CN116416884A - Testing device and testing method for display module - Google Patents
Testing device and testing method for display module Download PDFInfo
- Publication number
- CN116416884A CN116416884A CN202310688591.5A CN202310688591A CN116416884A CN 116416884 A CN116416884 A CN 116416884A CN 202310688591 A CN202310688591 A CN 202310688591A CN 116416884 A CN116416884 A CN 116416884A
- Authority
- CN
- China
- Prior art keywords
- test data
- model
- test
- display module
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 559
- 238000007405 data analysis Methods 0.000 claims abstract description 64
- 230000004044 response Effects 0.000 claims abstract description 62
- 238000000034 method Methods 0.000 claims abstract description 53
- 230000010354 integration Effects 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000011156 evaluation Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 20
- 238000003066 decision tree Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 12
- 238000007637 random forest analysis Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000010998 test method Methods 0.000 claims description 5
- 230000008030 elimination Effects 0.000 claims description 3
- 238000003379 elimination reaction Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 19
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000013515 script Methods 0.000 description 5
- 208000025174 PANDAS Diseases 0.000 description 4
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 4
- 240000004718 Panda Species 0.000 description 4
- 235000016496 Panda oleosa Nutrition 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000007635 classification algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000002441 reversible effect Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013079 data visualisation Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/006—Electronic inspection or testing of displays and display drivers, e.g. of LED or LCD displays
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to the field of artificial intelligence, and discloses a testing device and a testing method of a display module, which are used for improving the testing accuracy of the display module. The method comprises the following steps: respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model; obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data; inputting the image quality test data and the response time test data into a target test meta model for display module test data analysis to obtain a plurality of test data analysis results; and integrating the results of the analysis of the plurality of test data, and outputting a target test result corresponding to the second display module through the target test meta-model.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a testing device and a testing method of a display module.
Background
Currently, performance parameter detection of display modules is often dependent on specialized instrumentation and tools, but these methods tend to be relatively complex, time consuming and expensive. Therefore, the display module is tested by using the artificial intelligence model, so that the accuracy and the intelligence of the test can be improved, and the lower test cost and the faster test speed can be realized.
Traditional testing methods often require specialized personnel to process and analyze the data, and the test data may be affected by factors such as environment and equipment, and the accuracy of the test results cannot be guaranteed. Because of the large number of test parameters, the test data becomes very huge, and the conventional data analysis method may be faced with the problems of complex calculation and data confusion. The traditional test method can only obtain static test results and cannot adjust and optimize real-time data in time.
Disclosure of Invention
The invention provides a testing device and a testing method of a display module, which are used for improving the testing accuracy of the display module.
The first aspect of the present invention provides a method for testing a display module, where the method for testing a display module includes:
acquiring first test data of a plurality of first display modules, and classifying the first test data to obtain a plurality of second test data;
Respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model;
obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data;
inputting the image quality test data and the response time test data into the target test meta-model for display module test data analysis to obtain a plurality of test data analysis results;
and integrating the results of the analysis of the plurality of test data, and outputting a target test result corresponding to the second display module through the target test meta-model.
With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the obtaining first test data of the plurality of first display modules and classifying the first test data to obtain a plurality of second test data includes:
acquiring a plurality of first display modules, determining the number of the modules of the plurality of first display modules, acquiring first test data of each display module, and transmitting the first test data to a preset first test system;
Carrying out data cleaning on the first test data of each display module through the first test system to obtain cleaned first test data;
performing repeated value elimination and missing value filling on the cleaned first test data to obtain preprocessed first test data;
performing feature classification on the preprocessed first test data according to a preset feature classification index to obtain a plurality of second test data, wherein the feature classification index comprises: resolution, color depth, contrast, and response time.
With reference to the first aspect, in a second implementation manner of the first aspect of the present invention, the separately constructing a test data base model of each second test data includes:
performing data division on the plurality of second test data to obtain characteristic data and tag data;
constructing a plurality of decision trees based on the characteristic data, and generating a random forest according to the decision trees;
and classifying and carrying out regression training on the random forest to obtain a test data base model of each second test data.
With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the performing model integration on the test data base model to generate a target test meta-model includes:
Evaluating the test data base model through the tag data to obtain model performance evaluation parameters, and selecting an optimal model architecture according to the model performance evaluation parameters;
according to the model architecture, carrying out model series connection on the test data base model to obtain a stacking model;
and performing model architecture and weight optimization on the stacking model to generate a target test meta-model.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, the obtaining target test data corresponding to the second display module to be processed, and extracting features of the target test data, to obtain image quality test data and response time test data, includes:
acquiring target test data corresponding to a second display module to be processed based on a preset second test system;
performing image feature recognition on the target test data to obtain initial image test data, and performing response feature extraction on the target test data to obtain initial response test data;
and respectively carrying out feature normalization processing on the initial image test data and the initial response test data to obtain image quality test data and response time test data.
With reference to the first aspect, in a fifth implementation manner of the first aspect of the present invention, inputting the image quality test data and the response time test data into the target test meta-model to perform display module test data analysis, to obtain a plurality of test data analysis results, includes:
inputting the image quality test data and the response time test data into the target test meta-model;
performing display module test data analysis through a plurality of test data base models in the target test meta-model to obtain a test data analysis result corresponding to each test data base model;
and taking the test data analysis result corresponding to each test data base model as the target test meta-model and outputting the test data analysis result to obtain a plurality of test data analysis results.
With reference to the first aspect, in a sixth implementation manner of the first aspect of the present invention, the integrating the results of analyzing the plurality of test data, and outputting, by using the target test meta-model, a target test result corresponding to the second display module includes:
acquiring weight evaluation coefficients of the plurality of test data base models based on the model architecture;
Performing result integration on the test data analysis results corresponding to each test data base model according to the weight evaluation coefficients;
and outputting a target test result corresponding to the second display module through the target test meta model, and transmitting the target test result to the second test system.
The second aspect of the present invention provides a testing device for a display module, the testing device for a display module includes:
the device comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring first test data of a plurality of first display modules and classifying the first test data to obtain a plurality of second test data;
the integration module is used for respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model;
the extraction module is used for acquiring target test data corresponding to the second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data;
the analysis module is used for inputting the image quality test data and the response time test data into the target test meta-model to perform display module test data analysis to obtain a plurality of test data analysis results;
And the integration module is used for integrating the results of the analysis results of the plurality of test data and outputting the target test result corresponding to the second display module through the target test meta-model.
A third aspect of the present invention provides a test apparatus for a display module, comprising: a memory and at least one processor, the memory having instructions stored therein; and the at least one processor calls the instructions in the memory to enable the testing equipment of the display module to execute the testing method of the display module.
A fourth aspect of the present invention provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the method of testing a display module as described above.
In the technical scheme provided by the invention, a test data base model of each second test data is respectively constructed, and the test data base models are subjected to model integration to generate a target test meta-model; obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data; inputting the image quality test data and the response time test data into a target test meta model for display module test data analysis to obtain a plurality of test data analysis results; the invention performs objective and accurate test on the performance parameters of the display module by the artificial intelligent model, avoids the influence of artificial factors in the test process, and improves the test accuracy. By using the artificial intelligent model, automatic analysis and processing of test data are realized without complicated manual intervention, so that the intelligence of the test is greatly improved, the labor and time cost is reduced, and lower test equipment cost is realized. And the real-time data analysis and processing are realized, the test result is quickly responded and adjusted, and the optimization and improvement can be timely carried out, so that the test accuracy of the display module is improved.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a testing method of a display module according to the present invention;
FIG. 2 is a flow chart of constructing a test data base model in an embodiment of the invention;
FIG. 3 is a flow chart of model integration in an embodiment of the invention;
FIG. 4 is a flow chart of feature extraction in an embodiment of the invention;
FIG. 5 is a schematic diagram of an embodiment of a testing apparatus for a display module according to the present invention;
fig. 6 is a schematic diagram of an embodiment of a testing apparatus for a display module according to the present invention.
Detailed Description
The embodiment of the invention provides a testing device and a testing method of a display module, which are used for improving the testing accuracy of the display module. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For easy understanding, the following describes a specific flow of an embodiment of the present invention, referring to fig. 1, and an embodiment of a method for testing a display module in an embodiment of the present invention includes:
s101, acquiring first test data of a plurality of first display modules, and classifying the test data of the first test data to obtain a plurality of second test data;
it can be understood that the execution body of the present invention may be a testing device of a display module, and may also be a terminal or a server, which is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
Specifically, the server is connected to a plurality of first display modules and reads first test data from each module, where the server reads the first test data from each module through a data acquisition device and a corresponding interface and communication protocol, and further groups the first test data according to characteristics of the data, such as a numerical value, a frequency, a waveform, a change rate, a timestamp, and the like, to generate a plurality of second test data, and after classifying the first test data, extracts a group of data from each classification as the second test data, and it is to be noted that the second test data can compare effects under different test conditions, for example, 3 first display modules and various acquisition devices. Firstly, reading first test data from each module, such as measuring parameters of brightness, color, resolution and the like of a display; then, according to the previously set classification rule, the first test data is divided into three groups, such as that all brightness data are placed in one group, all color data are placed in the second group, all resolution data are placed in the third group, and finally, one group of data is extracted from the 3 classifications respectively as the second test data, for example, the maximum brightness value in the brightness group, the red saturation in the color group and the horizontal resolution value in the resolution group.
S102, respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model;
it should be noted that the test data base model is a base model constructed according to features and attributes in the second test data, specifically, each second test data may independently construct a test data base model. Further, the server integrates the test data base model of each second test data, wherein the target test meta-model is the result of the whole model integration and is a comprehensive description and prediction of all the test data. Specifically, the server processes the integrated model to generate a target test meta-model. In this step, for example, the machine life of a display needs to be tested. Three characteristics including CPU occupancy rate and screen brightness are selected from the extracted second test data respectively, a test data base model is built for each characteristic, and three base models are obtained by using different algorithms and technologies; and then, carrying out integrated processing on the base models, such as an integrated algorithm such as a decision tree with multiple outputs by using a neural network layer, so as to obtain a comprehensive target test meta-model, wherein the model can predict the machine life of the display.
S103, obtaining target test data corresponding to the second display module to be processed, and extracting features of the target test data to obtain image quality test data and response time test data;
it should be noted that, first, the server analyzes the way to obtain the data of the second display module, where if the data is real-time data collected by an external sensor, the sensor can be directly accessed and read. If collected through a website, the data is extracted through web crawler technology. For example, crawling target test data on a website is achieved through a Requests library and a Beautifuge library written by Python, data is obtained from an API of the website through the Requests library, then the Beautifuge library is used for extracting data and cleaning the data, reliability and accuracy of the collected data are ensured, it is to be noted that feature extraction is a process of extracting useful information from original data, in the invention, feature extraction of the target test data is achieved through Fourier transformation, wavelet transformation and the like, for example, fourier transformation converts a time domain signal into a frequency domain signal to obtain information about signal frequency and amplitude. Wavelet transforms explore the local and global properties of a signal by decomposing the signal and extracting features in different frequency bins. After feature extraction, image quality test data and response time test data are obtained using these features. The image quality test data is classified, for example, by a classifier such as a Support Vector Machine (SVM) or random forest. Regression analysis is performed on the response time test data by linear regression or neural networks.
S104, inputting the image quality test data and the response time test data into a target test meta model for display module test data analysis to obtain a plurality of test data analysis results;
specifically, the image quality test data and the response time test data are input into a target test meta-model to analyze the test data of the display module. Firstly, before model training, a training data set is divided, and the training data set is generally divided into a training set and a testing set, and then a target test meta-model is constructed through a Support Vector Machine (SVM). Finally, the image quality test data and the response time test data are input into the model by Python for data analysis, and typically the output of the model typically contains a plurality of indicators that can be used to analyze the quality and performance of the display module. For example, the performance quality of the display module under different environments (e.g., indoor and outdoor) and performance under different display sizes and resolutions are predicted by the model. In addition, in the embodiment of the invention, the root cause of the performance problem of the display module is determined by a reverse reasoning technology. The reverse reasoning technique can analyze the output of the model and determine the factors of the output results, thereby helping to identify and solve display module performance problems. For example, if the results of the model output show a slower response time, then the use of reverse reasoning techniques can determine which factors caused the problem, e.g., screen resolution that is too high resulting in a reduced response time or other factors that result in a too long response time.
S105, integrating the results of the analysis of the plurality of test data, and outputting a target test result corresponding to the second display module through the target test meta model.
Specifically, first, a tool or platform is selected that analyzes a plurality of test data. For example, through Pandas data analysis library in Python programming language, multiple data files are read into data frame, and then data are combined, processed and integrated for unified output and analysis. Other tools similar to R or Matlab may be used to perform data analysis and integration, and further, by writing Python or Matlab programs, key parameters corresponding to the target test meta-model are extracted from the plurality of test data, these parameters are input into the model, the analysis result is output through the model, and the result data is output to the second display module. For example, assume that a batch of test data for a display is being processed, and a corresponding target test result is output. Test data were read and processed by using Pandas and NumPy libraries in Python and classification or regression models in Scikit-learn libraries were used as target test metamodels. The performance and quality of the display will be predicted by inputting key parameters of the test data, such as brightness, color, contrast, etc., and the analysis results will be derived.
In the embodiment of the invention, a test data base model of each second test data is respectively constructed, and the test data base models are subjected to model integration to generate a target test meta-model; obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data; inputting the image quality test data and the response time test data into a target test meta model for display module test data analysis to obtain a plurality of test data analysis results; the invention performs objective and accurate test on the performance parameters of the display module by the artificial intelligent model, avoids the influence of artificial factors in the test process, and improves the test accuracy. By using the artificial intelligent model, automatic analysis and processing of test data are realized without complicated manual intervention, so that the intelligence of the test is greatly improved, the labor and time cost is reduced, and lower test equipment cost is realized. And the real-time data analysis and processing are realized, the test result is quickly responded and adjusted, and the optimization and improvement can be timely carried out, so that the test accuracy of the display module is improved.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Acquiring a plurality of first display modules, determining the number of the modules of the plurality of first display modules, acquiring first test data of each display module, and transmitting the first test data to a preset first test system;
(2) Carrying out data cleaning on the first test data of each display module through a first test system to obtain cleaned first test data;
(3) Performing repeated value elimination and missing value filling on the cleaned first test data to obtain preprocessed first test data;
(4) Performing feature classification on the preprocessed first test data according to a preset feature classification index to obtain a plurality of second test data, wherein the feature classification index comprises: resolution, color depth, contrast, and response time.
Specifically, the server first marks or stores a plurality of first display modules in a database for tracking and management. And meanwhile, the number of each module is determined, and each acquired display module can work normally. And finally, acquiring the first test data of each display module, transmitting the first test data to a preset first test system, and transmitting the data to the first test system and cleaning the data after acquiring the first test data of each module. The data cleansing may include the following steps: and removing invalid data, denoising, processing abnormal values, smoothing data and the like. The problem that the first test data after cleaning may still have a repeated value and a missing value, etc. it should be noted that, for the repeated value, redundant records in the data are deleted by a "deduplication" method; for missing values, correlation methods (e.g., mean interpolation, KNN interpolation, regression interpolation, EM algorithm, etc.) may be used to populate the missing data. Further, the first test data after preprocessing is subjected to feature classification, wherein indexes of the feature classification comprise parameters such as resolution, color depth, contrast, response time and the like. The method comprises the steps of classifying features through a classification algorithm and a clustering method in a Scikit-learn library in a Python programming language, classifying and dividing the preprocessed first test data according to a preset feature classification index, and obtaining a plurality of second test data. For example, assume that multiple computer display modules are being tested and feature classified. The module data is read and processed through NumPy and Pandas libraries in Matlab or Python programming languages, and then the number and the category of the modules are determined by using a cluster analysis method. And further, the characteristic classification indexes are predicted and evaluated through classification or regression models in the Scikit-learn library, so that the display effect and quality of the module can be better known.
In a specific embodiment, as shown in fig. 2, the process of executing step S102 may specifically include the following steps:
s201, data division is carried out on a plurality of second test data to obtain feature data and tag data;
s202, constructing a plurality of decision trees based on the feature data, and generating a random forest according to the decision trees;
and S203, classifying and carrying out regression training on the random forest to obtain a test data base model of each second test data.
Specifically, the server performs data division on a plurality of second test data to obtain feature data and tag data, and first, performs data division on the second test data to obtain feature data and tag data. Feature data refers to key feature parameters extracted from test data, such as resolution, color depth, contrast, response time, etc., by the preceding feature classification. The label data refers to label and label data obtained according to actual test results, such as characteristic parameters of brightness, color, contrast and the like of a screen, a plurality of decision trees are constructed through a decision tree algorithm based on the characteristic data, and integrated processing is carried out according to a random forest mode. Decision trees are classical machine learning algorithms that classify and judge data in a manner similar to human thinking. By judging and analyzing the characteristic data, the decision tree can automatically classify the data and obtain classification results. The random forest is an integrated model formed by a plurality of decision trees, and can effectively avoid the problems of over fitting and insufficient generalization capability of a single decision tree, thereby obtaining more accurate classification results and model prediction capability. For random forests, training and prediction is performed by classification and regression algorithms. The classification algorithm classifies test data into specific classifications, such as screen resolution, color depth, response time, etc., and then performs test simulation and prediction based on the classification results. And the regression algorithm can predict and estimate parameters such as color accuracy, contrast, and brightness of the relevant characteristics and performance of the display module. For example, assuming that display testing is required, the above procedure may be implemented using the Python programming language and the Scikit-learn library. The plurality of second test data is divided into feature data and tag data, and a random forest model is constructed using a decision tree algorithm. The test data is then classified into specific categories, e.g., high definition, 4K, HDR, etc., by a classification algorithm, and parameters such as color accuracy, contrast, brightness, etc., are predicted using a regression algorithm. Finally, the resulting test data base model is integrated and evaluated to understand the characteristics and performance of the display.
In a specific embodiment, as shown in fig. 3, the process of executing step S102 further includes the following steps:
s301, evaluating the test data base model through label data to obtain model performance evaluation parameters, and selecting an optimal model architecture according to the model performance evaluation parameters;
s302, carrying out model series connection on the test data base model according to a model architecture to obtain a stacked model;
and S303, performing model architecture and weight optimization on the stack model to generate a target test meta-model.
Specifically, the server first evaluates the test data base model by using the tag data, and in the embodiment of the present invention, the performance of the model is evaluated by using different evaluation indexes, such as accuracy, recall, and F1 value, and an optimal model architecture is selected. On the other hand, the conditions of over fitting, under fitting and the like need to be paid attention to, and the model is ensured to have generalization capability and adaptability to new data. And carrying out model series connection on the test data base model based on the optimal model architecture to obtain a stacked model. The stacking model is an integrated learning method, and can aggregate the prediction results of a plurality of basic models, thereby obtaining more accurate model prediction capability. Model series flow is realized through the existing Python or R library, and proper integration strategy and model optimization method are selected. And performing model architecture and weight tuning on the stacked model to generate a target test meta-model. And for generating the target test meta-model, performing further model architecture and weight tuning on the stacked model. Model architectural tuning typically involves adding, deleting, or modifying network structures and hierarchies of the model in order to increase the representation and generalization capabilities of the model. Weight tuning is preferably achieved by adjusting parameters and super-parameters of the model, such as learning rate, regularization factors, etc., to minimize model loss or error, thereby obtaining more accurate model predictions. For example, suppose that an intelligent display test metamodel needs to be designed. Model evaluation and tuning procedures are implemented by a Python programming language and a Scikit-learn library. First, it is necessary to evaluate the performance of the test data base model and select the optimal model architecture. Then, a plurality of basic models are combined into a stacked model by a model series method. And further, the stacking model is subjected to model architecture and weight optimization, so that an accurate and robust target test meta-model is obtained.
In a specific embodiment, as shown in fig. 4, the process of performing step S103 may specifically include the following steps:
s401, acquiring target test data corresponding to a second display module to be processed based on a preset second test system;
s402, performing image feature recognition on target test data to obtain initial image test data, and performing response feature extraction on the target test data to obtain initial response test data;
and S403, performing feature normalization processing on the initial image test data and the initial response test data respectively to obtain image quality test data and response time test data.
Specifically, the server first needs to obtain target test data corresponding to the second display module to be processed based on a preset second test system. These target test data are stored in a database for subsequent processing and analysis. And meanwhile, the acquired test data is subjected to preliminary processing and preprocessing so as to ensure the accuracy and the integrity of the data. And extracting the image and response characteristics of the target test data so as to better understand the characteristics and the performances of the display module. Image characteristics may include parameters such as resolution, color depth, contrast, and brightness, while response characteristics may include parameters such as response time, refresh rate, and brightness. The feature extraction is performed on the target test data by various image processing algorithms and feature extraction methods, such as Convolutional Neural Network (CNN), image segmentation, edge detection, face recognition, and the like. The extracted image and response features are feature normalized to better process the data and prepare them for subsequent analysis. The feature normalization process may include various methods such as mean variance normalization, range normalization, maximum minimum normalization, and the like. Through the feature normalization processing, different feature parameters are standardized, and the situation that the different feature parameters are affected mutually due to different scales is eliminated. For example, assume a new display module is being tested. Test data is read and processed through the Python programming language and OpenCV library, and Convolutional Neural Networks (CNNs) are used to extract key feature information from the images. Further, the extracted response characteristics are analyzed and processed by a response detection algorithm and a data visualization tool, such as a Seaborn library in Matlab or Python. Finally, the extracted image and response characteristics are normalized by a characteristic normalization method, so that more accurate and reliable test data and characteristic parameters are obtained.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Inputting the image quality test data and the response time test data into a target test meta-model;
(2) Performing display module test data analysis through a plurality of test data base models in the target test meta-model to obtain a test data analysis result corresponding to each test data base model;
(3) And taking the test data analysis result corresponding to each test data base model as a target test meta model and outputting the target test meta model to obtain a plurality of test data analysis results.
Specifically, first, image quality test data and response time test data are input into a target test meta-model for test data analysis and processing. Through Python or R programming language, through calling the API or function in the target test meta-model, the data is input and converted so as to meet the input requirement of the model. And further, performing display module test data analysis through a plurality of test data base models in the target test meta-model to obtain test data analysis results corresponding to each test data base model. These base models may be various machine learning and data processing models, such as decision trees, linear regression, neural networks, PCA, and the like. By inputting test data, the base models analyze the test data for respective target features (image quality or response time, etc.), and obtain corresponding prediction results and evaluation indexes. And finally, taking the test data analysis result corresponding to each test data base model as the output of the target test meta-model, and integrating and summarizing to obtain a plurality of test data analysis results. Script scripts are written through Python or R programming language, the results of all the base models are integrated, a plurality of test data analysis reports and result files are generated, and visualization and data analysis can be carried out according to the requirements. For example, assuming a new display is being tested, the image quality test data and response time test data have been extracted using the previous process and a target test meta-model has been generated. Now, it is necessary to analyze and process test data using a target test meta-model and to obtain a plurality of test data analysis results. Test data may be input into the target test meta-model and test data analysis may be performed on a plurality of base models. Finally, the data is analyzed by data analysis and visualization tools, such as Pandas and Matplotlib kits, before outputting the results of the multiple test data analyses.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Based on a model architecture, acquiring weight evaluation coefficients of a plurality of test data base models;
(2) Carrying out result integration on the test data analysis results corresponding to each test data base model according to the weight evaluation coefficients;
(3) And outputting a target test result corresponding to the second display module through the target test meta model, and transmitting the target test result to the second test system.
Specifically, first, it is necessary to obtain weight evaluation coefficients of a plurality of test data base models based on a model architecture. The weight evaluation coefficients are typically a set of real numbers that indicate the importance and contribution of each test data base model in the overall test data analysis. The weight coefficient of each model is obtained by various model evaluation methods and weight coefficient calculation methods, such as PCA, lasso, and ridge regression. And integrating the results of the test data analysis corresponding to each test data base model according to the weight evaluation coefficients. Integration may include various methods such as weighted averaging and base model selection. Through the guidance of the weight evaluation coefficients, a plurality of test data analysis results can be better integrated and integrated, and more accurate and reliable test results are obtained. And finally, outputting a target test result corresponding to the second display module through the target test meta model, and transmitting the target test result to the second test system. Script scripts are written through Python or R programming language, test data analysis results are integrated, target test results are generated, and the results are transmitted to a second test system, so that the performance and characteristics of the display module can be better known and evaluated. For example, assuming a new dashboard display is being tested, the previous process has been completed, the weight evaluation coefficients for a plurality of test data base models are derived, and the test data analysis results are integrated and integrated to generate the target test results. And further, a script is written through a Python programming language, the target test result is output in a PDF or other visual format, and the result is transmitted to a second test system of the automobile instrument panel display.
The method for testing the display module in the embodiment of the present invention is described above, and the following describes a testing device for the display module in the embodiment of the present invention, referring to fig. 5, and one embodiment of the testing device for the display module in the embodiment of the present invention includes:
the acquiring module 501 is configured to acquire first test data of a plurality of first display modules, and classify the first test data to obtain a plurality of second test data;
the integration module 502 is configured to respectively construct a test data base model of each second test data, and perform model integration on the test data base model to generate a target test meta-model;
the extracting module 503 is configured to obtain target test data corresponding to the second display module to be processed, and perform feature extraction on the target test data to obtain image quality test data and response time test data;
the analysis module 504 is configured to input the image quality test data and the response time test data into the target test meta model to perform display module test data analysis, so as to obtain a plurality of test data analysis results;
and the integrating module 505 is configured to integrate the results of the analysis of the plurality of test data, and output a target test result corresponding to the second display module through the target test meta-model.
Respectively constructing a test data base model of each second test data through the cooperative cooperation of the components, and carrying out model integration on the test data base model to generate a target test meta-model; obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data; inputting the image quality test data and the response time test data into a target test meta model for display module test data analysis to obtain a plurality of test data analysis results; the invention performs objective and accurate test on the performance parameters of the display module by the artificial intelligent model, avoids the influence of artificial factors in the test process, and improves the test accuracy. By using the artificial intelligent model, automatic analysis and processing of test data are realized without complicated manual intervention, so that the intelligence of the test is greatly improved, the labor and time cost is reduced, and lower test equipment cost is realized. And the real-time data analysis and processing are realized, the test result is quickly responded and adjusted, and the optimization and improvement can be timely carried out, so that the test accuracy of the display module is improved.
The above fig. 5 describes the testing apparatus of the display module in the embodiment of the present invention in detail from the point of view of the modularized functional entity, and the following describes the testing device of the display module in the embodiment of the present invention in detail from the point of view of hardware processing.
Fig. 6 is a schematic structural diagram of a testing device for a display module according to an embodiment of the present invention, where the testing device 600 for a display module may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) storing applications 633 or data 632. Wherein the memory 620 and the storage medium 630 may be transitory or persistent storage. The program stored on the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations in the test device 600 of the display module. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the test device 600 of the display module.
The display module testing apparatus 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input/output interfaces 660, and/or one or more operating systems 631, such as Windows Serve, macOS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the test device structure of the display module shown in fig. 6 does not constitute a limitation of the test device of the display module, and may include more or fewer components than shown, or may combine certain components, or may have a different arrangement of components.
The invention also provides a testing device of the display module, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor executes the steps of the testing method of the display module in the embodiments.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, when the instructions are executed on a computer, cause the computer to perform the steps of the method for testing a display module.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (randomacceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments is still modified or some technical features thereof are replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The method for testing the display module is characterized by comprising the following steps of:
acquiring first test data of a plurality of first display modules, and classifying the first test data to obtain a plurality of second test data;
respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model;
obtaining target test data corresponding to a second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data;
Inputting the image quality test data and the response time test data into the target test meta-model for display module test data analysis to obtain a plurality of test data analysis results;
and integrating the results of the analysis of the plurality of test data, and outputting a target test result corresponding to the second display module through the target test meta-model.
2. The method for testing a display module according to claim 1, wherein the obtaining the first test data of the plurality of first display modules and classifying the test data of the first test data to obtain the plurality of second test data includes:
acquiring a plurality of first display modules, determining the number of the modules of the plurality of first display modules, acquiring first test data of each display module, and transmitting the first test data to a preset first test system;
carrying out data cleaning on the first test data of each display module through the first test system to obtain cleaned first test data;
performing repeated value elimination and missing value filling on the cleaned first test data to obtain preprocessed first test data;
Performing feature classification on the preprocessed first test data according to a preset feature classification index to obtain a plurality of second test data, wherein the feature classification index comprises: resolution, color depth, contrast, and response time.
3. The method for testing a display module according to claim 1, wherein the constructing a test data base model of each second test data respectively includes:
performing data division on the plurality of second test data to obtain characteristic data and tag data;
constructing a plurality of decision trees based on the characteristic data, and generating a random forest according to the decision trees;
and classifying and carrying out regression training on the random forest to obtain a test data base model of each second test data.
4. A method of testing a display module according to claim 3, wherein the model integrating the test data base model to generate a target test meta-model comprises:
evaluating the test data base model through the tag data to obtain model performance evaluation parameters, and selecting an optimal model architecture according to the model performance evaluation parameters;
According to the model architecture, carrying out model series connection on the test data base model to obtain a stacking model;
and performing model architecture and weight optimization on the stacking model to generate a target test meta-model.
5. The method for testing a display module according to claim 4, wherein the obtaining target test data corresponding to the second display module to be processed, and extracting features of the target test data, obtain image quality test data and response time test data, includes:
acquiring target test data corresponding to a second display module to be processed based on a preset second test system;
performing image feature recognition on the target test data to obtain initial image test data, and performing response feature extraction on the target test data to obtain initial response test data;
and respectively carrying out feature normalization processing on the initial image test data and the initial response test data to obtain image quality test data and response time test data.
6. The method for testing a display module according to claim 5, wherein inputting the image quality test data and the response time test data into the target test meta-model for display module test data analysis, obtaining a plurality of test data analysis results, comprises:
Inputting the image quality test data and the response time test data into the target test meta-model;
performing display module test data analysis through a plurality of test data base models in the target test meta-model to obtain a test data analysis result corresponding to each test data base model;
and taking the test data analysis result corresponding to each test data base model as the target test meta-model and outputting the test data analysis result to obtain a plurality of test data analysis results.
7. The method for testing a display module according to claim 6, wherein the integrating the results of the analysis of the plurality of test data and outputting the target test result corresponding to the second display module through the target test meta-model comprises:
acquiring weight evaluation coefficients of the plurality of test data base models based on the model architecture;
performing result integration on the test data analysis results corresponding to each test data base model according to the weight evaluation coefficients;
and outputting a target test result corresponding to the second display module through the target test meta model, and transmitting the target test result to the second test system.
8. The utility model provides a testing arrangement of display module assembly, its characterized in that, testing arrangement of display module assembly includes:
the device comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring first test data of a plurality of first display modules and classifying the first test data to obtain a plurality of second test data;
the integration module is used for respectively constructing a test data base model of each second test data, and carrying out model integration on the test data base model to generate a target test meta-model;
the extraction module is used for acquiring target test data corresponding to the second display module to be processed, and extracting characteristics of the target test data to obtain image quality test data and response time test data;
the analysis module is used for inputting the image quality test data and the response time test data into the target test meta-model to perform display module test data analysis to obtain a plurality of test data analysis results;
and the integration module is used for integrating the results of the analysis results of the plurality of test data and outputting the target test result corresponding to the second display module through the target test meta-model.
9. A test apparatus for a display module, the test apparatus comprising: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invoking the instructions in the memory to cause the testing device of the display module to perform the method of testing a display module of any of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement a method of testing a display module according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310688591.5A CN116416884B (en) | 2023-06-12 | 2023-06-12 | Testing device and testing method for display module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310688591.5A CN116416884B (en) | 2023-06-12 | 2023-06-12 | Testing device and testing method for display module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116416884A true CN116416884A (en) | 2023-07-11 |
| CN116416884B CN116416884B (en) | 2023-08-18 |
Family
ID=87054726
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310688591.5A Active CN116416884B (en) | 2023-06-12 | 2023-06-12 | Testing device and testing method for display module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116416884B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664551A (en) * | 2023-07-21 | 2023-08-29 | 深圳市长荣科机电设备有限公司 | Display screen detection method, device, equipment and storage medium based on machine vision |
| CN117558220A (en) * | 2024-01-09 | 2024-02-13 | 四川信特农牧科技有限公司 | Liquid crystal display quality monitoring system and method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107396095A (en) * | 2017-08-28 | 2017-11-24 | 方玉明 | One kind is without with reference to three-dimensional image quality evaluation method |
| CN108873401A (en) * | 2018-06-22 | 2018-11-23 | 西安电子科技大学 | Liquid crystal display response time prediction technique based on big data |
| CN109087281A (en) * | 2018-07-02 | 2018-12-25 | 北京百度网讯科技有限公司 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
| CN110796200A (en) * | 2019-10-30 | 2020-02-14 | 深圳前海微众银行股份有限公司 | Data classification method, terminal, device and storage medium |
| CN114091360A (en) * | 2022-01-21 | 2022-02-25 | 武汉格蓝若智能技术有限公司 | Multi-model fused voltage transformer error state evaluation method |
| CN114118192A (en) * | 2020-09-01 | 2022-03-01 | 中国移动通信有限公司研究院 | Training method, prediction method, device and storage medium of user prediction model |
| CN114997051A (en) * | 2022-05-30 | 2022-09-02 | 浙大城市学院 | Aero-engine service life prediction and health assessment method based on transfer learning |
-
2023
- 2023-06-12 CN CN202310688591.5A patent/CN116416884B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107396095A (en) * | 2017-08-28 | 2017-11-24 | 方玉明 | One kind is without with reference to three-dimensional image quality evaluation method |
| CN108873401A (en) * | 2018-06-22 | 2018-11-23 | 西安电子科技大学 | Liquid crystal display response time prediction technique based on big data |
| CN109087281A (en) * | 2018-07-02 | 2018-12-25 | 北京百度网讯科技有限公司 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
| CN110796200A (en) * | 2019-10-30 | 2020-02-14 | 深圳前海微众银行股份有限公司 | Data classification method, terminal, device and storage medium |
| CN114118192A (en) * | 2020-09-01 | 2022-03-01 | 中国移动通信有限公司研究院 | Training method, prediction method, device and storage medium of user prediction model |
| CN114091360A (en) * | 2022-01-21 | 2022-02-25 | 武汉格蓝若智能技术有限公司 | Multi-model fused voltage transformer error state evaluation method |
| CN114997051A (en) * | 2022-05-30 | 2022-09-02 | 浙大城市学院 | Aero-engine service life prediction and health assessment method based on transfer learning |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664551A (en) * | 2023-07-21 | 2023-08-29 | 深圳市长荣科机电设备有限公司 | Display screen detection method, device, equipment and storage medium based on machine vision |
| CN116664551B (en) * | 2023-07-21 | 2023-10-31 | 深圳市长荣科机电设备有限公司 | Display screen detection method, device, equipment and storage medium based on machine vision |
| CN117558220A (en) * | 2024-01-09 | 2024-02-13 | 四川信特农牧科技有限公司 | Liquid crystal display quality monitoring system and method |
| CN117558220B (en) * | 2024-01-09 | 2024-04-05 | 四川信特农牧科技有限公司 | Liquid crystal display quality monitoring system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116416884B (en) | 2023-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111178456B (en) | Abnormal index detection method and device, computer equipment and storage medium | |
| CN116416884B (en) | Testing device and testing method for display module | |
| US7627620B2 (en) | Data-centric automatic data mining | |
| CN112087443B (en) | Sensing data anomaly detection method under physical attack of industrial sensing network information | |
| CN112288021A (en) | Medical wastewater monitoring data quality control method, device and system | |
| CN114861788A (en) | Load abnormity detection method and system based on DBSCAN clustering | |
| CN111242387A (en) | Talent departure prediction method and device, electronic equipment and storage medium | |
| CN107944005B (en) | Data display method and device | |
| CN112580780A (en) | Model training processing method, device, equipment and storage medium | |
| CN117235524A (en) | Learning training platform of automatic valuation model | |
| CN110320802B (en) | Complex system signal time sequence identification method based on data visualization | |
| CN116451081A (en) | Data drift detection method, device, terminal and storage medium | |
| CN116861331A (en) | Expert model decision-fused data identification method and system | |
| US11715204B2 (en) | Adaptive machine learning system for image-based biological sample constituent analysis | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| CN113935413A (en) | Distribution network wave recording file waveform identification method based on convolutional neural network | |
| CN117015812A (en) | System for clustering data points | |
| Guidi et al. | A new procedure to optimize the selection of groups in a classification tree: Applications for ecological data | |
| CN114004138A (en) | Building monitoring method and system based on big data artificial intelligence and storage medium | |
| CN112580781A (en) | Processing method, device and equipment of deep learning model and storage medium | |
| Zabary et al. | A MATLAB pipeline for spatiotemporal quantification of monolayer cell migration | |
| CN117076454B (en) | Engineering quality acceptance form data structured storage method and system | |
| CN115831339B (en) | Medical system risk management and control pre-prediction method and system based on deep learning | |
| WO2023181230A1 (en) | Model analysis device, model analysis method, and recording medium | |
| CN116863481A (en) | Service session risk processing method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A testing device and testing method for a display module Effective date of registration: 20231208 Granted publication date: 20230818 Pledgee: Shenzhen high tech investment and financing Company limited by guarantee Pledgor: SHENZHEN OSTAR DISPLAY ELECTRONICS CO.,LTD. Registration number: Y2023980070323 |