Multi-dimensional temperature measurement method for organic electroluminescent display panel

Abstract

The present invention provides a multi-dimensional temperature measurement method for an organic electroluminescent display panel, which belongs to the technical field of organic electroluminescent panel temperature measurement, and includes: step 1: obtaining the type and properties of the display panel, and determining the corresponding thermal expansion coefficient sensor; step 2: measuring the temperature change data of the display panel according to the thermal expansion coefficient sensor, and extracting the characteristic parameters of the display panel; step 3: determining the temperature distribution of the panel surface at different positions and directions according to the infrared image reflected by the display panel surface captured by the infrared camera; step 4: obtaining the charge of each pixel of the display panel, and obtaining the state change of the pixel according to the heat conduction technology; step 5: constructing a multi-dimensional temperature prediction model, and measuring the multi-dimensional temperature of the display panel. The method solves the problem that the actual temperature distribution of each area of ​​the display panel cannot be accurately reflected, which reduces the performance of the light-emitting display panel, and at the same time, the measurement method is relatively single, resulting in inaccurate measurement results.

Description

Technical Field

[0001] The present invention relates to the technical field of organic electroluminescent panel temperature measurement, and in particular to a multi-dimensional temperature measurement method for an organic electroluminescent display panel. Background Art

[0002] With the development of science and technology, organic electroluminescent technology has been widely used. Organic electroluminescent display panels have the advantages of self-luminescence, high contrast, and low power consumption, so they have become the preferred material for the new generation of displays.

[0003] The traditional multi-dimensional temperature measurement method of organic electroluminescent display panels cannot accurately reflect the actual temperature distribution of each area of ​​the display panel, resulting in local overheating or overcooling, which greatly reduces the performance of the organic electroluminescent display panel. At the same time, the traditional measurement method is relatively single and cannot combine information from multiple dimensions to measure the temperature of the light-emitting panel, resulting in deviation and inaccuracy in the measurement results.

[0004] Therefore, the present invention proposes a multi-dimensional temperature measurement method for an organic electroluminescent display panel. Summary of the invention

[0005] The present invention provides a multi-dimensional temperature measurement method for an organic electroluminescent display panel, which is used to solve the problem that the prior art cannot accurately reflect the actual temperature distribution of each area of ​​the display panel, resulting in local overheating or overcooling, which greatly reduces the performance of the organic electroluminescent display panel. At the same time, the traditional measurement method is relatively single and cannot combine information from multiple dimensions to measure the temperature of the light-emitting panel, resulting in the defects of deviation and inaccuracy of the measurement results.

[0006] In one aspect, the present invention provides a multi-dimensional temperature measurement method for an organic electroluminescent display panel, comprising:

[0007] Step 1: obtaining the type and properties of the target organic electroluminescent display panel, and determining a thermal expansion coefficient sensor of corresponding type and specification according to the type and properties;

[0008] Step 2: measuring the temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor, and extracting characteristic parameters of the organic electroluminescent display panel based on the temperature change data and a machine learning algorithm model;

[0009] Step 3: capturing an infrared image reflected by the surface of the organic electroluminescent display panel using an infrared camera, and determining the temperature distribution of the panel surface at different positions and directions according to the infrared image;

[0010] Step 4: acquiring the charge of each pixel of the organic electroluminescent display panel, acquiring the state change of the pixel based on the charge of each pixel according to the thermal conduction technology, and acquiring the change image of the pixel state;

[0011] Step 5: Construct a multi-dimensional temperature prediction model based on the deep learning model according to the characteristic parameters, the temperature distribution on the panel surface, and the pixel state change image, and measure the multi-dimensional temperature of the organic electroluminescent display panel according to the multi-dimensional temperature prediction model.

[0012] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, the type and properties of a target organic electroluminescent display panel are obtained, and a thermal expansion coefficient sensor of a corresponding type and specification is determined according to the type and properties, including:

[0013] Obtaining material composition of a target organic electroluminescent display panel, and determining interface characteristics between materials according to the material composition;

[0014] Acquire the RGB arrangement and resolution of the target organic electroluminescent display panel, and determine the pixel structure of the panel according to the RGB arrangement and resolution;

[0015] Determining the type and properties of the target organic electroluminescent display panel according to the interface characteristics and pixel structure;

[0016] According to the type and property, a thermal expansion coefficient sensor of corresponding type and specification is determined based on multiple influencing factors, and the thermal expansion coefficient sensor is calibrated.

[0017] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, the temperature change data of the organic electroluminescent display panel is measured according to the thermal expansion coefficient sensor, and characteristic parameters of the organic electroluminescent display panel are extracted based on the temperature change data and a machine learning algorithm model, including:

[0018] Measuring temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor based on the working principle and temperature sensitivity characteristics of the organic electroluminescent display panel;

[0019] Performing data preprocessing on the temperature change data to obtain a preprocessing result;

[0020] Obtaining the scale and complexity of the data according to the preprocessing results, and determining the distribution characteristics of the data according to the scale and complexity;

[0021] Determine a corresponding machine learning algorithm model according to the distribution characteristics of the data;

[0022] The preprocessed data is input into the machine learning algorithm model and converted into a high-dimensional feature vector through the model's mapping;

[0023] The characteristic parameters of the organic electroluminescent display panel are extracted according to the high-dimensional characteristic vector.

[0024] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, the scale and complexity of data are obtained according to the preprocessing result, and the distribution characteristics of the data are determined according to the scale and complexity, including:

[0025] Determine the data volume statistical parameters of the temperature change data according to the preprocessing results, and determine the data scale according to the data volume statistical parameters;

[0026] Acquire multiple preset complexity measurement indicators, and perform feature evaluation on the temperature change data using the multiple preset complexity measurement indicators to acquire multiple candidate complex feature matrices;

[0027] Obtaining the relationship between row elements, the relationship between column elements and the directionality of their relationship of each candidate complex feature matrix;

[0028] Determine the relationship strength weight between the row and column elements of each candidate complex feature matrix based on the relationship between the row elements, the relationship between the column elements and the directionality of their relationship;

[0029] Determine the standard connection information flow between temperature data and time changes in the temperature change data under each candidate complex feature matrix according to the relationship strength weight;

[0030] Acquire the current contact information flow of the temperature data and the time change in the temperature change data, match the current contact information flow with the standard contact information flow to determine a first candidate complex feature matrix with a matching degree greater than or equal to a preset threshold;

[0031] Determine data sampling prior information according to the distribution characteristics of the depicted data of the first candidate complex feature matrix, and extract sample data from the temperature change data according to the data sampling prior information and a probability statistics method;

[0032] Obtaining global density data parameters and local density data parameters of the sample data, and determining data distribution estimation parameters corresponding to each first candidate complex feature matrix according to the global density data parameters and the local density data parameters;

[0033] Selecting a second candidate complex feature matrix from the first candidate complex feature matrix according to the data distribution estimation parameter and the data distribution of the temperature change data;

[0034] The data complexity is determined according to the feature quantity of the second candidate complex feature matrix, and the distribution characteristics of the data are determined according to the data scale and the data complexity.

[0035] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, an infrared camera captures an infrared image reflected by the surface of the organic electroluminescent display panel, and determines the temperature distribution of the panel surface at different positions and directions according to the infrared image, including:

[0036] Acquire the working environment and use conditions of the organic electroluminescent display panel, and set the parameters of the infrared camera according to the working environment and use conditions;

[0037] The infrared camera is configured to capture an infrared image reflected from the surface of the organic electroluminescent display panel, and perform image preprocessing on the image;

[0038] Generate a temperature distribution thermogram of the organic electroluminescent display panel at different positions and directions based on the preprocessed image based on the thermogram technology;

[0039] Convert the temperature distribution thermodynamic map into a color gradient image to obtain temperature differences in different positions and directions;

[0040] The temperature distribution of the surface of the organic electroluminescent display panel at different positions and directions is determined according to the temperature difference.

[0041] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, the charge of each pixel of the organic electroluminescent display panel is obtained, the state change of the pixel is obtained based on the charge of each pixel according to the heat conduction technology, and an image of the change of the pixel state is obtained, including:

[0042] Obtaining the electrophoretic response of each pixel area of ​​the organic electroluminescent display panel based on different voltages according to the electrophoretic imaging system;

[0043] determining the charge of each pixel region according to the electrophoretic response of each pixel region;

[0044] Determine the charge transfer and distribution in the pixel based on the material transport equation of the first principles according to the charge of each pixel area;

[0045] Determine a temperature change of each pixel based on a numerical method according to the charge transfer and distribution, and determine a color value change of each pixel according to the temperature change;

[0046] The state change of the pixel is obtained according to the change of the color value of each pixel, the state changes of all pixels are integrated together, the temperature change and final state of each pixel during the heat conduction process are obtained, and a pixel state change image is formed according to the temperature change and the final state.

[0047] According to a multi-dimensional temperature measurement method for an organic electroluminescent display panel provided by the present invention, after determining the charge of each pixel area according to the electrophoretic response of each pixel area, the method further includes:

[0048] Acquire the grayscale value and color information of each pixel area, and determine the data of each pixel area in the electrophoresis image according to the grayscale value and color information;

[0049] Acquire characteristic points of similar electrophoretic response based on local binary pattern entropy according to the data of each pixel area;

[0050] determining charge density data for each pixel region based on the area of ​​each pixel region according to the conductivity measurement results;

[0051] Normalizing the charge density data of each pixel region based on a scale factor;

[0052] The normalized charge density data and the characteristic points of similar electrophoretic responses are input into the deep learning model as feature vectors for training, and different types of electrophoretic patterns and electrophoretic anomalies are identified based on the trained model.

[0053] According to a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by the present invention, a multi-dimensional temperature prediction model is constructed based on a deep learning model according to the characteristic parameters, the temperature distribution on the panel surface, and the change image of the pixel state, and the multi-dimensional temperature of the organic electroluminescent display panel is measured according to the multi-dimensional temperature prediction model, including:

[0054] Acquire time series information of the organic electroluminescent display panel according to the characteristic parameters, the temperature distribution on the panel surface and the pixel state change image;

[0055] Acquire a recurrent neural network model according to the time series information, and train and optimize the recurrent neural network model according to cross-validation;

[0056] A multi-dimensional temperature prediction model is constructed based on the trained and optimized recurrent neural network model based on the panel surface temperature distribution data and corresponding features, and the multi-dimensional temperature of the organic electroluminescent display panel is measured based on the multi-dimensional temperature prediction model.

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

[0058] The temperature change data of the organic electroluminescent display panel is obtained through the thermal expansion coefficient sensor, and the characteristic parameters are extracted. The temperature distribution of the panel surface at different positions and directions determined by the infrared image and the change image of the state of each pixel on the panel are combined to build a multi-dimensional temperature prediction model. It can accurately reflect the actual temperature distribution of each area of ​​the display panel, avoid local overheating or overcooling, and greatly improve the performance of the organic electroluminescent display panel. At the same time, the traditional measurement method is diversified, and the temperature of the light-emitting panel can be measured by combining information from multiple dimensions to enhance the accuracy of the measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a schematic flow chart of a multi-dimensional temperature measurement method of an organic electroluminescent display panel provided by an embodiment of the present invention;

[0061] Figure 2 It is a schematic diagram of a process of determining a thermal expansion coefficient sensor according to the type and properties of an organic electroluminescent display panel provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0063] Embodiment 1:

[0064] An embodiment of the present invention provides a multi-dimensional temperature measurement method for an organic electroluminescent display panel, such as Figure 1 As shown, the method mainly includes the following steps:

[0065] Step 1: obtaining the type and properties of the target organic electroluminescent display panel, and determining a thermal expansion coefficient sensor of corresponding type and specification according to the type and properties;

[0066] Step 2: measuring the temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor, and extracting characteristic parameters of the organic electroluminescent display panel based on the temperature change data and a machine learning algorithm model;

[0067] Step 3: capturing an infrared image reflected by the surface of the organic electroluminescent display panel using an infrared camera, and determining the temperature distribution of the panel surface at different positions and directions according to the infrared image;

[0068] Step 4: acquiring the charge of each pixel of the organic electroluminescent display panel, acquiring the state change of the pixel based on the charge of each pixel according to the thermal conduction technology, and acquiring the change image of the pixel state;

[0069] Step 5: Construct a multi-dimensional temperature prediction model based on the deep learning model according to the characteristic parameters, the temperature distribution on the panel surface, and the pixel state change image, and measure the multi-dimensional temperature of the organic electroluminescent display panel according to the multi-dimensional temperature prediction model.

[0070] In this embodiment, the types of organic electroluminescence include: a traditional structured organic electroluminescent display panel, a flexible organic electroluminescent display panel, and a foldable organic electroluminescent display panel.

[0071] In this embodiment, the properties of the organic electroluminescent display panel include: high contrast, fast response, and wide color gamut.

[0072] In this embodiment, the thermal expansion coefficient sensor is a device used to measure the thermal expansion coefficient of a material.

[0073] In this embodiment, the characteristic parameters of the organic electroluminescent display panel refer to physical quantities or indicators related to image quality, display performance, energy efficiency, etc. generated during the operation of the organic electroluminescent display panel, such as: voltage and current, temperature, color coordinates, color gamut coverage, and response time.

[0074] In this embodiment, the infrared image reflected by the surface of the organic electroluminescent display panel refers to an infrared radiation image of the surface of the display panel captured by an infrared camera.

[0075] The beneficial effects of the above technical solution are: obtaining the temperature change data of the organic electroluminescent display panel through the thermal expansion coefficient sensor, extracting characteristic parameters, combining the temperature distribution of the panel surface at different positions and directions determined by the infrared image and the change image of the state of each pixel on the panel to build a multi-dimensional temperature prediction model, which can accurately reflect the actual temperature distribution of each area of ​​the display panel, avoid local overheating or overcooling, and greatly improve the performance of the organic electroluminescent display panel. At the same time, the traditional measurement method is diversified, and the temperature of the light-emitting panel can be measured in combination with information in multiple dimensions to enhance the accuracy of the measurement results.

[0076] Embodiment 2:

[0077] Based on Example 1, the embodiment of the present invention obtains the type and properties of the target organic electroluminescent display panel, and determines the corresponding type and specification of the thermal expansion coefficient sensor according to the type and properties, such as Figure 2 As shown, including:

[0078] S01: Obtaining material composition of a target organic electroluminescent display panel, and determining interface characteristics between materials according to the material composition;

[0079] S02: Acquire the RGB arrangement and resolution of the target organic electroluminescent display panel, and determine the pixel structure of the panel according to the RGB arrangement and resolution;

[0080] S03: determining the type and properties of the target organic electroluminescent display panel according to the interface characteristics and pixel structure;

[0081] S04: Determine a thermal expansion coefficient sensor of corresponding type and specification according to the type and property based on multiple influencing factors, and calibrate the thermal expansion coefficient sensor.

[0082] In this embodiment, the material components of the organic electroluminescent display panel include: organic material, inorganic material, flexible plastic, and glass fiber.

[0083] In this embodiment, the interface characteristics between materials refer to the properties and behaviors exhibited when the materials interact with each other.

[0084] In this embodiment, the RGB arrangement refers to the color combination of each pixel on the OLED display. This arrangement usually refers to arranging the three primary color elements of red, green and blue in a specific order and interval, such as: interlaced scanning, full refresh, and direct arrangement.

[0085] In this embodiment, the resolution of the organic electroluminescent display panel refers to the minimum detail and clarity of an image that can be presented on the screen.

[0086] In this embodiment, the pixel structure of the organic electroluminescent display panel refers to the basic unit constituting the display screen, which is composed of a grid array composed of thin film transistors.

[0087] In this embodiment, the multiple influencing factors include: response speed, accuracy, linearity, and installation method.

[0088] In this embodiment, calibration refers to: determining the zero point and full scale of the sensor, and determining its response characteristics under specific environmental conditions.

[0089] The beneficial effect of the above technical solution is: the type and properties of the target organic electroluminescent display panel are determined according to the interface characteristics and pixel structure, thereby determining the corresponding type and specification of the thermal expansion coefficient sensor, which can provide a more accurate and reliable thermal expansion coefficient sensor, thereby improving the accuracy of the acquired temperature data.

[0090] Embodiment 3:

[0091] Based on Example 2, the embodiment of the present invention measures the temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor, and extracts the characteristic parameters of the organic electroluminescent display panel based on the machine learning algorithm model according to the temperature change data, including:

[0092] Measuring temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor based on the working principle and temperature sensitivity characteristics of the organic electroluminescent display panel;

[0093] Performing data preprocessing on the temperature change data to obtain a preprocessing result;

[0094] Obtaining the scale and complexity of the data according to the preprocessing results, and determining the distribution characteristics of the data according to the scale and complexity;

[0095] Determine a corresponding machine learning algorithm model according to the distribution characteristics of the data;

[0096] The preprocessed data is input into the machine learning algorithm model and converted into a high-dimensional feature vector through the model's mapping;

[0097] The characteristic parameters of the organic electroluminescent display panel are extracted according to the high-dimensional characteristic vector.

[0098] In this embodiment, the thermal expansion coefficient sensor is a device used to measure the thermal expansion coefficient of a material.

[0099] In this embodiment, the working principle of the organic electroluminescent display panel is based on the characteristics of organic semiconductors. When current passes through the organic semiconductor, the electrons in the semiconductor gain energy and jump to the conduction band, and at the same time, holes are released to form carriers. When the positive and negative ions of the organic semiconductor meet, a chemical reaction occurs to generate light.

[0100] In this embodiment, the temperature sensitivity of the organic electroluminescent display panel refers to the degree to which its performance changes with temperature.

[0101] In this embodiment, data preprocessing may include: normalization, cleaning, and conversion.

[0102] In this embodiment, the distribution characteristics of the temperature change data of the organic electroluminescent display panel include:

[0103] Non-uniform distribution: Due to the differences in materials and structures inside the organic electroluminescent display panel, as well as the influence of the working environment, its temperature distribution is non-uniform. Even at the same moment, the temperatures of adjacent areas may vary greatly.

[0104] Periodicity: Due to the working principle of the display panel and environmental factors such as backlight temperature, current density, etc., it may show a certain periodicity, which may cause the temperature to change within a week, half a month or a certain period of time in a day.

[0105] Local hot spots: Due to uneven heat distribution in local components inside the display panel, the temperature in some areas may suddenly rise, forming local hot spots.

[0106] Randomness: Even at the same point in time, the temperature of the display panel may be affected by many uncertain factors, such as ambient temperature changes and operational errors.

[0107] In this embodiment, the characteristic parameters of the organic electroluminescent display panel refer to physical quantities or indicators related to image quality, display performance, energy efficiency, etc. generated during the operation of the organic electroluminescent display panel, such as: voltage and current, temperature, color coordinates, color gamut coverage, and response time.

[0108] The beneficial effect of the above technical solution is: by determining the distribution characteristics of the data through the scale and complexity of the temperature data of the organic electroluminescent display panel, and extracting the characteristic parameters of the organic electroluminescent display panel in combination with the machine learning algorithm model, the usage and load conditions of the organic electroluminescent display panel can be accurately obtained, thereby improving the reliability of the panel.

[0109] Embodiment 4:

[0110] Based on Example 3, the embodiment of the present invention obtains the scale and complexity of the data according to the preprocessing results, and determines the distribution characteristics of the data according to the scale and complexity, including:

[0111] Determine the data volume statistical parameters of the temperature change data according to the preprocessing results, and determine the data scale according to the data volume statistical parameters;

[0112] Acquire multiple preset complexity measurement indicators, and perform feature evaluation on the temperature change data using the multiple preset complexity measurement indicators to acquire multiple candidate complex feature matrices;

[0113] Obtaining the relationship between row elements, the relationship between column elements and the directionality of their relationship of each candidate complex feature matrix;

[0114] Determine the relationship strength weight between the row and column elements of each candidate complex feature matrix based on the relationship between the row elements, the relationship between the column elements and the directionality of their relationship;

[0115] Determine the standard connection information flow between temperature data and time changes in the temperature change data under each candidate complex feature matrix according to the relationship strength weight;

[0116] Acquire the current contact information flow of the temperature data and the time change in the temperature change data, match the current contact information flow with the standard contact information flow to determine a first candidate complex feature matrix with a matching degree greater than or equal to a preset threshold;

[0117] Determine data sampling prior information according to the distribution characteristics of the depicted data of the first candidate complex feature matrix, and extract sample data from the temperature change data according to the data sampling prior information and a probability statistics method;

[0118] Obtaining global density data parameters and local density data parameters of the sample data, and determining data distribution estimation parameters corresponding to each first candidate complex feature matrix according to the global density data parameters and the local density data parameters;

[0119] Selecting a second candidate complex feature matrix from the first candidate complex feature matrix according to the data distribution estimation parameter and the data distribution of the temperature change data;

[0120] The data complexity is determined according to the feature quantity of the second candidate complex feature matrix, and the distribution characteristics of the data are determined according to the data scale and the data complexity.

[0121] In this embodiment, the data volume statistical parameters of the temperature change data include:

[0122] Mean: reflects the average temperature level of the data set.

[0123] Standard deviation: It measures the degree of difference between each value in a data set and its mean value. It is an indicator of the degree of dispersion of the data.

[0124] Skewness coefficient: describes the shape of the data distribution and is used to distinguish between normal distribution, bell-shaped distribution, bimodal distribution, etc.

[0125] Peak coefficient: describes the tail shape of the data distribution and reflects the degree of skewness of the data distribution.

[0126] In this embodiment, data scale refers to the amount of data.

[0127] In this embodiment, the preset complexity measurement indicators include:

[0128] Statistical parameters such as the mean and standard deviation at each time point are used to evaluate the dispersion and stability of the data.

[0129] Check the distribution of pixel values ​​in the local area, such as whether there are hot spots, cold spots, etc., to evaluate the inhomogeneity of the data.

[0130] Analyze the peaks, valleys and other extreme points in the data to assess the amount of mutation in the data.

[0131] The time series characteristics of the data, such as whether there are periodic fluctuations, are evaluated through methods such as autocorrelation function and power spectral density estimation.

[0132] Analyze the correlation coefficient of data to evaluate the internal correlation of data

[0133] In this embodiment, the characteristics of the temperature change data include: time characteristics, space characteristics, temperature value characteristics, and temperature change rate characteristics.

[0134] In this embodiment, the complex feature matrix is ​​a data structure used to represent multi-dimensional nonlinear relationships. The matrix is ​​usually composed of multiple interrelated features, and each feature contains information of multiple dimensions.

[0135] In this embodiment, the relationship strength weight is used to determine the relative importance between row and column elements.

[0136] In this embodiment, the standard connection information flow of temperature data and time changes is a method of measuring the amount of information about the impact of temperature changes on other variables in time series data.

[0137] In this embodiment, data sampling prior information refers to the prior knowledge about data distribution obtained by discrete sampling of the original data set during data processing, which can be used to estimate and predict unknown components in the data, such as implicit states and noise levels.

[0138] In this embodiment, the global density data parameter is a method for describing the dense nature of sample data, and mainly focuses on the distribution of data points and the distances between adjacent points.

[0139] In this embodiment, the local density data parameter is mainly used to analyze the information of the local area in the data, so as to understand the overall structure and content of the data, as well as the details of different data.

[0140] In this embodiment, the data distribution estimation parameter is a statistic used to describe the distribution of data points in a data set.

[0141] The beneficial effects of the above technical solution are: determining the data complexity according to the feature quantity of the second candidate complex feature matrix, and determining the distribution characteristics of the temperature change data of the organic electroluminescent display panel according to the data scale and data complexity, so that data processing and analysis can be performed more efficiently, thereby obtaining valuable information more quickly. At the same time, it can more accurately predict possible problems, and respond and adjust more quickly when problems occur, reducing the risk of damage or performance degradation of the organic electroluminescent display panel due to abnormal temperature.

[0142] Embodiment 5:

[0143] Based on Example 4, the embodiment of the present invention captures an infrared image reflected by the surface of the organic electroluminescent display panel using an infrared camera, and determines the temperature distribution of the panel surface at different positions and directions according to the infrared image, including:

[0144] Acquire the working environment and use conditions of the organic electroluminescent display panel, and set the parameters of the infrared camera according to the working environment and use conditions;

[0145] The infrared camera is configured to capture an infrared image reflected from the surface of the organic electroluminescent display panel, and perform image preprocessing on the image;

[0146] Generate a temperature distribution thermogram of the organic electroluminescent display panel at different positions and directions based on the preprocessed image based on the thermogram technology;

[0147] Convert the temperature distribution thermodynamic map into a color gradient image to obtain temperature differences in different positions and directions;

[0148] The temperature distribution of the surface of the organic electroluminescent display panel at different positions and directions is determined according to the temperature difference.

[0149] In this embodiment, the working environment of the organic electroluminescent display panel includes: temperature, humidity, dustproof and shockproof.

[0150] In this embodiment, the use conditions of the organic electroluminescent display panel may be: stable voltage and current, and should be placed in a place without strong magnetic fields, static fields or other electromagnetic interference sources.

[0151] In this embodiment, the parameters of the infrared camera include: resolution, dynamic range and color depth.

[0152] In this embodiment, the infrared image reflected by the surface of the organic electroluminescent display panel refers to an infrared radiation image of the surface of the display panel captured by an infrared camera.

[0153] In this embodiment, image preprocessing includes: checking and repairing erroneous, blurred or damaged areas in the image, such as missing pixels and wrong color values ​​in the image; image enhancement, such as improving the contrast, brightness and color balance of the image.

[0154] In this embodiment, the heat map technology refers to treating the temperature of each data unit as a point in a two-dimensional coordinate system, and then sorting these points according to the size of their temperature values ​​to form a two-dimensional heat map matrix, such as: local scaling method.

[0155] In this embodiment, the temperature distribution heat map is a chart that uses color coding to visualize the temperature values ​​at different locations on the panel. By mapping the temperature value of each location to a different color on the chart, the temperature differences between the locations can be displayed simultaneously on one chart.

[0156] The beneficial effects of the above technical solution are: by capturing the infrared image reflected by the surface of the organic electroluminescent display panel with an infrared camera and performing image preprocessing, the image quality can be improved. Further, a temperature distribution thermogram of the organic electroluminescent display panel at different positions and directions is generated, and the temperature difference at different positions and directions is obtained, which can better control the temperature of different areas on the panel, avoid overheating or too low temperature, and improve the performance and service life of the organic electroluminescent display panel.

[0157] Embodiment 6:

[0158] Based on Example 5, the embodiment of the present invention obtains the charge of each pixel of the organic electroluminescent display panel, obtains the state change of the pixel based on the charge of each pixel according to the thermal conduction technology, and obtains the change image of the pixel state, including:

[0159] Obtaining the electrophoretic response of each pixel area of ​​the organic electroluminescent display panel based on different voltages according to the electrophoretic imaging system;

[0160] determining the charge of each pixel region according to the electrophoretic response of each pixel region;

[0161] Determine the charge transfer and distribution in the pixel based on the material transport equation of the first principles according to the charge of each pixel area;

[0162] Determine a temperature change of each pixel based on a numerical method according to the charge transfer and distribution, and determine a color value change of each pixel according to the temperature change;

[0163] The state change of the pixel is obtained according to the change of the color value of each pixel, the state changes of all pixels are integrated together, the temperature change and final state of each pixel during the heat conduction process are obtained, and a pixel state change image is formed according to the temperature change and the final state.

[0164] In this embodiment, the electrophoresis imaging system is a microscopy technology used to detect and image charged particles or molecules. It uses the interaction between electric fields and charges to achieve micro-analysis of samples through high-resolution image processing and analysis.

[0165] In this embodiment, in the organic electroluminescent display panel, each pixel region has a specific charge state, called electrophoretic response, which is caused by the voltage generated by the current passing through the display and determines the color emitted by each pixel region.

[0166] In this embodiment, each pixel area is composed of many tiny transistors that are responsible for controlling the flow of current. When current flows through these transistors, it releases the same number of electrons as the surrounding environment, so that each pixel area has a specific charge state.

[0167] In this embodiment, the first-principles material transport equation refers to the basic law describing the transport of matter in solid materials. In solid physics, there are two main modes of material transport: diffusion and migration.

[0168] In this embodiment, in the organic electroluminescent display panel, the temperature change of each pixel generally refers to the change of the energy state of electrons in the pixel.

[0169] In this embodiment, in the organic electroluminescent display panel, the change in the color value of each pixel refers to the change in the red, green and blue color values ​​corresponding to the brightness level of a specific color displayed on the screen.

[0170] In this embodiment, the pixel state change image of the organic electroluminescent display panel refers to the change of the current of each pixel over time during the organic electroluminescent display process.

[0171] The beneficial effects of the above technical solution are: determining the charge of each pixel area based on the electrophoretic response of each pixel area, thereby determining the charge transfer and distribution in the pixel and the change in the color value of each pixel, obtaining the temperature change and final state of each pixel during the heat conduction process, and forming a pixel state change image, which can quickly discover and locate faults and thermal management problems, thereby avoiding long-term display errors or damage and improving the reliability of the panel.

[0172] Embodiment 7:

[0173] Based on Embodiment 6, after determining the charge of each pixel region according to the electrophoretic response of each pixel region, the embodiment of the present invention further includes:

[0174] Acquire the grayscale value and color information of each pixel area, and determine the data of each pixel area in the electrophoresis image according to the grayscale value and color information;

[0175] Acquire characteristic points of similar electrophoretic response based on local binary pattern entropy according to the data of each pixel area;

[0176] determining charge density data for each pixel region based on the area of ​​each pixel region according to the conductivity measurement results;

[0177] Normalizing the charge density data of each pixel region based on a scale factor;

[0178] The normalized charge density data and the characteristic points of similar electrophoretic responses are input into the deep learning model as feature vectors for training, and different types of electrophoretic patterns and electrophoretic anomalies are identified based on the trained model.

[0179] In this embodiment, the grayscale value of each pixel area represents the color depth of each pixel in the digital image, and is an integer between 0 and 255, where 0 represents black and 255 represents white.

[0180] In this embodiment, the color information of each pixel area is composed of three channels: red, green and blue channels. These three channels correspond to the three primary colors of red, green and blue on the screen, and their intensities determine the final display color. Therefore, the color information of each pixel area can be represented by a three-dimensional array, in which the first element represents the intensity of the red channel, the second element represents the intensity of the green channel, and the third element represents the intensity of the blue channel. The value range of this array is usually between 0 and 1, representing the color of the minimum and maximum brightness respectively.

[0181] In this embodiment, in the electrophoresis diagram, the data of each pixel area represents the bandwidth of the molecules separated in the area. The electrophoresis technology will perform electrophoretic separation on the mixture so that molecules with different molecular weights and sizes move at different speeds.

[0182] In this embodiment, the local binary pattern entropy is a method for analyzing the statistical characteristics of local regions in an image, and is used to detect and segment regions with similar features.

[0183] In this embodiment, similar electrophoretic response means that in an electrophoretic experiment, two or more organic electroluminescent display panel samples exhibit similar electrophoretic behaviors under the same conditions.

[0184] In this embodiment, the charge density refers to the proportion of charged ions per unit area of ​​the light-emitting panel.

[0185] The beneficial effect of the above technical solution is: by obtaining the characteristic points of similar electrophoretic responses through the data of each pixel area in the electrophoresis image, and combining the normalized charge density data of each pixel area to train the deep learning model, different types of electrophoretic patterns and electrophoretic anomalies can be identified, which can greatly reduce the probability of failure of the organic electroluminescent display panel, ensure the normal operation of the panel, and improve the reliability and stability of the organic electroluminescent display panel.

[0186] Embodiment 8:

[0187] Based on Example 7, the embodiment of the present invention builds a multi-dimensional temperature prediction model based on the deep learning model according to the characteristic parameters, the temperature distribution on the panel surface, and the change image of the pixel state, and measures the multi-dimensional temperature of the organic electroluminescent display panel according to the multi-dimensional temperature prediction model, including:

[0188] Acquire time series information of the organic electroluminescent display panel according to the characteristic parameters, the temperature distribution on the panel surface and the pixel state change image;

[0189] Acquire a recurrent neural network model according to the time series information, and train and optimize the recurrent neural network model according to cross-validation;

[0190] A multi-dimensional temperature prediction model is constructed based on the trained and optimized recurrent neural network model based on the panel surface temperature distribution data and corresponding features, and the multi-dimensional temperature of the organic electroluminescent display panel is measured based on the multi-dimensional temperature prediction model.

[0191] In this embodiment, the time series information of the organic electroluminescent display panel refers to information of images displayed by the display panel at different time points, which can be used to obtain the grayscale and color space of the display panel at each time point, thereby generating time series image data.

[0192] In this embodiment, the recurrent neural network is a special neural network structure that can capture the timing information in time series data. The core idea of ​​the recurrent neural network is to pass part of the hidden state to the next layer based on the previous layer, so that the network can "remember" the previous input information.

[0193] In this embodiment, cross validation is a method for evaluating model performance. The basic idea is to divide the total sample size into K subsets, assign each subset to a model for training, and keep the remaining subsets for testing. After the model is trained, the subsets kept for testing are used to evaluate the generalization ability of the model.

[0194] In this embodiment, training and optimization may involve adjusting different hyperparameters, such as learning rate and batch size.

[0195] In this embodiment, the panel surface temperature distribution data refers to data recording the surface temperatures of different regions of the organic electroluminescent display panel, including: the temperature change trend of a specific region of the organic electroluminescent display panel over time, the temperature values ​​at different positions of the organic electroluminescent display panel, the temperature fluctuation range, the temperature peak and valley values, etc.

[0196] In this embodiment, the multi-dimensional temperature prediction model is a model for predicting the multi-dimensional temperature distribution of a specific area within a period of time in the future.

[0197] The beneficial effects of the above technical solution are: by obtaining a recurrent neural network model through the time series information of the organic electroluminescent display panel, and optimizing and training it, modeling can be performed in the time dimension to capture the long-term trend of temperature and changes at different time points. Furthermore, a multi-dimensional temperature prediction model is constructed based on the data of the panel surface temperature distribution and the corresponding characteristics, which can reduce the influence of a single factor and thus improve the accuracy of the prediction.

[0198] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0199] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-dimensional temperature measurement method for an organic electroluminescent display panel, characterized in that: include: Step 1: obtaining the type and properties of the target organic electroluminescent display panel, and determining a thermal expansion coefficient sensor of corresponding type and specification according to the type and properties; Step 2: measuring the temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor, and extracting characteristic parameters of the organic electroluminescent display panel based on the temperature change data and a machine learning algorithm model; Step 3: capturing an infrared image reflected by the surface of the organic electroluminescent display panel using an infrared camera, and determining the temperature distribution of the panel surface at different positions and directions according to the infrared image; Step 4: acquiring the charge of each pixel of the organic electroluminescent display panel, acquiring the state change of the pixel based on the charge of each pixel according to the thermal conduction technology, and acquiring the change image of the pixel state; Step 5: constructing a multi-dimensional temperature prediction model based on the deep learning model according to the characteristic parameters, the temperature distribution on the panel surface, and the pixel state change image, and measuring the multi-dimensional temperature of the organic electroluminescent display panel according to the multi-dimensional temperature prediction model; The method includes obtaining the charge of each pixel of the organic electroluminescent display panel, obtaining the state change of the pixel based on the charge of each pixel according to the thermal conduction technology, and obtaining the change image of the pixel state, including: Obtaining the electrophoretic response of each pixel area of ​​the organic electroluminescent display panel based on different voltages according to the electrophoretic imaging system; determining the charge of each pixel region according to the electrophoretic response of each pixel region; Determine the charge transfer and distribution in the pixel based on the material transport equation of the first principles according to the charge of each pixel area; Determine a temperature change of each pixel based on a numerical method according to the charge transfer and distribution, and determine a color value change of each pixel according to the temperature change; Obtain the state change of the pixel according to the change of the color value of each pixel, integrate the state changes of all pixels together, obtain the temperature change and final state of each pixel during the heat conduction process, and form a pixel state change image according to the temperature change and the final state; After determining the charge of each pixel region according to the electrophoretic response of each pixel region, the method further includes: Acquire the grayscale value and color information of each pixel area, and determine the data of each pixel area in the electrophoresis image according to the grayscale value and color information; Acquire characteristic points of similar electrophoretic response based on local binary pattern entropy according to the data of each pixel area; determining charge density data for each pixel region based on the area of ​​each pixel region according to the conductivity measurement results; Normalizing the charge density data of each pixel region based on a scale factor; The normalized charge density data and the characteristic points of similar electrophoretic responses are input into the deep learning model as feature vectors for training, and different types of electrophoretic patterns and electrophoretic anomalies are identified based on the trained model.

2. The multi-dimensional temperature measurement method of an organic electroluminescent display panel according to claim 1, characterized in that: Acquiring the type and properties of the target organic electroluminescent display panel, and determining the corresponding type and specification of the thermal expansion coefficient sensor according to the type and properties, including: Obtaining material composition of a target organic electroluminescent display panel, and determining interface characteristics between materials according to the material composition; Acquire the RGB arrangement and resolution of the target organic electroluminescent display panel, and determine the pixel structure of the panel according to the RGB arrangement and resolution; Determining the type and properties of the target organic electroluminescent display panel according to the interface characteristics and pixel structure; According to the type and property, a thermal expansion coefficient sensor of corresponding type and specification is determined based on multiple influencing factors, and the thermal expansion coefficient sensor is calibrated.

3. The multi-dimensional temperature measurement method of an organic electroluminescent display panel according to claim 1, characterized in that: Measuring temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor, and extracting characteristic parameters of the organic electroluminescent display panel based on the temperature change data and a machine learning algorithm model, including: Measuring temperature change data of the organic electroluminescent display panel according to the thermal expansion coefficient sensor based on the working principle and temperature sensitivity characteristics of the organic electroluminescent display panel; Performing data preprocessing on the temperature change data to obtain a preprocessing result; Obtaining the scale and complexity of the data according to the preprocessing results, and determining the distribution characteristics of the data according to the scale and complexity; Determine a corresponding machine learning algorithm model according to the distribution characteristics of the data; The preprocessed data is input into the machine learning algorithm model and converted into a high-dimensional feature vector through the model's mapping; The characteristic parameters of the organic electroluminescent display panel are extracted according to the high-dimensional characteristic vector.

4. The multi-dimensional temperature measurement method of the organic electroluminescent display panel according to claim 3, characterized in that: The size and complexity of the data are obtained according to the preprocessing results, and the distribution characteristics of the data are determined according to the size and complexity, including: Determine the data volume statistical parameters of the temperature change data according to the preprocessing results, and determine the data scale according to the data volume statistical parameters; Acquire multiple preset complexity measurement indicators, and perform feature evaluation on the temperature change data using the multiple preset complexity measurement indicators to acquire multiple candidate complex feature matrices; Obtaining the relationship between row elements, the relationship between column elements and the directionality of their relationship of each candidate complex feature matrix; Determine the relationship strength weight between the row and column elements of each candidate complex feature matrix based on the relationship between the row elements, the relationship between the column elements and the directionality of their relationship; Determine the standard connection information flow between temperature data and time changes in the temperature change data under each candidate complex feature matrix according to the relationship strength weight; Acquire the current contact information flow of the temperature data and the time change in the temperature change data, match the current contact information flow with the standard contact information flow to determine a first candidate complex feature matrix with a matching degree greater than or equal to a preset threshold; Determine data sampling prior information according to the distribution characteristics of the depicted data of the first candidate complex feature matrix, and extract sample data from the temperature change data according to the data sampling prior information and a probability statistics method; Obtaining global density data parameters and local density data parameters of the sample data, and determining data distribution estimation parameters corresponding to each first candidate complex feature matrix according to the global density data parameters and the local density data parameters; Selecting a second candidate complex feature matrix from the first candidate complex feature matrix according to the data distribution estimation parameter and the data distribution of the temperature change data; The data complexity is determined according to the feature quantity of the second candidate complex feature matrix, and the distribution characteristics of the data are determined according to the data scale and the data complexity.

5. The multi-dimensional temperature measurement method of an organic electroluminescent display panel according to claim 1, characterized in that: The method captures an infrared image reflected by the surface of the organic electroluminescent display panel using an infrared camera, and determines the temperature distribution of the panel surface at different positions and directions according to the infrared image, including: Acquire the working environment and use conditions of the organic electroluminescent display panel, and set the parameters of the infrared camera according to the working environment and use conditions; The infrared camera is configured to capture an infrared image reflected from the surface of the organic electroluminescent display panel, and perform image preprocessing on the image; Generate a temperature distribution thermogram of the organic electroluminescent display panel at different positions and directions based on the preprocessed image based on the thermogram technology; Convert the temperature distribution thermodynamic map into a color gradient image to obtain temperature differences in different positions and directions; The temperature distribution of the surface of the organic electroluminescent display panel at different positions and directions is determined according to the temperature difference.

6. The multi-dimensional temperature measurement method of an organic electroluminescent display panel according to claim 1, characterized in that: According to the characteristic parameters, the temperature distribution on the panel surface and the change image of the pixel state, a multi-dimensional temperature prediction model is constructed based on a deep learning model, and the multi-dimensional temperature of the organic electroluminescent display panel is measured according to the multi-dimensional temperature prediction model, including: Acquire time series information of the organic electroluminescent display panel according to the characteristic parameters, the temperature distribution on the panel surface and the pixel state change image; Acquire a recurrent neural network model according to the time series information, and train and optimize the recurrent neural network model according to cross-validation; A multi-dimensional temperature prediction model is constructed based on the trained and optimized recurrent neural network model based on the panel surface temperature distribution data and corresponding features, and the multi-dimensional temperature of the organic electroluminescent display panel is measured based on the multi-dimensional temperature prediction model.

Images