Intelligent electronic equipment program control test system based on quality monitoring information

Abstract

The invention relates to an electrical performance and electrical fault testing device. In particular to an intelligent electronic equipment program control testing system based on quality monitoring information, which comprises a collecting unit, a quality monitoring unit, a testing unit, a testing unit and a testing unit, and is characterized in that the collecting unit is used for automatically collecting the quality monitoring information through various channel sources and storing the quality monitoring information into a storage library; the extraction processing unit is used for processing the information in the storage library and performing feature extraction; the test rule generation unit is used for designing a program control test rule by utilizing the extracted implementation characteristics; the connection test unit is used for respectively connecting the test rule and the intelligent electronic equipment with the test host and carrying out program control test; and the feedback unit is used for collecting test results of each stage of the electronic equipment and sending the test results to the monitoring terminal for display. According to the invention, by processing the collected quality monitoring information, the accuracy and reliability of the measured data can be improved.

Description

Technical Field

[0001] The invention relates to a testing device for electrical performance and electrical faults; in particular, to a program-controlled testing system for intelligent electronic equipment based on quality monitoring information. Background Art

[0002] Intelligent electronic devices are devices that can sense, process, interact and manage information, and can achieve autonomous learning and self-upgrade. With the development of artificial intelligence and Internet of Things technologies, more and more traditional electronic devices are gradually transforming into intelligent ones, and quality testing of intelligent electronic devices is required during the research and development process.

[0003] Testing of intelligent electronic devices is very important for ensuring product quality, improving customer satisfaction, reducing maintenance costs and ensuring information security. The main purpose of testing intelligent electronic devices is to ensure that the equipment can work stably, reliably and efficiently, and will not bring negative impacts or safety hazards to users. Through testing, problems and defects in products can be detected, and the design and development process can be adjusted in time to improve product quality and performance. Before the product is officially released, discovering and solving problems can avoid the cost of later repairs and maintenance of products, and reduce the possibility of customer complaints and returns. Good testing can ensure that product functions and performance meet the expected level and meet customer needs and expectations, thereby improving user experience and satisfaction. During the testing process, security issues of the device, such as data leakage, network attacks, etc., can be detected, and measures can be taken to strengthen them to ensure the security and privacy of user information.

[0004] The existing program-controlled test method usually connects devices with different test contents to the electronic device to be tested, and then sequentially inputs the test instructions of the test device into the electronic device for testing.

[0005] The Chinese patent with the announcement number CN113468048B discloses a system testing method, device, equipment and computer-readable storage medium, which specifically discloses: receiving an input test instruction and determining a test task corresponding to the test instruction; controlling a preset mechanical arm to perform a simulation test according to the test task, and obtaining a test picture corresponding to the simulation test; comparing the test picture with a preset standard picture, and determining the test result of the simulation test based on the comparison result. The present invention improves the intelligence of system testing.

[0006] The Chinese patent with publication number CN109614330A discloses a storage system service test method, device, system, storage controller and medium; specifically discloses: including: controlling a server to receive and identify basic test data; wherein the basic test data includes: LUN setting parameters, test cases; calling a LUN creation tool to create a LUN in the storage server to be tested according to the LUN setting parameters; creating a link between each test server and the storage server to be tested; sending a corresponding test case to the test server, controlling the test server to interact with the tested storage server for service data; collecting working status data during the service data interaction process to obtain service detection data; generating test results according to the service detection data. This method can realize the automation of the storage system test process, thereby improving the test efficiency.

[0007] A Chinese patent with publication number CN105955878A discloses a server-side testing method and system; specifically disclosed are: intercepting test request feedback to a test terminal; parsing the test request feedback to determine the test request; determining the expected operation result based on the determined test request using a rule engine library, wherein the rule engine library includes each test request and each corresponding expected operation result; comparing the test result generated by executing the test task corresponding to the test request with the expected operation result, and generating a test report. The present disclosure also provides a server-side testing system. The present disclosure can intercept the test on the server side and generate a test report, thereby overcoming the problems of inefficiency, poor accuracy, and the need to invade the test end to damage the test end caused by the tester's manual analysis of log files. The operation is simple and convenient, and the test automation is realized.

[0008] If there is no testing standard during the testing of the above test methods, the test results may be inaccurate. Since there is no reliable test specification or process, there will be certain differences in the methods used by each experimenter during the test, which will lead to inaccurate test data. If there is no testing standard, then due to various reasons, including the equipment used, operating procedures and environment, etc., they cannot be accurately defined, making it difficult to compare and communicate the test results. In short, the absence of testing standards during testing will bring various inconveniences to enterprises, consumers and society, and even cause quality problems and disputes, and hinder the improvement and optimization of products or services. Therefore, it is very necessary and important to establish testing standards on the testing production line.

[0009] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0010] In view of the deficiencies in the prior art, the present invention provides an intelligent electronic device program-controlled testing system based on quality monitoring information, which has the advantage of high test quality and thus solves the problem that errors may occur in test results.

[0011] In order to achieve the above advantages of high test quality, the specific technical solutions adopted by the present invention are as follows: An intelligent electronic device program-controlled test system based on quality monitoring information, characterized in that the system is connected to a test unit, a collection unit, an extraction processing unit, a test rule generation unit and a feedback unit; The connection test unit is used to connect to the signal port of the intelligent electronic device and send a time-domain current signal of a preset waveform to the signal port; The collecting unit is used to collect the feedback current signal of the signal port of the intelligent electronic device, and store the feedback current signal in a storage repository in the form of a time domain waveform; The extraction processing unit is used to perform clutter filtering and feature extraction on the feedback current signal of the signal port of the intelligent electronic device in the storage library; The test rule generating unit is used to generate a program-controlled test rule according to the extracted implementation characteristics of the current variation of the signal port of the intelligent electronic device; The feedback unit is used to collect the test results of the electronic equipment at each stage and send the test results to the monitoring terminal for display.

[0012] Preferably, the extraction processing unit includes a generation module, an analysis module and a synthesis module; A generation module, used for generating a difference map between two adjacent sets of quality monitoring information; An analysis module, used for analyzing and processing the same regions and different regions between two sets of difference images; The synthesis module is used to synthesize features from the difference map analysis processing results to obtain implementation features.

[0013] Preferably, the step of generating a difference map between two adjacent sets of quality monitoring information comprises: S11, pre-processing two adjacent sets of quality monitoring information collected from multiple channels; S12, performing visualization imaging processing on the preprocessed information to obtain two two-dimensional images of the same monitoring scene; S13, constructing two initial difference maps through the two two-dimensional images, which are a difference map and a logarithmic ratio map; S14, performing Fourier transform on the difference map and the logarithmic ratio map respectively to obtain the amplitude and phase of each initial difference map; S15, combining the phase of the difference map and the amplitude of the logarithmic ratio map, and generating a new difference map through inverse Fourier transform.

[0014] Preferably, the step of analyzing and processing the same regions and different regions between the two sets of difference maps comprises: S21, using a generation module to obtain two sets of difference maps; S22, respectively calculating the difference in the grayscale values ​​of the pixels corresponding to the two sets of difference images, and performing normalization processing to obtain the grayscale matrix of the difference image; S23, respectively calculating the quotient logarithms of the grayscale values ​​of the pixels corresponding to the two sets of difference images, and performing normalization processing to obtain the grayscale matrix of the ratio image; S24, calculating the local variance value of the gray value of each pixel in the difference map, and performing threshold segmentation to obtain the optimal threshold; S25, according to the variance matrix and the optimal threshold, the difference difference map and the ratio difference map are fused to obtain the gray value of each pixel of the fused final difference map; S26. Perform FCM two-cluster segmentation on the fused image to obtain the same area and the different area.

[0015] Preferably, the gray value difference is calculated as follows: In the formula, is the difference in grayscale values, is a pixel, and They are the first group of difference maps and the second group of difference maps respectively; The first set of difference maps at pixel points The gray value of The second set of difference maps at pixel points The gray value of .

[0016] Preferably, the calculation formula for the normalization process is: In the formula, is the difference map S at the pixel point The gray value of is the matrix generated by subtracting the grayscale of the corresponding pixels of the first set of difference maps from the second set of difference maps, is the difference in grayscale values, is the maximum value of the matrix, is the minimum value of the matrix.

[0017] Preferably, the step of synthesizing features from the difference map analysis results to obtain implementation features comprises: S31, establish a DOG scale space pyramid model, and compare each point in the DOG scale space with the points of adjacent scales and adjacent positions one by one, and the obtained local extreme points are the feature points; S32, deleting low-end comparison pairs and unstable edge response points in feature points; S33, deleting low-contrast extreme points and unstable edge response points according to the improved threshold value to obtain accurate feature points; S34, calculate and obtain the standard value corresponding to each group of precise points. If the standard value is less than the threshold, it is low contrast and is filtered out. Otherwise, it is retained. S35, generating feature descriptors for the retained feature points, and using the feature descriptors of the first new difference image and the second new difference image to determine whether the feature points corresponding to the two images are matching points, and if so, determining them as implementation features.

[0018] Preferably, the test rule generation unit includes a formulation module and a construction module; The formulation module is used to formulate program control test rules according to the implementation characteristics and performance test requirements obtained through analysis and processing; The construction module is used to complete the construction of the performance model according to the performance test requirements and relevant information of the production system, and guide the performance test process and the generation of test results.

[0019] Preferably, formulating program-controlled test rules according to the implementation characteristics and performance test requirements obtained through analysis and processing includes: Attribute constraints are performed based on performance tests, and redundant attributes are deleted to form a simplified attribute set; Calculate the importance of each attribute in the reduced attribute set and select the largest attribute as the splitting attribute; Repeat the above steps until the decision attributes of each category are consistent, generating a leaf node; Generate program-controlled rules through reading, and calculate the confidence and support of the rules to obtain the rule set; Verify whether the obtained rule set complies with the program-controlled test rules. If not, repeat the above steps. Otherwise, directly output the rule set.

[0020] Preferably, the construction of the performance model is completed according to the performance test requirements and the relevant information of the production system, and the generation of the performance test process and the test results is guided, including: The results of the program control rule set, performance test requirements and relevant information of the production system are combined and divided into two categories: normal data set and mixed data set. At the same time, the association analysis algorithm is used to obtain the association patterns of the normal data set and the mixed data set respectively; Based on the association algorithm, the sequence analysis algorithm is used to calculate the sequence rules of the normal data set and the mixed data set; The sequence rule coding of the normal data set and the mixed data set are obtained by using the sequence rule coding algorithm, and the encoded patterns are compared to obtain the sequence coding pattern that conforms to the program control rules; Apply statistical feature construction algorithms to sequence coding patterns to construct statistical feature attributes that can clearly distinguish normal and abnormal behaviors; Add statistical feature attributes to the training data set and the test data set and calculate all feature attribute values. Post-process the obtained data files and convert them into a data format that can be recognized by the classification algorithm. Finally, generate a classification detection model based on the classification algorithm.

[0021] Compared with the prior art, the present invention provides an intelligent electronic device program-controlled test system based on quality monitoring information, which has the following beneficial effects: (1) The present invention provides a program-controlled test system for intelligent electronic devices based on quality monitoring information, which can automate and optimize the test process, improve test efficiency, and avoid the influence of human factors on the test results. By analyzing the quality monitoring information, key indicators and performance parameters can be obtained to determine the test requirements and test objectives, and coverage testing of functional modules, data structures and logical algorithms in more specific scenarios can be achieved.

[0022] (2) The present invention can also remove irrelevant or fully verified test cases based on quality monitoring information, avoid repeated testing and redundant testing, and make the testing work more centralized and effective. In addition, the program-controlled test has high reliability and consistency, which can greatly reduce the test error rate, and has the advantages of automated detection and verification, thereby ensuring the accuracy and reliability of the test process and test results. The program-controlled test based on quality monitoring information can promote the implementation of continuous testing based on real-time monitoring data and test reports, periodically evaluate and feedback the performance of the product, and provide timely feedback and improvements on the problems found, making the test a continuously optimized work process.

[0023] (3) The present invention processes the collected quality monitoring information and extracts rules that meet quality standards from the information as standards for testing intelligent electronic devices, thereby improving the accuracy and reliability of measurement data, saving a lot of testing time and costs, helping to ensure the accuracy and reliability of test results, and reducing the test error rate, thereby improving test quality.

[0024] (4) The present invention first performs Fourier transform on the generated difference map to obtain the respective amplitude and phase, and combines the difference map phase and the logarithmic ratio map amplitude to generate a new difference map. The phase information is used in the generation of the difference map. The phase characteristics of the edge of the difference map change area are obvious, which makes it easy to extract features from the difference map in the later stage, laying a good foundation for the detection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 It is a principle block diagram of an intelligent electronic device program-controlled testing system based on quality monitoring information according to an embodiment of the present invention.

[0027] In the figure: 1. Collection unit; 2. Extraction and processing unit; 3. Test rule generation unit; 4. Connection test unit; 5. Feedback unit. DETAILED DESCRIPTION

[0028] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0029] According to an embodiment of the present invention, a program-controlled test system for an intelligent electronic device based on quality monitoring information is provided.

[0030] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the intelligent electronic device program-controlled test system based on quality monitoring information of an embodiment of the present invention, the system includes a collection unit 1 for collecting the current change amount of the signal port of the intelligent electronic device, an extraction processing unit 2, a test rule generation unit 3, a connection test unit 4 and a feedback unit 5; The connection test unit 4 and the collection unit 1 have an I / O port ( Figure 1 not shown).

[0031] Wherein, the collection unit 1 for collecting the current variation of the signal port of the intelligent electronic device is used to automatically collect quality monitoring information through multiple sources and store it in a storage repository; Specifically, the steps of collecting quality monitoring information from multiple sources and storing it in the repository include the following aspects: Determine what quality monitoring information needs to be collected, such as quality indicators (voltage, current, power supply, anti-interference, safety, stability, etc.), monitoring cycle, monitoring range, etc.; Select channels for obtaining quality monitoring information based on demand, such as production data, equipment sensors, manual inspection, etc. Collect data using automated tools or manually, and format and integrate the collected data into a readable form; Store the collected data in a unified repository to ensure data integrity, security, and availability, such as a database or cloud storage.

[0032] Extraction processing unit 2, used for processing the information in the repository and performing feature extraction; Specifically, the extraction processing unit 2 includes a generation module, an analysis module and a synthesis module.

[0033] The generation module is used to generate a difference map between two adjacent sets of quality monitoring information, and can be performed according to the following steps: Data preprocessing: Clean, denoise and normalize the raw data collected by the collection unit 1 to facilitate further analysis and visualization, and perform preprocessing, including removing outliers and filling missing data; Data fusion: Combining information from multiple data sources, effectively integrating them through methods such as feature extraction, feature matching, and data association, data fusion can help improve the accuracy and reliability of monitoring results; Information analysis: further analyze the preprocessed and fused data to extract useful information and patterns. The analysis methods include image recognition, target detection, classification, clustering, etc. Visualization: Convert the analysis results into a two-dimensional image format and process the image appropriately, such as adjusting color, brightness, contrast, etc. In addition, multiple two-dimensional images can be integrated into a more expressive scene representation using techniques such as image overlay and fusion. Result presentation: Finally, the two two-dimensional images are displayed to the user end for subsequent processing and use.

[0034] Two initial difference maps are constructed from two two-dimensional images, namely, a difference map and a logarithmic ratio map; Fourier transform is performed on the difference map and the logarithmic ratio map respectively to obtain the amplitude and phase of each difference map; the difference map phase and the logarithmic ratio map amplitude are combined, and a new difference map is generated by inverse Fourier transform.

[0035] The analysis module is used to analyze and process the same regions and different regions between the two sets of difference maps, and includes the following steps: Based on different quality monitoring information, two sets of difference maps are obtained using the generation module; The differences in the grayscale values ​​of the pixels corresponding to the two sets of difference maps are calculated respectively, and normalized to obtain the grayscale matrix of the difference map; Specifically, the gray value difference is calculated as follows: In the formula, is the difference in grayscale values, is a pixel, and They are the first group of difference maps and the second group of difference maps respectively; The first set of difference maps at pixel points The gray value of The second set of difference maps at pixel points The gray value of .

[0036] Specifically, the calculation formula for the normalization process is: In the formula, is the difference map S at the pixel point The gray value of is the matrix generated by subtracting the grayscale of the corresponding pixels of the first set of difference maps from the second set of difference maps, is the difference in grayscale values, is the maximum value of the matrix, is the minimum value of the matrix.

[0037] The quotient logarithms of the grayscale values ​​of the pixels corresponding to the two sets of difference maps are calculated respectively, and normalized to obtain the grayscale matrix of the ratio map.

[0038] Calculate the local variance value of the grayscale value of each pixel in the difference map, perform threshold segmentation, and obtain the optimal threshold; According to the variance matrix and the optimal threshold, the difference difference map and the ratio difference map are fused to obtain the gray value of each pixel in the final fused difference map.

[0039] The fused image is segmented into two FCM clusters to obtain the same area and different areas.

[0040] FCM is a clustering method based on fuzzy theory. It assigns each sample point the probability of belonging to a different cluster, rather than assigning each sample point to a certain cluster for sure like traditional hard clustering. In FCM, the center of each cluster is defined as the weighted average of all data points in the cluster. The probability that a sample point belongs to a specific cluster is calculated by the distance between the data point in the class and the center of the cluster, while considering the influence of other clusters on the cluster to which the point belongs.

[0041] The synthesis module is used to synthesize features from the difference map analysis processing results, including the following steps: Establish a DOG scale space pyramid model, and compare each point in the DOG scale space with the points of adjacent scales and adjacent positions one by one. The local extreme value points obtained are the required feature points. Delete low-end contrast pairs and unstable edge response points in feature points; According to the improved threshold, the low-contrast extreme points and unstable edge response points are deleted to obtain accurate feature points; Calculate the standard value corresponding to each group of precise points. If the standard value is less than the threshold, it is low contrast and is filtered out. Otherwise, it is retained. The retained feature points are used to generate feature descriptors, and the feature descriptors of the first new difference image and the second new difference image are used to determine whether the feature points corresponding to the two images are matching points. If they are matching points, they are determined to be implementation features.

[0042] DOG scale (Difference of Gaussian, DOG) space is a core concept in multi-scale image processing, which is used to detect feature points at different scales. In DOG scale space, the image will undergo multiple Gaussian blurs and differential processing at two adjacent scales to generate a series of Gaussian difference maps at different scales. Each Gaussian difference map represents the information in the image at different scales.

[0043] A test rule generating unit 3, used for designing program-controlled test rules using the extracted implementation features; The test rule generating unit 3 includes a formulation module and a construction module.

[0044] The formulation module is used to formulate program control test rules according to the implementation characteristics of the analysis and processing and the performance test requirements, including the following steps: Attribute constraints are performed based on performance tests, and redundant attributes are deleted to form a simplified attribute set; Calculate the importance of each attribute in the reduced attribute set and select the largest attribute as the splitting attribute; Repeat the above steps until the decision attributes of each category are consistent, generating a leaf node; Generate program-controlled rules through reading, and calculate the confidence and support of the rules to obtain the rule set; Verify whether the obtained rule set complies with the program-controlled test rules. If not, repeat the above steps. Otherwise, directly output the rule set.

[0045] The construction module is used to complete the construction of the performance model according to the performance test requirements and relevant information of the production system, and guide the performance test process and the generation of test results, including the following steps: The results of the program control rule set, performance test requirements and relevant information of the production system are combined and divided into two categories: normal data set and mixed data set. At the same time, the association analysis algorithm is used to obtain the association patterns of the normal data set and the mixed data set respectively; Based on the association algorithm, the sequence analysis algorithm is used to calculate the sequence rules of the normal data set and the mixed data set; The sequence rule coding of the normal data set and the mixed data set are obtained by using the sequence rule coding algorithm, and the encoded patterns are compared to obtain the sequence coding pattern that conforms to the program control rules; Apply statistical feature construction algorithms to sequence coding patterns to construct statistical feature attributes that can clearly distinguish normal and abnormal behaviors; Add statistical feature attributes to the training data set and the test data set and calculate all feature attribute values. Post-process the obtained data files and convert them into a data format that can be recognized by the classification algorithm. Finally, generate a classification detection model based on the classification algorithm.

[0046] A connection test unit 4 is used to connect the test rules and the intelligent electronic device to the test host and perform program-controlled testing; Specifically, all intelligent electronic devices are connected to the test host through wireless connection, and each intelligent electronic device is numbered to receive test instructions and execute the test content of the test instructions.

[0047] The feedback unit 5 is used to collect the test results of the electronic device at each stage and send the test results to the monitoring terminal for display.

[0048] Among them, collecting the test results of each stage of the electronic equipment and sending the test results to the monitoring terminal for display includes the following aspects: Collect test results: collect and organize the test results of each test phase and save them in the database using data structures for subsequent data analysis and reporting; Send test results: Send test results to the monitoring terminal for monitoring, and issue an alarm notification when abnormal conditions occur; Display test results: The monitoring terminal can obtain test results in real time, conduct real-time monitoring and record relevant data, and set various reports, charts or other visualization tools to display test results; Analyze test data: Perform dynamic analysis such as data mining and machine learning, as well as statistical data on large amounts of test data, analyze the quality of test results, and make targeted optimization suggestions to improve production efficiency and quality.

[0049] In summary, with the help of the above technical solution of the present invention, the present invention provides a program-controlled test system for intelligent electronic devices based on quality monitoring information, which can automate and optimize the test process, improve test efficiency, and avoid the influence of human factors on the test results. By analyzing the quality monitoring information, key indicators and performance parameters can be obtained, so as to determine the test requirements and test targets, and realize coverage testing of functional modules, data structures and logic algorithms in more specific scenarios. The present invention can also remove irrelevant or fully verified test cases based on quality monitoring information, avoid repeated tests and redundant tests, make the test work more centralized and effective, and the program-controlled test has high reliability and consistency, can greatly reduce the test error rate, and has the advantages of automated detection and verification, thereby ensuring the accuracy and reliability of the test process and test results. The program-controlled test based on quality monitoring information can promote the implementation of continuous testing based on real-time monitoring data and test reports, periodically evaluate and feedback the performance of the product, and timely feedback and improve the problems found, so that the test becomes a continuously optimized work process.

[0050] The present invention processes the collected quality monitoring information, extracts rules that meet the quality standards from the information as the standard for testing intelligent electronic devices, thereby improving the accuracy and reliability of the measurement data, saving a lot of test time and cost, helping to ensure the accuracy and reliability of the test results, and reducing the test error rate, thereby improving the test quality. The present invention first performs a Fourier transform on the generated difference map, obtains the respective amplitudes and phases, and combines the difference map phase and the logarithmic ratio map amplitude to generate a new difference map. The phase information is applied to the generation of the difference map. The phase characteristics of the edge of the difference map change area are obvious, which can facilitate the extraction of features from the difference map in the later stage, laying a good foundation for the detection work.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A program-controlled test system for intelligent electronic devices based on quality monitoring information, characterized in that: The system includes: a connection test unit, a collection unit, an extraction processing unit, a test rule generation unit and a feedback unit; The connection test unit is used to connect to the signal port of the intelligent electronic device and send a time-domain current signal of a preset waveform to the signal port; The collecting unit is used to collect the feedback current signal of the signal port of the intelligent electronic device, and store the feedback current signal in a storage repository in the form of a time domain waveform; The extraction processing unit is used to perform clutter filtering and feature extraction on the feedback current signal of the signal port of the intelligent electronic device in the storage library; The test rule generating unit is used to generate a program-controlled test rule according to the extracted implementation characteristics of the current variation of the signal port of the intelligent electronic device; The feedback unit is used to collect the test results of the electronic equipment at each stage and send the test results to the monitoring terminal for display.

2. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 1, characterized in that: The extraction processing unit includes a generation module, an analysis module and a synthesis module; The generating module is used to generate a difference map between two adjacent sets of quality monitoring information; The analysis module is used to analyze and process the same regions and different regions between the two sets of difference maps; The synthesis module is used to synthesize features from the difference map analysis processing results to obtain implementation features.

3. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 2, characterized in that: The step of generating a difference map between two adjacent sets of quality monitoring information comprises: S11, pre-processing two adjacent sets of quality monitoring information collected from multiple channels; S12, performing visualization imaging processing on the preprocessed information to obtain two two-dimensional images of the same monitoring scene; S13, constructing two initial difference maps through the two two-dimensional images, which are a difference map and a logarithmic ratio map; S14, performing Fourier transform on the difference map and the logarithmic ratio map respectively to obtain the amplitude and phase of each initial difference map; S15, combining the phase of the difference map and the amplitude of the logarithmic ratio map, and generating a new difference map through inverse Fourier transform.

4. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 3 is characterized in that: The step of analyzing and processing the same regions and different regions between the two sets of difference maps comprises: S21, using a generation module to obtain two sets of difference maps; S22, respectively calculating the difference in the grayscale values ​​of the pixels corresponding to the two sets of difference images, and performing normalization processing to obtain the grayscale matrix of the difference image; S23, respectively calculating the quotient logarithms of the grayscale values ​​of the pixels corresponding to the two sets of difference images, and performing normalization processing to obtain the grayscale matrix of the ratio image; S24, calculating the local variance value of the gray value of each pixel in the difference map, and performing threshold segmentation to obtain the optimal threshold; S25, according to the variance matrix and the optimal threshold, the difference difference map and the ratio difference map are fused to obtain the gray value of each pixel of the fused final difference map; S26. Perform FCM two-cluster segmentation on the fused image to obtain the same area and the different area.

5. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 4, characterized in that: The calculation formula of the gray value difference is: In the formula, is the difference in grayscale values, is a pixel, and They are the first group of difference maps and the second group of difference maps respectively; The first set of difference maps at pixel points The gray value of The second set of difference maps at pixel points The gray value of .

6. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 5, characterized in that: The calculation formula for the normalization process is: In the formula, is the difference map S at the pixel point The gray value of is the matrix generated by subtracting the grayscale of the corresponding pixels of the first set of difference maps from the second set of difference maps, is the difference in grayscale values, is the maximum value of the matrix, is the minimum value of the matrix.

7. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 6, characterized in that: The feature synthesis is performed from the difference map analysis processing result to obtain the implementation feature, which includes: S31, establish a DOG scale space pyramid model, and compare each point in the DOG scale space with the points of adjacent scales and adjacent positions one by one, and the obtained local extreme points are the feature points; S32, deleting low-end comparison pairs and unstable edge response points in feature points; S33, deleting low-contrast extreme points and unstable edge response points according to the improved threshold value to obtain accurate feature points; S34, calculate and obtain the standard value corresponding to each group of precise points. If the standard value is less than the threshold, it is low contrast and is filtered out. Otherwise, it is retained. S35, generating feature descriptors for the retained feature points, and using the feature descriptors of the first new difference image and the second new difference image to determine whether the feature points corresponding to the two images are matching points, and if so, determining them as implementation features.

8. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 7, characterized in that: The test rule generation unit includes a formulation module and a construction module; The formulation module is used to formulate program control test rules according to the implementation characteristics and performance test requirements obtained through analysis and processing; The construction module is used to complete the construction of the performance model according to the performance test requirements and relevant information of the production system, and guide the performance test process and the generation of test results.

9. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 8, characterized in that: The formulating of program control test rules according to the implementation characteristics and performance test requirements obtained through analysis and processing includes: Attribute constraints are performed based on performance tests, and redundant attributes are deleted to form a simplified attribute set; Calculate the importance of each attribute in the simplified attribute set and select the largest attribute as the splitting attribute; Repeat the above steps until the decision attributes of each category are consistent, generating a leaf node; Generate program-controlled rules through reading, and calculate the confidence and support of the rules to obtain the rule set; Verify whether the obtained rule set complies with the program-controlled test rules. If not, repeat the above steps. Otherwise, directly output the rule set.

10. The intelligent electronic device program-controlled test system based on quality monitoring information according to claim 9, characterized in that: The construction of the performance model is completed according to the performance test requirements and the relevant information of the production system, and the generation of the performance test process and test results is guided, including: The results of the program control rule set, performance test requirements and relevant information of the production system are combined and divided into two categories: normal data set and mixed data set. At the same time, the association analysis algorithm is used to obtain the association patterns of the normal data set and the mixed data set respectively; Based on the association algorithm, the sequence analysis algorithm is used to calculate the sequence rules of the normal data set and the mixed data set; The sequence rule coding of the normal data set and the mixed data set are obtained through the sequence rule coding algorithm, and the encoded patterns are compared to obtain the sequence coding pattern that conforms to the program control rules; Apply statistical feature construction algorithms to sequence coding patterns to construct statistical feature attributes that can clearly distinguish normal and abnormal behaviors; Statistical feature attributes are added to the training data set and the test data set and all feature attribute values ​​are calculated. The obtained data files are post-processed and converted into a data format that can be recognized by the classification algorithm. Finally, a classification detection model is generated according to the classification algorithm.

Images