Aging test method and system for display panel

Abstract

The invention relates to the technical field of display panels, in particular to an aging test method and system for a display panel. The method comprises the following steps: firstly, processing image data by using a display panel feature recognition model to obtain brightness histogram data, color histogram data and defect feature data; then, the response duration data, the brightness histogram data and the color histogram data are evaluated, and a first display panel test value is obtained; evaluating the power data and the defect characteristic data to obtain a second display panel test value; finally, weighting the first display panel test value and the second display panel test value to obtain an aging test value, and comparing the aging test value with an aging test threshold value; if the threshold value is exceeded, judging that the product is unqualified; otherwise, inputting the aging test data into the test environment prediction model, and generating optimal test environment information according to a model output result; according to the invention, the accuracy and reliability of the aging test of the display panel can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of display panels, and in particular to an aging test method and system for display panels. Background Art

[0002] Display panels are hardware devices used to display image or video content and are widely used in electronic devices such as TVs, computer monitors and smartphones.

[0003] After the display panel is produced, it is usually necessary to test the various performance and parameters of the display panel to eliminate defective products that do not meet the requirements. Therefore, aging testing of the display panel is an essential operating procedure.

[0004] At present, the existing technology still has some deficiencies in the aging test of display panels. On the one hand, the existing technology does not take into account the aging factors of the display panel such as surface defects, display effects and power conversion, which will reduce the reliability and accuracy of the display panel aging test; on the other hand, the existing technology does not take into account the impact of the test environment on the aging test process of the display panel, and lacks testing and evaluation of environmental conditions, which will also reduce the reliability and accuracy of the display panel aging test.

[0005] To this end, a display panel aging test method and system are proposed. Summary of the invention

[0006] The object of the present invention is to provide an aging test method and system for a display panel. First, image data, power data and response time data are obtained; then the image data is processed using a display panel feature recognition model to obtain brightness histogram data, color histogram data and defect feature data; then, the brightness histogram data, the color histogram data and the response time data are evaluated to obtain a first display panel test value; the defect feature data and the power data are evaluated to obtain a second display panel test value; finally, the first display panel test value and the second display panel test value are weighted to obtain an aging test value and compared with an aging test threshold; if the threshold is exceeded, it is judged as unqualified; otherwise, the aging test data is input into a test environment prediction model, and optimal test environment information is generated according to the model output result.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A display panel aging test method, comprising:

[0009] Acquire image data, power data and response time data of the display panel;

[0010] Using a display panel feature recognition model to perform recognition processing on the image data to obtain brightness histogram data, color histogram data and defect feature data;

[0011] Evaluate the brightness histogram data, the color histogram data and the response time data to obtain a first display panel test value;

[0012] Evaluate the defect characteristic data and the power data to obtain a second display panel test value;

[0013] The first display panel test value and the second display panel test value are weighted to obtain a display panel aging test value; the calculation formula of the display panel aging test value is:

[0014] LHCS=T S *[ω*FICS+(1-ω)*SECS];

[0015]

[0016] Wherein, LHCS represents the aging test value of the display panel; T S represents the weight of the power-on working state; ω represents the weight of the first display panel test value; FICS represents the first display panel test value; SECS represents the second display panel test value;

[0017] The aging test value of the display panel is compared with the aging test threshold. If it exceeds the aging test threshold, it is judged as unqualified; otherwise, the environmental prediction parameters and the aging test data are input into the pre-trained test environment prediction model for parameter update, and the optimal test environment information is generated according to the data results of the test environment prediction model.

[0018] Furthermore, the specific implementation process of using the display panel feature recognition model to perform recognition processing on the image data to obtain brightness histogram data, color histogram data and defect feature data includes:

[0019] Obtain historical preprocessed sample image data;

[0020] Further, a camera is used to collect images from the display panel to obtain image data;

[0021] Further, the image data is preprocessed to obtain preprocessed image data;

[0022] Further, a display panel feature recognition model is constructed, and the historical pre-processed sample image data is input into the display panel feature recognition model for training to obtain a final display panel feature recognition model;

[0023] Furthermore, the pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are output.

[0024] Furthermore, the specific implementation process of evaluating the brightness histogram data, the color histogram data and the response time data to obtain the first display panel test value includes:

[0025] Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data;

[0026] Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value;

[0027] The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using a brightness weight factor, a color weight factor and a response time weight factor to obtain a first display panel test value.

[0028] Furthermore, the specific implementation process of evaluating the defect characteristic data and the power data to obtain the second display panel test value includes:

[0029] Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features;

[0030] Evaluating the image defect feature, the panel defect feature and the power data to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value;

[0031] Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation.

[0032] Furthermore, the environmental prediction parameters and the aging test data are input into the pre-trained test environment prediction model for parameter update, and the specific implementation process of generating the optimal test environment information according to the data results of the test environment prediction model includes:

[0033] Obtain historical test environment data, historical aging test data, and current aging test data;

[0034] Further, the historical test environment data and the historical aging test data are input into a test environment prediction model for training to obtain environment prediction parameters;

[0035] Further, the environmental prediction parameters and the current aging test data are input into the test environment prediction model, the model parameters are updated, and the real-time environmental model weights are output; wherein the real-time environmental model weights include: real-time illumination environmental parameters, real-time temperature environmental parameters, and real-time humidity environmental parameters;

[0036] Furthermore, a Hadamard product operation is performed on the real-time environment information and the real-time environment model weight to obtain optimal test environment information.

[0037] A display panel aging test system comprises: a system control module, a data acquisition module, a data processing module, an aging test module and an output module; wherein the system control module is used to control the start, pause and stop of the system; the data acquisition module is used to receive the collected image data, power data and response time data; the data processing module is used to preprocess the acquired data; the data processing module uses a display panel feature recognition model to perform feature recognition on the preprocessed image; the aging test module is used to perform an aging test using the feature recognition result, the power data and the response time data, and output an aging test value; the output module comprises a judgment unit and a feedback unit; the judgment unit is used to output an aging test judgment result according to the aging test value; the feedback unit is used to provide feedback to a staff member according to the aging test judgment result.

[0038] Furthermore, the specific implementation process of the data processing module performing feature recognition on the pre-processed image using the display panel feature recognition model includes:

[0039] Obtain historical preprocessed sample image data;

[0040] Further, a camera is used to collect images from the display panel to obtain image data;

[0041] Further, the image data is preprocessed to obtain preprocessed image data;

[0042] Further, a display panel feature recognition model is constructed, and the historical pre-processed sample image data is input into the display panel feature recognition model for training to obtain a final display panel feature recognition model;

[0043] Furthermore, the pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are output.

[0044] Furthermore, the aging test module evaluates the brightness histogram data, the color histogram data and the response time data to obtain a specific implementation process of the first display panel test value, including:

[0045] Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data;

[0046] Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value;

[0047] The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using a brightness weight factor, a color weight factor and a response time weight factor to obtain a first display panel test value.

[0048] Furthermore, the aging test module evaluates the defect characteristic data and the power data to obtain a specific implementation process of the second display panel test value including:

[0049] Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features;

[0050] Evaluating the image defect feature, the panel defect feature and the power data to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value;

[0051] Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation.

[0052] Further, the judgment unit compares the aging test value with the aging test threshold, and if it exceeds the aging test threshold, it is judged as unqualified; otherwise, the aging test data is input into the test environment prediction model, and the specific process of generating the optimal test environment information according to the data result of the test environment prediction model includes:

[0053] Obtain historical test environment data, historical aging test data, and current aging test data;

[0054] Further, the historical test environment data and the historical aging test data are input into a test environment prediction model for training to obtain environment prediction parameters;

[0055] Further, the environmental prediction parameters and the current aging test data are input into the test environment prediction model, the model parameters are updated, and the real-time environmental model weights are output; wherein the real-time environmental model weights include: real-time illumination environment weights, real-time temperature environment weights, and real-time humidity environment weights;

[0056] Furthermore, a Hadamard product operation is performed on the real-time environment information and the real-time environment model weight to obtain optimal test environment information.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. The present invention proposes a visual aging test function for testing the aging degree of visual display effects; the function recognizes and processes the display panel image through a display panel feature recognition model to obtain visual display effect data; the visual display effect data records the distribution of the brightness and color of the picture; the function obtains a first display panel test value by evaluating the brightness histogram data, the color histogram data and the response time data; the first display panel test value reflects the aging degree of the display panel in terms of visual effects; the function can effectively improve the reliability and accuracy of the display panel aging test.

[0059] 2. The present invention proposes a defect performance aging test function for testing defect characteristics and the degree of aging in performance; this function identifies and processes the display panel image through a display panel feature recognition model to obtain defect feature data; the defect feature data records the characteristic distribution of image defects and panel surface defects; this function obtains a second display panel test value by evaluating the defect feature data and power data; the second display panel test value reflects the degree of aging of the display panel in terms of defect performance; this function can effectively improve the reliability and accuracy of the display panel aging test.

[0060] 3. The present invention proposes an aging test environment prediction function for obtaining optimal aging test environment conditions; this function obtains environment prediction parameters by training a test environment prediction model using historical test environment data and historical aging test data; the environment prediction parameters record changes in light, temperature and humidity in the test environment; this function obtains optimal test environment information by performing a Hadamard product operation on real-time environment information and real-time environment model weights; the real-time environment model weights are obtained by inputting the environment prediction parameters and current aging test data into the test environment prediction model; this function can adaptively adjust the test environment conditions, thereby effectively improving the reliability and accuracy of the display panel aging test. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of a flow chart of an aging test method for a display panel according to the present invention;

[0062] Figure 2 It is a structural schematic diagram of a display panel feature recognition model of the present invention;

[0063] Figure 3 It is a structural schematic diagram of the test environment prediction model of the present invention;

[0064] Figure 4 The figure is a schematic structural diagram of an aging test system for a display panel according to the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Display panels are hardware devices used to display image or video content and are widely used in electronic devices such as TVs, computer monitors and smartphones.

[0067] After the display panel is produced, it is usually necessary to test the various performance and parameters of the display panel to eliminate defective products that do not meet the requirements. Therefore, aging testing of the display panel is an essential operating procedure.

[0068] At present, the existing technology still has some deficiencies in the aging test of display panels. On the one hand, the existing technology tests the circuit conditions under high voltage, and does not take into account the aging factors of the display panel such as surface defects, display effects and power conversion, which will reduce the reliability and accuracy of the display panel aging test; on the other hand, the existing technology does not take into account the impact of the test environment on the aging test process of the display panel, and lacks testing and evaluation of environmental conditions, which will also reduce the reliability and accuracy of the display panel aging test.

[0069] Embodiment 1

[0070] The specific implementation process in the embodiment of the present application is implemented by an aging test method for a display panel of the present invention, see Figure 1 The process of the method proposed in the present invention is described in the following; the aging test method of a display panel comprises:

[0071] S10. Obtaining image data, power data and response time data of the display panel;

[0072] S20. Inputting the image data into a display panel feature recognition model to obtain brightness histogram data, color histogram data and defect feature data;

[0073] S30. Evaluate the brightness histogram data, the color histogram data and the response time data to obtain a first display panel test value;

[0074] S40. Evaluate the defect characteristic data and the power data to obtain a second display panel test value;

[0075] S50. Weighting the first display panel test value and the second display panel test value to obtain a display panel aging test value;

[0076] S60. Compare the aging test value of the display panel with the aging test threshold. If it exceeds the aging test threshold, it is judged as unqualified; otherwise, input the aging test data into the test environment prediction model, and generate the optimal test environment information according to the data results of the test environment prediction model.

[0077] Furthermore, a specific implementation process of a display panel aging test method is as follows:

[0078] Acquire image data, power data and response time data of the display panel;

[0079] Using a display panel feature recognition model to perform recognition processing on the image data to obtain brightness histogram data, color histogram data and defect feature data;

[0080] Evaluate the brightness histogram data, the color histogram data and the response time data to obtain a first display panel test value;

[0081] Evaluate the defect characteristic data and the power data to obtain a second display panel test value;

[0082] The first display panel test value and the second display panel test value are weighted to obtain a display panel aging test value; the calculation formula of the display panel aging test value is:

[0083] LHCS=T S *[ω*FICS+(1-ω)*SECS];

[0084]

[0085] Wherein, LHCS represents the aging test value of the display panel; T Srepresents the weight of the power-on working state; ω represents the weight of the first display panel test value; FICS represents the first display panel test value; SECS represents the second display panel test value;

[0086] The aging test value of the display panel is compared with the aging test threshold. If it exceeds the aging test threshold, it is judged as unqualified; otherwise, the environmental prediction parameters and the aging test data are input into the pre-trained test environment prediction model for parameter update, and the optimal test environment information is generated according to the data results of the test environment prediction model.

[0087] In this embodiment, a display panel aging test method is proposed. First, image data, power data and response time data are obtained; then, the image data is processed using a display panel feature recognition model to obtain brightness histogram data, color histogram data and defect feature data; then, the brightness histogram data, the color histogram data and the response time data are evaluated to obtain a first display panel test value; the defect feature data and the power data are evaluated to obtain a second display panel test value; finally, the first display panel test value and the second display panel test value are weighted to obtain an aging test value and compared with an aging test threshold; if the threshold is exceeded, it is judged as unqualified; otherwise, the aging test data is input into a test environment prediction model, and optimal test environment information is generated according to the model output result; this method can effectively improve the reliability and accuracy of the display panel aging test.

[0088] For specific explanation, the present invention is described in conjunction with the following embodiments, as follows:

[0089] Acquire image data, power data and response time data of the display panel;

[0090] In this embodiment, the image data is collected by a camera; the power data and the response time data are acquired by a sensor; wherein the power data and the response time data are average numerical data obtained by collecting five times and calculating.

[0091] Furthermore, the specific implementation process of using the display panel feature recognition model to perform recognition processing on the image data to obtain brightness histogram data, color histogram data and defect feature data includes:

[0092] Obtain historical preprocessed sample image data;

[0093] Further, a camera is used to collect images from the display panel to obtain image data;

[0094] Further, the image data is preprocessed to obtain preprocessed image data;

[0095] In this embodiment, the preprocessing process includes Gaussian denoising, edge enhancement and normalization operations; wherein the Gaussian denoising is used to suppress noise to improve the recognition accuracy of the model; the edge enhancement is used to enhance the high-frequency details of the image to improve the accuracy of the model in identifying some high-frequency defect features; the edge enhancement algorithm can use the Sobel operator, Prewitt operator and Canny operator, etc.; the normalization operation is used to compress the data into the same interval to improve the calculation efficiency of the model.

[0096] Further, a display panel feature recognition model is constructed, and the historical pre-processed sample image data is input into the display panel feature recognition model for training to obtain a final display panel feature recognition model;

[0097] Further, the pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are obtained as output;

[0098] The structure of the display panel feature recognition model is as follows: Figure 2 As shown, it includes: convolution layer, feature extraction layer, feature labeling layer, feature attention layer, feature recognition layer and output layer; the specific implementation process of the model includes:

[0099] Inputting the image data into the convolution layer to obtain a display panel feature map;

[0100] Further, the display panel feature map is input into the feature extraction layer to obtain the display panel convolution feature; wherein the feature extraction layer includes 5 residual blocks for extracting the convolution feature;

[0101] Further, the display panel convolution feature is input into the feature marking layer to obtain an anchor frame; wherein the anchor frame is set to five different sizes;

[0102] Further, extracting features from the anchor frame and inputting them into the feature attention layer to obtain enhanced display panel features; wherein the feature attention layer is a dual-branch layer, including: a channel attention layer and a spatial attention layer;

[0103] Further, the enhanced display panel feature is input into the feature recognition layer to obtain a recognition result;

[0104] Furthermore, the recognition result is input into the output layer to obtain recognition data; wherein the recognition data includes: brightness histogram data, color histogram data and defect feature data.

[0105] In this embodiment, a display panel feature recognition model is proposed to obtain brightness histogram data, color histogram data and defect feature data; the model includes: a convolution layer, a feature extraction layer, a feature marking layer, a feature attention layer, a feature recognition layer and an output layer; the model can provide accurate data support for subsequent test calculations, further improving the reliability and accuracy of the display panel aging test.

[0106] Furthermore, the specific implementation process of evaluating the brightness histogram data, the color histogram data and the response time data to obtain the first display panel test value includes:

[0107] Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data;

[0108] Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value;

[0109] The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using the brightness weight factor, the color weight factor and the response time weight factor to obtain a first display panel test value; wherein the calculation formula of the first display panel test value is:

[0110]

[0111] α1+α2+α3=1.0;

[0112] Wherein, FICS represents the first display panel test value, and the smaller the value, the closer the visual effect of the display panel is to the standard effect; α1 represents the brightness weight factor; n represents the number of image data; hld i hld is the brightness histogram data of the i-th image; th It is represented as the standard brightness histogram data; α2 is represented as the color weight factor; hsc j The color histogram data represented as the j-th image; hsc th is represented by the standard color histogram data; α3 is represented by the response time weight factor; m is represented by the number of response time measurements; trs k Represented as the kth response duration data; trs th Represented as the standard response time data.

[0113] In this embodiment, the brightness weight factor, color weight factor and response time weight factor are set to 0.3, 0.35 and 0.35 respectively; the setting of the weight factor can be flexibly adjusted according to actual conditions.

[0114] In this embodiment, the standard brightness histogram data, standard color histogram data and standard response time data are set by technicians in this field based on historical experience and display panel parameters. Therefore, the setting of standard values ​​is not unique and can be adjusted according to actual conditions.

[0115] In this embodiment, a visual aging test function is proposed to test the aging degree of visual display effects; this function recognizes and processes the display panel image through a display panel feature recognition model to obtain visual display effect data; the visual display effect data records the distribution of the brightness and color of the picture; this function obtains a first display panel test value by evaluating the brightness histogram data, the color histogram data and the response time data; the first display panel test value reflects the aging degree of the display panel in terms of visual effects; this function can effectively improve the reliability and accuracy of the display panel aging test.

[0116] Furthermore, the specific implementation process of evaluating the defect characteristic data and the power data to obtain the second display panel test value includes:

[0117] Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features;

[0118] The image defect characteristics, the panel defect characteristics and the power data are evaluated to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value; wherein the image defect characteristics include: afterimage characteristics and screen bright and dark spot characteristics; the panel defect characteristics include: scratch characteristics and bubble characteristics;

[0119] Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation; wherein the calculation formula of the second display panel test value is:

[0120]

[0121] β1+β2+β3+β4=1.0;

[0122] Wherein, SECS represents the test value of the second display panel. The smaller the value, the closer the quality of the display panel is to the standard requirement. N represents the number of image feature samples, that is, the number of display images and panel images is consistent. β1 represents the residual image feature coefficient. cx jIt is expressed as the maximum number of residual images of the jth sample; It is expressed as the j1th residual image area of ​​the jth sample; β2 is expressed as the bright and dark point characteristic coefficient; pild j Expressed as the maximum number of bright spots in the jth sample; piad j Expressed as the maximum number of dark spots in the jth sample; pizs j It is represented by the total number of screen pixels of the jth sample; β3 is represented by the scratch characteristic coefficient; hh j It is expressed as the maximum number of scratches of the jth sample; It is expressed as the length of the j3rd scratch of the jth sample; It is represented by the depth of the j3rd scratch of the jth sample; β4 is represented by the bubble characteristic coefficient; qp j Expressed as the maximum number of bubbles in the jth sample; is the diameter of the j4th bubble of the jth sample; M is the number of times the power conversion efficiency is tested; η i Expressed as the i-th power conversion rate; wherein the power conversion rate is the ratio of output power to input power; η th Expressed as the threshold value of power conversion rate.

[0123] In this embodiment, the settings of the afterimage characteristic coefficient, the bright and dark spot characteristic coefficient, the scratch characteristic coefficient and the bubble characteristic coefficient are set according to the degree of influence on the display panel aging test, and are set to 0.3, 0.3, 0.2 and 0.2 respectively; the settings of different characteristic coefficients can be adjusted according to actual conditions.

[0124] In this embodiment, the thresholds of the image defect evaluation value, the panel defect evaluation value and the power conversion rate are set by technicians in this field based on historical experience and display panel parameters. Therefore, the setting of the standard value is not unique and can be adjusted according to actual conditions.

[0125] In this embodiment, a defect performance aging test function is proposed to test the aging degree of defect characteristics and performance; this function recognizes and processes the display panel image through a display panel feature recognition model to obtain defect feature data; the defect feature data records the characteristic distribution of image defects and panel surface defects; this function obtains a second display panel test value by evaluating the defect feature data and power data; the second display panel test value reflects the aging degree of the display panel in terms of defect performance; this function can effectively improve the reliability and accuracy of the display panel aging test.

[0126] Furthermore, the first display panel test value and the second display panel test value are weighted to obtain a display panel aging test value; the calculation formula of the display panel aging test value is:

[0127] LHCS=T S *[ω*FICS+(1-ω)*SECS];

[0128]

[0129] Wherein, LHCS represents the aging test value of the display panel; T S It represents the weight of the power-on working state; ω represents the weight of the test value of the first display panel; the weight is set to 0.5; FICS represents the test value of the first display panel; SECS represents the test value of the second display panel.

[0130] Furthermore, the environmental prediction parameters and aging test data are input into the pre-trained test environment prediction model for parameter update, and the specific implementation process of generating the optimal test environment information according to the data results of the test environment prediction model is as follows:

[0131] Obtain historical test environment data, historical aging test data, and current aging test data;

[0132] Further, the historical test environment data and the historical aging test data are input into a test environment prediction model for training to obtain environment prediction parameters;

[0133] The test environment prediction model in this embodiment uses a hybrid network combining CNN and Transformer. The structure of the model is as follows: Figure 3 As shown in the figure; the hybrid network adopts a dual-branch structure, and the input data is input into the CNN and Transformer sub-networks for feature processing respectively, and finally the output features of different networks are fused and output to obtain the prediction results; CNN uses three residual blocks to extract spatial convolution features, and uses convolution operations to learn the contextual relationship between aging test data and environmental data; Transformer uses three Transformer layers to extract long-distance dependency features, and uses the self-attention mechanism to learn the global relationship between data.

[0134] Further, the environmental prediction parameters and the current aging test data are input into the test environment prediction model, the model parameters are updated, and the real-time environmental model weights are output; wherein the real-time environmental model weights include: real-time illumination environment weights, real-time temperature environment weights, and real-time humidity environment weights;

[0135] Furthermore, the real-time environment information and the real-time environment model weight are subjected to a Hadamard product operation to obtain the optimal test environment information; wherein the calculation formula of the optimal test environment information is:

[0136] PEI=Had(PE ,I E )=(ω gz Inf gz ,ω wd Inf wd ,ω sd Inf sd );

[0137] P E =(ω gz ,ω wd ,ω sd );

[0138] I E =(Inf gz ,Inf wd ,Inf sd );

[0139] Wherein, PEI represents the optimal test environment information; Had represents the Hadamard product operation; P E I is represented by the real-time environment model weight; E Represented as the real-time environmental information; ω gz Represented as the real-time lighting environment weight; Inf gz Represented as real-time lighting information; ω wd Inf is the real-time temperature environment weight; wd Represented as real-time temperature information; ω sd Inf is the real-time humidity environment weight; sd Represents real-time humidity information.

[0140] In this embodiment, an aging test environment prediction function is proposed to obtain the optimal aging test environment conditions; this function obtains environment prediction parameters by training the test environment prediction model using historical test environment data and historical aging test data; the environment prediction parameters record the changes in light, temperature and humidity of the test environment; this function obtains the optimal test environment information by performing Hadamard product operation on real-time environment information and real-time environment model weights; the real-time environment model weights are obtained by inputting the environment prediction parameters and current aging test data into the test environment prediction model; this function can adaptively adjust the test environment conditions, thereby effectively improving the reliability and accuracy of the display panel aging test.

[0141] Embodiment 2

[0142] As an embodiment of the present invention, refer to Figure 4 , an aging test system for a display panel, comprising: a system control module, a data acquisition module, a data processing module, an aging test module and an output module;

[0143] Wherein, the system control module is used to control the start, pause and stop of the system;

[0144] The data acquisition module is used to receive the collected image data, power data and response time data;

[0145] The data processing module is used to pre-process the acquired data;

[0146] The data processing module uses a display panel feature recognition model to perform feature recognition on the pre-processed image;

[0147] Furthermore, the specific implementation process of the data processing module performing feature recognition on the pre-processed image using the display panel feature recognition model includes:

[0148] Obtain historical preprocessed sample image data;

[0149] Further, a camera is used to collect images from the display panel to obtain image data;

[0150] Further, the image data is preprocessed to obtain preprocessed image data;

[0151] Further, a display panel feature recognition model is constructed, and the historical pre-processed sample image data is input into the display panel feature recognition model for training to obtain a final display panel feature recognition model;

[0152] Furthermore, the pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are output.

[0153] The aging test module is used to perform an aging test using the feature recognition result, the power data and the response duration data, and output an aging test value;

[0154] Furthermore, the aging test module evaluates the brightness histogram data, the color histogram data and the response time data to obtain a specific implementation process of the first display panel test value, including:

[0155] Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data;

[0156] Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value;

[0157] The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using the brightness weight factor, the color weight factor and the response time weight factor to obtain a first display panel test value; wherein the calculation formula of the first display panel test value is:

[0158]

[0159] α1+α2+α3=1.0;

[0160] Wherein, FICS represents the first display panel test value, and the smaller the value, the closer the visual effect of the display panel is to the standard effect; α1 represents the brightness weight factor; n represents the number of image data; hld i hld is the brightness histogram data of the i-th image; th It is represented as the standard brightness histogram data; α2 is represented as the color weight factor; hsc j The color histogram data represented as the j-th image; hsc th is represented by the standard color histogram data; α3 is represented by the response time weight factor; m is represented by the number of response time measurements; trs k Represented as the kth response duration data; trs th Represented as the standard response time data.

[0161] Furthermore, the aging test module evaluates the defect characteristic data and the power data to obtain a specific implementation process of the second display panel test value including:

[0162] Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features;

[0163] Evaluating the image defect feature, the panel defect feature and the power data to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value;

[0164] Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation; wherein the calculation formula of the second display panel test value is:

[0165]

[0166] β1+β2+β3+β4=1.0;

[0167] Wherein, SECS represents the test value of the second display panel. The smaller the value, the closer the quality of the display panel is to the standard requirement. N represents the number of image feature samples, that is, the number of display images and panel images is consistent. β1 represents the residual image feature coefficient. cx j It is expressed as the maximum number of residual images of the jth sample; It is expressed as the j1th residual image area of ​​the jth sample; β2 is expressed as the bright and dark point characteristic coefficient; pild j Expressed as the maximum number of bright spots in the jth sample; piad j Expressed as the maximum number of dark spots in the jth sample; pizs j It is represented by the total number of screen pixels of the jth sample; β3 is represented by the scratch characteristic coefficient; hh j It is expressed as the maximum number of scratches of the jth sample; It is expressed as the length of the j3rd scratch of the jth sample; It is represented by the depth of the j3rd scratch of the jth sample; β4 is represented by the bubble characteristic coefficient; qp j Expressed as the maximum number of bubbles in the jth sample; is the diameter of the j4th bubble of the jth sample; M is the number of times the power conversion efficiency is tested; η i Expressed as the i-th power conversion rate; wherein the power conversion rate is the ratio of output power to input power; η th Expressed as the threshold value of power conversion rate.

[0168] Furthermore, the first display panel test value and the second display panel test value are weighted to obtain a display panel aging test value; the calculation formula of the display panel aging test value is:

[0169] LHCS=T S *[ω*FICS+(1-ω)*SECS];

[0170]

[0171] Wherein, LHCS represents the aging test value of the display panel; T S It represents the weight of the power-on working state; ω represents the weight of the test value of the first display panel; FICS represents the test value of the first display panel; SECS represents the test value of the second display panel.

[0172] The output module includes a judgment unit and a feedback unit; the judgment unit is used to output an aging test judgment result according to the aging test value; and the feedback unit is used to provide feedback to the staff according to the aging test judgment result.

[0173] Furthermore, the judgment unit compares the display panel aging test value with the aging test threshold. If the display panel aging test value is equal to 0 or exceeds the aging test threshold, it is judged as unqualified; otherwise, the aging test data is input into the test environment prediction model, and the optimal test environment information is generated according to the data results of the test environment prediction model.

[0174] Furthermore, the specific implementation process of obtaining the optimal test environment information includes:

[0175] Obtain historical test environment data, historical aging test data, and current aging test data;

[0176] Further, the historical test environment data and the historical aging test data are input into a test environment prediction model for training to obtain environment prediction parameters;

[0177] Further, the environmental prediction parameters and the current aging test data are input into the test environment prediction model, the model parameters are updated, and the real-time environmental model weights are output; wherein the real-time environmental model weights include: real-time illumination environment weights, real-time temperature environment weights, and real-time humidity environment weights;

[0178] Furthermore, a Hadamard product operation is performed on the real-time environment information and the real-time environment model weight to obtain optimal test environment information.

[0179] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A display panel aging test method, characterized in that: include: Acquire image data, power data and response time data of the display panel; Using a display panel feature recognition model to perform recognition processing on the image data to obtain brightness histogram data, color histogram data and defect feature data; Evaluate the brightness histogram data, the color histogram data and the response time data to obtain a first display panel test value; Evaluating the defect characteristic data and the power data to obtain a second display panel test value; The first display panel test value and the second display panel test value are weighted to obtain an aging test value; the calculation formula of the aging test value is: LHCS=T S *[ω*FICS+(1-ω)*SECS]; Wherein, LHCS represents the aging test value; T S represents the weight of the power-on working state; ω represents the weight of the first display panel test value; FICS represents the first display panel test value; SECS represents the second display panel test value; The aging test value is compared with the aging test threshold. If it exceeds the aging test threshold, it is judged as unqualified; otherwise, the environmental prediction parameters and the aging test data are input into the pre-trained test environment prediction model for parameter update, and the optimal test environment information is generated according to the data results of the test environment prediction model.

2. The aging test method for a display panel according to claim 1, characterized in that: The specific implementation process of using the display panel feature recognition model to recognize and process the image data to obtain brightness histogram data, color histogram data and defect feature data includes: Obtain historical preprocessed sample image data; Use a camera to collect image data from the display panel; Preprocessing the image data to obtain preprocessed image data; Constructing a display panel feature recognition model, inputting the historical preprocessed sample image data into the display panel feature recognition model for training, and obtaining a final display panel feature recognition model; The pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are obtained as output.

3. The aging test method for a display panel according to claim 1, characterized in that: The specific implementation process of evaluating the brightness histogram data, the color histogram data and the response time data to obtain the first display panel test value includes: Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data; Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value; The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using a brightness weight factor, a color weight factor and a response time weight factor to obtain a first display panel test value.

4. The aging test method for a display panel according to claim 1, characterized in that: The specific implementation process of evaluating the defect characteristic data and the power data to obtain the second display panel test value includes: Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features; Evaluating the image defect feature, the panel defect feature and the power data to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value; Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation.

5. The aging test method for a display panel according to claim 1, characterized in that: The specific implementation process of inputting the environmental prediction parameters and the aging test data into the pre-trained test environment prediction model to update the parameters and generating the optimal test environment information according to the data results of the test environment prediction model includes: Obtain historical test environment data, historical aging test data, and current aging test data; Inputting the historical test environment data and the historical aging test data into a test environment prediction model for training to obtain environment prediction parameters; Input the environmental prediction parameters and the current aging test data into the test environment prediction model, update the model parameters, and output the real-time environmental model weights; wherein the real-time environmental model weights include: real-time illumination environmental parameters, real-time temperature environmental parameters, and real-time humidity environmental parameters; The real-time environment information and the real-time environment model weight are subjected to a Hadamard product operation to obtain optimal test environment information.

6. A display panel aging test system, characterized in that: include: System control module, data acquisition module, data processing module, aging test module and output module; wherein, the system control module is used to control the start, pause and stop of the system; the data acquisition module is used to receive the collected image data, power data and response time data; the data processing module is used to pre-process the acquired data; the data processing module uses the display panel feature recognition model to perform feature recognition on the pre-processed image; the aging test module is used to perform aging test using the feature recognition result, the power data and the response time data, and output the aging test value; the output module includes a judgment unit and a feedback unit; the judgment unit is used to output the aging test judgment result according to the aging test value; The feedback unit is used to provide feedback to the staff according to the aging test judgment result.

7. The aging test system for a display panel according to claim 6, characterized in that: The specific implementation process of the data processing module using the display panel feature recognition model to perform feature recognition on the pre-processed image includes: Obtain historical preprocessed sample image data; Use a camera to collect image data from the display panel; Preprocessing the image data to obtain preprocessed image data; Constructing a display panel feature recognition model, inputting the historical preprocessed sample image data into the display panel feature recognition model for training, and obtaining a final display panel feature recognition model; The pre-processed image data is input into the final display panel feature recognition model for recognition processing, and brightness histogram data, color histogram data and defect feature data are obtained as output.

8. The aging test system for a display panel according to claim 6, characterized in that: The aging test module evaluates the brightness histogram data, the color histogram data and the response time data to obtain the first display panel test value. The specific implementation process includes: Obtaining brightness histogram data, color histogram data and response time data and corresponding standard brightness histogram data, standard color histogram data and standard response time data; Comparing the brightness histogram data, the color histogram data and the response time data with the standard brightness histogram data, the standard color histogram data and the standard response time data, respectively, to obtain a brightness histogram evaluation value, a color histogram evaluation value and a response time evaluation value; The brightness histogram evaluation value, the color histogram evaluation value and the response time evaluation value are weighted by using a brightness weight factor, a color weight factor and a response time weight factor to obtain a first display panel test value.

9. The aging test system for a display panel according to claim 6, characterized in that: The specific implementation process of the aging test module evaluating the defect characteristic data and the power data to obtain the second display panel test value includes: Acquire defect feature data and power data; wherein the defect feature data includes: image defect features and panel defect features; Evaluating the image defect feature, the panel defect feature and the power data to obtain an image defect evaluation value, a panel defect evaluation value and a power conversion evaluation value; Corresponding evaluation weights are set for the image defect evaluation value, the panel defect evaluation value, and the power conversion evaluation value, and a second display panel test value is obtained by weighted sum calculation.

10. The aging test system for a display panel according to claim 6, characterized in that: The judgment unit compares the aging test value with the aging test threshold value, and judges it as unqualified if it exceeds the aging test threshold value; otherwise, the aging test data is input into the test environment prediction model, and the specific process of generating the optimal test environment information according to the data result of the test environment prediction model includes: Obtain historical test environment data, historical aging test data, and current aging test data; Inputting the historical test environment data and the historical aging test data into a test environment prediction model for training to obtain environment prediction parameters; Input the environmental prediction parameters and the current aging test data into the test environment prediction model, update the model parameters, and output the real-time environmental model weights; wherein the real-time environmental model weights include: real-time illumination environment weights, real-time temperature environment weights, and real-time humidity environment weights; The real-time environment information and the real-time environment model weight are subjected to a Hadamard product operation to obtain optimal test environment information.

Images