate threshold YL, it indicates that all devices are working normally; After receiving the adjustment signal, the adjustment module obtains the weight compliance rate DL of the quasi-unqualified product in combination with the product standard weight GB, compares and analyzes it with the preset weight compliance rate threshold GYL, obtains a maintenance signal, and sends the maintenance signal to the maintenance module; The maintenance module dispatches maintenance personnel to perform maintenance on the corresponding faulty equipment according to the type of the maintenance signal.

2. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 1 is characterized in that: The image processing module performs a preliminary judgment on the product, specifically: When (LB-M)≤actual length dimension LS of the product≤(LB+M), where M is the allowable error of the length dimension of the product and is obtained based on actual experience, the product is marked as a qualified product; When the actual length dimension LS of the product is less than (LB-M) or the actual length dimension LS of the product is greater than (LB+M), the product is marked as a quasi-unqualified product.

3. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 1 is characterized in that: The light sensing recognition module is used to periodically collect image information of products, specifically: The monitoring device is installed directly above the product conveyor belt by means of a bracket. When the product conveyor belt is working normally and conveying the actually completed products, a high-definition image of each product passing under the monitoring device is captured.

4. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 3 is characterized in that: The image processing module obtains the size parameters of the product according to the image information of the product, specifically: The product image information collected by the light sensing recognition detection module is obtained, the product image information is processed, and the actual length dimension LS of the product is obtained from the image information.

5. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 1 is characterized in that: The adjustment module obtains the weight compliance rate DL of the quasi-unqualified products in combination with the product standard weight GB, specifically: After receiving the adjustment signal, the adjustment module measures the weight of the quasi-unqualified product through the gravity sensor and compares it with the product standard weight GB. When (GB-W)≤weight of quasi-unqualified product≤(GB+W), where W is the allowable error value of product weight, the quasi-unqualified product is marked as a product that meets the weight standard, and the total number of products that meet the weight standard is marked as GD. The weight compliance rate DL is obtained.

6. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 5 is characterized in that: The maintenance signal includes a monitoring equipment maintenance signal and a processing equipment maintenance signal.

7. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 6 is characterized in that: The specific process of obtaining the maintenance signal of the monitoring equipment is as follows: If the weight compliance rate DL ≥ the preset weight compliance rate threshold GYL, a monitoring equipment maintenance signal is generated.

8. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 6 is characterized in that: The specific process of obtaining the processing equipment maintenance signal is as follows: If the weight compliance rate DL is less than the preset weight compliance rate threshold GYL, a processing equipment maintenance signal is generated.

9. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 7 is characterized in that: After receiving the maintenance signal of the monitoring device, the maintenance module obtains the product deviation coefficient PC and adjusts the preset quasi-failure rate threshold YL according to the product deviation coefficient PC, specifically: By formula Get the product deviation coefficient PC, where a1, a2, and a3 are coefficient factors; The preset quasi-failure rate threshold YL is adjusted according to the product deviation coefficient PC, specifically: Set several adjustment levels and mark them as Sb, b = 1, 2, ..., 5; and S1 < S2 < ... < S5, where b is the adjustment level number; different product deviation coefficients correspond to corresponding adjustment levels; Each adjustment level Sb corresponds to a range of product deviation coefficients, specifically: the preset range of product deviation coefficients corresponding to S1 is (0, PC1], the preset range of product deviation coefficients corresponding to S2 is (PC1, PC2], ……, the preset range of product deviation coefficients corresponding to S5 is (PC4, PC5], where 0 < PC1 < PC2 < …… < PC5, and PC1, PC2, PC3, PC4, PC5 are preset product deviation coefficient thresholds; When PC∈(PC b-1 , PC b ], the adjustment level corresponding to the product deviation coefficient is Sb; the preset quasi-failure rate threshold YL is adjusted according to the obtained adjustment level Sb; the re-planned preset quasi-failure rate threshold XYL is obtained through the formula XYL=YL+Sb.

10. The industrial manufacturing intelligent perception system based on light-sensing feature recognition according to claim 1 is characterized in that: The maintenance module dispatches maintenance personnel to repair the corresponding faulty equipment, specifically: According to the maintenance signal, collect the information of maintenance personnel for the corresponding equipment type, summarize the currently on-duty maintenance personnel, and retrieve the maintenance records of the maintenance personnel. The maintenance records include the maintenance time and the specific maintenance content; Obtain the time difference between the last maintenance completion time of the maintenance personnel and the current time; Sort the maintenance personnel in descending order of the time difference; Select the maintenance personnel ranked first for equipment maintenance.

Images