A method and system for detecting production quality of LED screens

Abstract

The present invention discloses a method and system for detecting the production quality of an LED screen, which specifically relates to the field of LED screen quality detection, and comprises the following steps: presetting LED screen production quality detection items, and configuring corresponding detection equipment for detection; sampling detection of LED screens that have passed the verification, and judging whether the detection accuracy of the detection equipment is insufficient; if there are hidden dangers, performing an abnormality analysis to obtain a deviation abnormality index and a consistency abnormality index; comprehensively analyzing to generate a comprehensive detection quality coefficient, and dividing the equipment into a low-quality or high-quality operating state; applying a fuzzy reasoning decision optimization method to equipment in a low-quality operating state; the present invention can comprehensively evaluate the detection results of the detection equipment, taking into account the consistency and accuracy of the detection results, and can more realistically and comprehensively reflect the operating state of the detection equipment, maintain high decision-making accuracy and stability in a complex and changeable detection environment, and ensure that the detection equipment is always in the best operating state.

Description

Technical Field

[0001] The present invention relates to the field of LED screen quality inspection, and more specifically, to a method and system for inspecting the production quality of an LED screen. Background Art

[0002] LED screen production quality testing involves comprehensive inspection and evaluation of each component and overall performance of LED displays during production to ensure product quality and reliability. This process includes testing LED chips, driver circuits, display modules, housings, and connectors, examining parameters such as brightness, color consistency, operating voltage, heat dissipation, and durability. These tests can identify and eliminate potential defects and issues during the production process, ensuring the stability and safety of the final product during use.

[0003] With the development of industrial automation and intelligent manufacturing, quality inspection equipment is increasingly being used in production processes. However, existing quality inspection equipment may experience insufficient accuracy or inconsistent test results after long-term operation, thus affecting product quality and production efficiency. Traditional inspection equipment optimization methods often rely on manual experience and simple threshold judgments, which cannot fully and accurately reflect the actual status of the inspection equipment and optimization needs. Therefore, this paper proposes a method and system for LED screen production quality inspection. Summary of the Invention

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for detecting the production quality of an LED screen comprises the following steps:

[0006] Pre-set the inspection items for the production quality of LED screens, and then configure the corresponding inspection equipment according to the pre-set inspection items for the production quality of LED screens, and use the inspection equipment to conduct the inspection work for the corresponding inspection items of the production quality of LED screens;

[0007] Conduct sampling tests on LED screens that have passed the verification of different testing equipment, and make a preliminary judgment based on the sampling test results whether the current testing equipment has the hidden danger of insufficient detection accuracy;

[0008] If the current detection equipment has the hidden danger of insufficient detection accuracy, then the current detection equipment is analyzed for abnormalities to obtain the deviation abnormality index and the consistency abnormality index;

[0009] Based on the deviation anomaly index and consistency anomaly index, a comprehensive analysis is performed to generate a comprehensive detection quality coefficient. The current detection equipment is divided into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient;

[0010] For detection equipment in low-quality operating conditions, determine how it needs to be optimized.

[0011] In a preferred embodiment, when sampling and testing LED screens that have passed the verification of different testing equipment, passing the verification means that the quality test of the testing equipment is passed and it is determined that the quality test is qualified.

[0012] In a preferred embodiment, during sampling inspection, the sampling plan is determined according to the preset sample size and sampling frequency and random sampling is performed, the sampled samples are uniquely identified and the inspection environment is ensured to meet the requirements, the sampling inspection equipment is used to inspect the sampled samples according to the pre-set inspection items, the sampling inspection data is recorded and the sampling inspection data is organized into an analyzable format.

[0013] In a preferred embodiment, the preliminary determination of whether the current detection equipment has a hidden danger of insufficient detection accuracy based on the sampling test results refers to:

[0014] Under a fixed number of sampling times, the sampling detection data of the detection items corresponding to the current target detection device is obtained, and then the average value and standard deviation of the sampling detection data are calculated, and then the average value and standard deviation of the sampling detection data are compared with the preset benchmark average value and standard deviation threshold value respectively. If the average value of the sampling detection data is greater than or equal to the preset benchmark average value and the standard deviation of the sampling detection data is less than or equal to the preset standard deviation threshold value, a normal signal is generated. If the average value of the sampling detection data is greater than or equal to the preset benchmark average value and the standard deviation of the sampling detection data is less than or equal to the preset standard deviation threshold value, an abnormal signal is generated. When the abnormal signal is generated, it indicates that the current target detection device has a hidden danger of insufficient detection accuracy.

[0015] In a preferred embodiment, the logic for obtaining the deviation anomaly index is:

[0016] Within a fixed time window, all detection data of the current target detection equipment are collected to obtain the time series X: X1, X2, ..., Xn; n is the total number of detection data, and the benchmark value data series B: B1, B2, ..., Bm; m is the total number of benchmark value data, and m and n have the same value. The deviation value corresponding to each detection data is calculated. The deviation value is obtained by subtracting the detection data from the benchmark value data with the same sequence number. Then, the absolute deviation value corresponding to each detection data is calculated. The absolute deviation value is the absolute value of the deviation value. Then, normalization is performed, and finally the weighted average method is used to calculate the deviation anomaly index. The calculation formula is: DAI is the deviation anomaly index, |Zi| is the absolute deviation value corresponding to the detection data with sequence number i, and Wi is the proportional coefficient of the absolute deviation value |Zi|.

[0017] In a preferred embodiment, Wi is determined by:

[0018] In a preferred embodiment, the logic for obtaining the consistency anomaly index is:

[0019] Within a fixed time window, all detection data of the current target detection device are collected to obtain the time series X: X1, X2, ..., Xn, where n is the total number of detection data. The LOF algorithm is used to calculate the LOF score of each sample, that is, each detection data. The mean, standard deviation, and absolute median of the LOF scores of all samples are calculated, and then the consistency anomaly index is calculated. The calculation formula is:

[0020] MAD is the absolute median difference, μD is the mean of the LOF scores of all samples, σD is the standard deviation of the LOF scores of all samples, and CAI is the consistency anomaly index.

[0021] In a preferred embodiment, the logic for obtaining the comprehensive detection quality coefficient is:

[0022] Obtain the consistency anomaly index and deviation anomaly index of the current target detection device, and then substitute the consistency anomaly index and deviation anomaly index into the pre-trained machine learning model to obtain the comprehensive detection quality coefficient;

[0023] Classifying the current detection equipment into a low-quality operating state and a high-quality operating state means: comparing the comprehensive detection quality coefficient with the preset standard parameter threshold. If the comprehensive detection quality coefficient is greater than or equal to the preset standard parameter threshold, the current detection equipment is classified as a low-quality operating state. If the comprehensive detection quality coefficient is less than the preset standard parameter threshold, the current detection equipment is classified as a high-quality operating state.

[0024] In a preferred embodiment, for a detection device in a low-quality operating state, the method of determining whether it needs to be optimized is:

[0025] Obtain the consistency anomaly index, deviation anomaly index, and comprehensive detection quality coefficient respectively, apply fuzzy reasoning, define the consistency anomaly index, deviation anomaly index, and comprehensive detection quality coefficient as input variables, define the output variable as the method that needs to be optimized, fuzzify the input variables, and convert the values of the input variables into fuzzy sets, fuzzify the output variables, and convert the output variables into fuzzy sets, formulate fuzzy rules, describe the optimization requirements under different data type combinations, and reason the fuzzified input variables through fuzzy rules to determine the method that needs to be optimized.

[0026] In a preferred embodiment, a LED screen production quality inspection system includes a classification inspection module, a sampling inspection module, an abnormality analysis module, a comprehensive analysis module, and a decision-making module;

[0027] The classification detection module is used to pre-set the detection items of LED screen production quality, and then configure the corresponding detection equipment according to the pre-set detection items of LED screen production quality, and use the detection equipment to detect the corresponding detection items of LED screen production quality;

[0028] The sampling detection module is used to perform sampling detection on LED screens that have passed the verification of different detection equipment, and preliminarily judge whether the current detection equipment has the hidden danger of insufficient detection accuracy based on the sampling detection results;

[0029] The anomaly analysis module is used to perform an anomaly analysis on the current detection equipment that has the hidden danger of insufficient detection accuracy, and obtain the deviation anomaly index and consistency anomaly index;

[0030] The comprehensive analysis module is used to conduct comprehensive analysis based on the deviation anomaly index and the consistency anomaly index to generate a comprehensive detection quality coefficient, and to classify the current detection equipment into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient;

[0031] The decision module is used to determine how to optimize the detection equipment in low-quality operating conditions.

[0032] The technical effects and advantages of the present invention are as follows:

[0033] This method comprehensively evaluates the test results of testing equipment using the consistency anomaly index (CAI), deviation anomaly index (DAI), and comprehensive test quality index (CQI). This method balances the consistency and accuracy of the test results, providing a more realistic and comprehensive reflection of the operating status of the testing equipment. Using fuzzy reasoning, the CAI, DAI, and CQI are used as input variables, and the optimization method (maintenance or replacement) is defined as the output variable. Through fuzzy processing and fuzzy rule reasoning, accurate decisions on the optimization method for testing equipment are made, avoiding the limitations of single threshold judgments in traditional methods.

[0034] By comprehensively considering various abnormal situations (such as deviations and consistency issues in detection results), the present invention can maintain high decision-making accuracy and stability in complex and changeable detection environments, and is suitable for optimizing quality inspection equipment in various industrial scenarios.

[0035] This invention accurately determines the optimal approach for testing equipment, enabling timely maintenance or replacement, ensuring optimal operation of testing equipment at all times. This reduces equipment failures and downtime, and improves overall production efficiency and product quality. This invention utilizes a data-driven, intelligent optimization decision-making approach, reducing reliance on manual experience and enhancing the intelligence of testing equipment management, resulting in more efficient equipment management and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0037] Figure 1 This is a schematic diagram of a method for detecting the production quality of an LED screen in the present invention.

[0038] Figure 2 This is a schematic diagram of a LED screen production quality inspection system in the present invention. DETAILED DESCRIPTION

[0039] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0040] Reference Figure 1 - Figure 2 The following examples were obtained:

[0041] Example 1: A method for detecting the production quality of an LED screen, comprising the following steps:

[0042] Pre-set the inspection items for the production quality of LED screens, and then configure the corresponding inspection equipment according to the pre-set inspection items for the production quality of LED screens, and use the inspection equipment to conduct the inspection work for the corresponding inspection items of the production quality of LED screens;

[0043] Conduct sampling tests on LED screens that have passed the verification of different testing equipment, and make a preliminary judgment based on the sampling test results whether the current testing equipment has the hidden danger of insufficient detection accuracy;

[0044] If the current detection equipment has the hidden danger of insufficient detection accuracy, then the current detection equipment is analyzed for abnormalities to obtain the deviation abnormality index and the consistency abnormality index;

[0045] Based on the deviation anomaly index and consistency anomaly index, a comprehensive analysis is performed to generate a comprehensive detection quality coefficient. The current detection equipment is divided into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient;

[0046] For detection equipment in low-quality operating conditions, determine how it needs to be optimized.

[0047] In the sampling inspection of LED screens that have passed the verification of different testing equipment, verification and passing refers to the products that have passed the quality inspection of the testing equipment and are judged to be qualified in the quality inspection and thus retained.

[0048] During sampling testing, a sampling plan is determined based on the pre-set sample size and sampling frequency, and random sampling is conducted. Samples are uniquely identified, and the testing environment ensures compliance with requirements. Samples are tested using sampling equipment according to pre-set test items, and the sampling test data is recorded and organized into an analyzable format. More specifically, the following steps are performed: Determine the sampling plan; Determine the sample size: The sample size for each sampling session is determined based on the size of the production batch. This can typically be calculated using industry standards or statistical methods to ensure statistical significance. Determine the sampling frequency: The sampling frequency is determined based on the production pace and quality control requirements. For example, 10 LED screens may be sampled for testing for every 100 produced. Random sampling: Randomness: Using random sampling ensures that the sampling process is not biased towards products from specific batches or time periods. Random number tables or computer-generated random numbers can be used to determine sampling points. Sample identification: Uniquely identify the sampled samples, such as with numbers or labels, for subsequent tracking and record-keeping. Data recording: Record the basic information of each sample in detail, such as production date, batch number, test time, etc. Preparation of the test environment: Ensure that the test environment meets the requirements to avoid external factors affecting the test results. For example, temperature, humidity, light, etc. should remain stable. Equipment calibration: Before each test, calibrate the test equipment to ensure that it is in the best working condition. Testing process: According to the pre-set test items and standards, use the test equipment to conduct a comprehensive test of the sampled samples, including but not limited to brightness uniformity, color consistency, operating voltage and heat dissipation performance. Data collection: Accurately record the test data of each sample to ensure that the data is complete and accurate. Data collation: Organize the sampling test data into an analyzable format in preparation for subsequent statistical analysis.

[0049] Based on the sampling test results, it is preliminarily judged whether the current testing equipment has the hidden danger of insufficient detection accuracy:

[0050] At a fixed sampling frequency, sampled test data corresponding to the test item of the current target detection device is obtained. The mean and standard deviation of the sampled test data are then calculated and compared with a preset baseline mean and standard deviation threshold, respectively. If the mean of the sampled test data is greater than or equal to the preset baseline mean and the standard deviation of the sampled test data is less than or equal to the preset standard deviation threshold, a normal signal is generated. If the mean of the sampled test data is not greater than or equal to the preset baseline mean and the standard deviation of the sampled test data is less than or equal to the preset standard deviation threshold, an abnormal signal is generated. The generation of an abnormal signal indicates that the current target detection device has a potential problem of insufficient detection accuracy. By comparing the mean and standard deviation of the sampled test data with the preset baseline and threshold values, it is possible to quickly determine whether the current detection device has insufficient detection accuracy. A normal signal indicates that the device's detection results are reliable, the device's accuracy is sufficient, and it can continue to be used. An abnormal signal indicates that the device's detection results may be unreliable, the device's accuracy may be insufficient, and further calibration or replacement of the device is required to ensure the accuracy of the test results. Promptly identifying and addressing insufficient detection equipment accuracy can prevent product quality problems caused by inaccurate detection, thereby improving product consistency and reliability. By regularly conducting sampling tests and promptly correcting equipment, you can maintain efficient equipment operation, reduce production stoppages and rework caused by equipment problems, and improve overall production efficiency. This method is simple, intuitive, and effective, and can serve as an important tool in production quality control. It can systematically determine whether the accuracy of testing equipment is sufficient, ensure the reliability of test results, and thus ensure the stability and consistency of product quality.

[0051] The logic for obtaining the deviation anomaly index is as follows: within a fixed time window, all detection data of the current target detection equipment are collected to obtain the time series X: X1, X2, ..., Xn; n is the total number of detection data, and the benchmark value data series B: B1, B2, ..., Bm is obtained; m is the total number of benchmark value data, and m and n have the same value. The benchmark value data series B can be detected and determined by a higher-precision device, and the deviation value corresponding to each detection data is calculated. The deviation value is obtained by subtracting the detection data from the benchmark value data with the same sequence number, and then the absolute deviation value corresponding to each detection data is calculated. The absolute deviation value is the absolute value of the deviation value, and then normalization is performed. Finally, the weighted average method is used to calculate the deviation anomaly index. The calculation formula is: DAI stands for the Deviation Anomaly Index, |Zi| is the absolute deviation value corresponding to the test data with sequence number i, and Wi is the proportional coefficient of the absolute deviation value |Zi|. The Deviation Anomaly Index measures the degree of deviation in the test data of the testing equipment. A larger DAI indicates a greater deviation between the test results of the testing equipment and the reference value, and a lower detection accuracy. Conversely, a smaller DAI indicates a smaller deviation between the test results of the testing equipment and the reference value, and a higher detection accuracy.

[0052] Wi is determined as follows: It can be assigned using a Gaussian distribution strategy: Wi is the proportional coefficient, or weight, of the absolute deviation value |Zi|, reflecting the importance of each data point in calculating the deviation anomaly index. Weights are assigned using a Gaussian distribution strategy, inversely proportional to the square of the standardized absolute deviation value Zi. Specifically, the larger the standardized absolute deviation value, the greater the deviation of the corresponding data point, and the smaller its weight Wi; the smaller the standardized absolute deviation value, the smaller the deviation of the corresponding data point, and the larger its weight Wi. This weight distribution method can better balance the influence of data points with large and small deviations in the dataset in the deviation anomaly index calculation, making the calculation result more reflective of the deviation of the overall data.

[0053] The logic for obtaining the consistency anomaly index is:

[0054] Within a fixed time window, all detection data of the current target detection device are collected to obtain the time series X: X1, X2, ..., Xn, where n is the total number of detection data. The LOF algorithm is used to calculate the LOF score of each sample, that is, each detection data. The LOF score measures the local density of each sample relative to the density of its neighbors and is expressed as: Where lrd(i) represents the local reachability density of the i-th sample, which is defined as: N k (i) represents the k nearest neighbor set of the i-th sample, reach-disk k (i, j) represents the reachable distance from the i-th sample to the j-th sample, defined as: reach-disk k (i, j) = max(k-disk(i, j), d(i, j)); k-disk(i, j) represents the k distance of the j-th sample, that is, the distance from the j-th sample to its k-th neighbor, d(i, j) represents the Euclidean distance between the i-th sample and the j-th sample, and the mean, standard deviation, and absolute median of the LOF scores of all samples are calculated. The calculation formulas for the mean, standard deviation, and absolute median are prior art and will not be elaborated on here. Then, the consistency anomaly index is calculated. The calculation formula is: MAD is the median absolute deviation, μD is the mean LOF score of all samples, σD is the standard deviation of the LOF scores of all samples, and CAI is the consistency anomaly index, which is used to measure the consistency of the test results of the detection device. A larger CAI indicates inconsistent test results, more outliers, and poorer consistency of the detection device. Conversely, a smaller CAI indicates better consistency of test results, fewer outliers, and higher consistency of the detection device.

[0055] The logic for obtaining the comprehensive detection quality coefficient is:

[0056] Obtain the consistency anomaly index and deviation anomaly index of the current target detection device, and then substitute the consistency anomaly index and deviation anomaly index into the pre-trained machine learning model to obtain the comprehensive detection quality coefficient; the machine learning model is not specifically limited here, and any model that can comprehensively analyze the consistency anomaly index and the deviation anomaly index to generate a comprehensive detection quality coefficient is acceptable. In order to implement the technical solution of the present invention, the present invention provides a specific implementation method: CQI=q1*DAI+q2*CAI; CQI is a comprehensive detection quality coefficient, which reflects the overall detection quality of the detection equipment. The larger the value, the worse the overall quality of the detection equipment. q1 and q2 are both preset proportional coefficients, which are used to adjust the contribution ratio of the consistency anomaly index and the deviation anomaly index in the comprehensive detection quality coefficient, and can be adjusted according to actual conditions.

[0057] Classifying the current detection equipment into a low-quality operating state and a high-quality operating state means: comparing the comprehensive detection quality coefficient with the preset standard parameter threshold. If the comprehensive detection quality coefficient is greater than or equal to the preset standard parameter threshold, the current detection equipment is classified as a low-quality operating state. If the comprehensive detection quality coefficient is less than the preset standard parameter threshold, the current detection equipment is classified as a high-quality operating state.

[0058] For testing equipment with low-quality operation, the way to decide whether it needs to be optimized is:

[0059] Obtain the consistency anomaly index (CAI), deviation anomaly index (DAI), and comprehensive inspection quality coefficient, respectively. Apply fuzzy reasoning and define the consistency anomaly index, deviation anomaly index, and comprehensive inspection quality coefficient as input variables. Define the output variable as the optimization method to be optimized. Fuzzify the input variables, converting their values into fuzzy sets. Fuzzy rules are then formulated to describe the optimization requirements for different data type combinations. Fuzzy rules are then used to reason about the fuzzified input variables to determine the optimization method to be optimized. Input variables include the consistency anomaly index (CAI), deviation anomaly index (DAI), and comprehensive inspection quality coefficient (CQI). Output variables include the optimization method to be optimized: maintenance or replacement. Fuzzy processing of the input variables: converting the input variable values into fuzzy sets: CAI (low, medium, high); DAI (low, medium, high); and CQI (low, medium, high). Converting the output variable into a fuzzy set: optimization method: maintenance or replacement. Fuzzy rules are then formulated based on different input variable combinations. For example, if the CAI is high, the DAI is high, and the CQI is high, then the optimization method is replacement. If the CAI is medium, the DAI is medium, and the CQI is high, the optimization approach is replacement. If the CAI is low, the DAI is low, and the CQI is low, the optimization approach is maintenance. If the CAI is medium, the DAI is low, and the CQI is medium, the optimization approach is maintenance. Fuzzy reasoning: Fuzzy rules are used to reason about the fuzzified input variables to determine the optimization approach. For each input variable, its membership in each fuzzy set is calculated. Based on the fuzzy rules, the membership of the output variable is calculated. The output variable is defuzzified using the maximum membership method or the center of gravity method to obtain the final optimization decision.

[0060] Example 2: A LED screen production quality inspection system, including a classification inspection module, a sampling inspection module, an abnormality analysis module, a comprehensive analysis module, and a decision module;

[0061] The classification detection module is used to pre-set the detection items of LED screen production quality, and then configure the corresponding detection equipment according to the pre-set detection items of LED screen production quality, and use the detection equipment to detect the corresponding detection items of LED screen production quality;

[0062] The sampling detection module is used to perform sampling detection on LED screens that have passed the verification of different detection equipment, and preliminarily judge whether the current detection equipment has the hidden danger of insufficient detection accuracy based on the sampling detection results;

[0063] The anomaly analysis module is used to perform an anomaly analysis on the current detection equipment that has the hidden danger of insufficient detection accuracy, and obtain the deviation anomaly index and consistency anomaly index;

[0064] The comprehensive analysis module is used to conduct comprehensive analysis based on the deviation anomaly index and the consistency anomaly index to generate a comprehensive detection quality coefficient, and to classify the current detection equipment into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient;

[0065] The decision module is used to determine how to optimize the detection equipment in low-quality operating conditions.

[0066] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0067] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0068] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0069] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0070] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for detecting the production quality of LED screens, characterized in that: The following steps are involved: Pre-set the inspection items for the production quality of LED screens, and then configure the corresponding inspection equipment according to the pre-set inspection items for the production quality of LED screens, and use the inspection equipment to conduct the inspection work for the corresponding inspection items of the production quality of LED screens; Conduct sampling tests on LED screens that have passed the verification of different testing equipment, and make a preliminary judgment based on the sampling test results whether the current testing equipment has the hidden danger of insufficient detection accuracy; If the current detection equipment has the hidden danger of insufficient detection accuracy, then the current detection equipment is analyzed for abnormalities to obtain the deviation abnormality index and the consistency abnormality index; Based on the deviation anomaly index and consistency anomaly index, a comprehensive analysis is performed to generate a comprehensive detection quality coefficient. The current detection equipment is divided into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient; For testing equipment in low-quality operating conditions, determine how to optimize it; The logic for obtaining the deviation anomaly index is: In a fixed time window, all detection data of the current target detection device are collected to obtain the time series X: X1, X2,…, X n ; n is the total number of test data, and obtains the reference value data sequence B: B1, B2, ..., B m ; m is the total number of reference value data, m and n have the same value, calculate the deviation value corresponding to each test data, the deviation value is obtained by subtracting the test data from the reference value data with the same serial number, and then calculate the absolute deviation value corresponding to each test data. The absolute deviation value is the absolute value of the deviation value, and then perform normalization processing. Finally, use the weighted average method to calculate the deviation anomaly index. The calculation formula is: DAI is the deviation anomaly index, |Zi| is the absolute deviation value corresponding to the detection data with sequence number i, and Wi is the proportional coefficient of the absolute deviation value |Zi|; Wi is determined by: The logic for obtaining the consistency anomaly index is: In a fixed time window, all detection data of the current target detection device are collected to obtain the time series X: X1, X2,…, X n ; n is the total number of test data. The LOF algorithm is used to calculate the LOF score of each sample, that is, each test data. The mean, standard deviation, and absolute median of the LOF scores of all samples are calculated, and then the consistency anomaly index is calculated. The calculation formula is: MAD is the absolute median difference, μD is the mean of the LOF scores of all samples, σD is the standard deviation of the LOF scores of all samples, and CAI is the consistency anomaly index; For testing equipment with low-quality operation, the way to decide whether it needs to be optimized is: Obtain the consistency anomaly index, deviation anomaly index, and comprehensive detection quality coefficient respectively, apply fuzzy reasoning, define the consistency anomaly index, deviation anomaly index, and comprehensive detection quality coefficient as input variables, define the output variable as the method that needs to be optimized, fuzzify the input variables, and convert the values of the input variables into fuzzy sets, fuzzify the output variables, and convert the output variables into fuzzy sets, formulate fuzzy rules, describe the optimization requirements under different data type combinations, and reason the fuzzified input variables through fuzzy rules to determine the method that needs to be optimized.

2. A method for detecting the production quality of an LED screen according to claim 1, characterized in that: In the sampling inspection of LED screens that have passed the verification of different testing equipment, verification and passing means that they have passed the quality inspection of the testing equipment and have been judged to be qualified in the quality inspection.

3. A method for detecting the production quality of an LED screen according to claim 2, characterized in that: During sampling testing, the sampling plan is determined based on the preset sample size and sampling frequency, and random sampling is carried out. The samples taken are uniquely identified and the testing environment is ensured to meet the requirements. According to the pre-set test items, the sampling samples are tested using sampling testing equipment, and the sampling test data is recorded and organized into an analyzable format.

4. A method for detecting the production quality of an LED screen according to claim 3, characterized in that: Based on the sampling test results, it is preliminarily judged whether the current testing equipment has the hidden danger of insufficient detection accuracy: Under a fixed number of sampling times, the sampling detection data of the detection items corresponding to the current target detection device is obtained, and then the average value and standard deviation of the sampling detection data are calculated, and then the average value and standard deviation of the sampling detection data are compared with the preset benchmark average value and standard deviation threshold value respectively. If the average value of the sampling detection data is greater than or equal to the preset benchmark average value and the standard deviation of the sampling detection data is less than or equal to the preset standard deviation threshold value, a normal signal is generated. If the average value of the sampling detection data is greater than or equal to the preset benchmark average value and the standard deviation of the sampling detection data is less than or equal to the preset standard deviation threshold value, an abnormal signal is generated. When the abnormal signal is generated, it indicates that the current target detection device has a hidden danger of insufficient detection accuracy.

5. A method for detecting the production quality of an LED screen according to claim 4, characterized in that: The logic for obtaining the comprehensive detection quality coefficient is: Obtain the consistency anomaly index and deviation anomaly index of the current target detection device, and then substitute the consistency anomaly index and deviation anomaly index into the pre-trained machine learning model to obtain the comprehensive detection quality coefficient; Classifying the current detection equipment into a low-quality operating state and a high-quality operating state means: comparing the comprehensive detection quality coefficient with the preset standard parameter threshold. If the comprehensive detection quality coefficient is greater than or equal to the preset standard parameter threshold, the current detection equipment is classified as a low-quality operating state. If the comprehensive detection quality coefficient is less than the preset standard parameter threshold, the current detection equipment is classified as a high-quality operating state.

6. A LED screen production quality inspection system, based on the LED screen production quality inspection method according to any one of claims 1 to 5, characterized in that: Including classification detection module, sampling detection module, anomaly analysis module, comprehensive analysis module, and decision-making module; The classification detection module is used to pre-set the detection items of LED screen production quality, and then configure the corresponding detection equipment according to the pre-set detection items of LED screen production quality, and use the detection equipment to detect the corresponding detection items of LED screen production quality; The sampling detection module is used to perform sampling detection on LED screens that have passed the verification of different detection equipment, and preliminarily judge whether the current detection equipment has the hidden danger of insufficient detection accuracy based on the sampling detection results; The anomaly analysis module is used to perform an anomaly analysis on the current detection equipment that has the hidden danger of insufficient detection accuracy, and obtain the deviation anomaly index and consistency anomaly index; The comprehensive analysis module is used to perform comprehensive analysis based on the deviation anomaly index and the consistency anomaly index to generate a comprehensive detection quality coefficient, and to classify the current detection equipment into a low-quality operation state and a high-quality operation state according to the comprehensive detection quality coefficient; The decision module is used to determine how to optimize the detection equipment in low-quality operating conditions.

Images