Brightness and chrominance correction method and system for LED display screen

Abstract

The invention discloses a brightness and chrominance correction method and system for an LED display screen. The method comprises the following steps: acquiring ambient light data and original brightness and chrominance data of the LED display screen; performing preprocessing operation on the ambient light data to obtain ambient light standardized data; inputting the original brightness and chrominance data and the ambient light standardized data into a pre-trained ambient light depth compensation model to obtain a brightness and chrominance correction coefficient matrix; inputting the brightness and chrominance correction coefficient matrix and the original brightness and chrominance data into a pre-trained display correction optimization model to obtain a correction instruction; and sending the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen. According to the method, through application of ambient light compensation and display correction optimization, the brightness and the chromaticity of the display screen can be automatically corrected under different ambient light conditions, and the consistency and the stability of the brightness and the chromaticity are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of LED display screen correction, and in particular to a brightness and chromaticity correction method and system for an LED display screen. Background Art

[0002] At present, in modern display technology, LED display screens have become an important application for indoor and outdoor display, but due to the dynamic changes in ambient light, the stability of its brightness and chromaticity is greatly affected. For example, changes in ambient light at different times or locations will cause significant fluctuations in display effects, which in turn affects the user's viewing comfort. In addition, the brightness and chromaticity differences of the LED components themselves can also cause problems such as uneven display.

[0003] In one existing technology, an ambient light sensor is used to detect the ambient light intensity, and an automatic brightness adjustment algorithm is used to compensate for the display effect to improve the readability of the screen and user experience. This technology relies on a single compensation coefficient to convert the ambient light intensity into the brightness adjustment amount of the LED display screen, which can alleviate the impact of ambient light on the display effect to a certain extent.

[0004] However, the existing technology relies on a single compensation coefficient calculation method that cannot dynamically adapt to the changing characteristics of lighting conditions, thus lacking real-time and accurate control of the brightness and chromaticity of the display screen. Under changing and complex ambient light conditions, the existing system cannot automatically adjust the brightness and chromaticity of the display screen in real time, making it difficult to maintain the stability and consistency of the display effect. Summary of the invention

[0005] The present invention provides a brightness and chromaticity correction method and system for an LED display screen, so as to reduce the influence of ambient light on the display effect and improve the stability and consistency of the display effect.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a method for calibrating brightness and chromaticity of an LED display screen, comprising: Obtain ambient light data and original brightness and chromaticity data of LED display screen; Performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; Inputting the original luminance and chromaticity data and the ambient light normalized data into a pre-trained ambient light depth compensation model to obtain a luminance and chromaticity correction coefficient matrix; Inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; The correction instruction is sent to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

[0007] In an optional implementation, the training process of the ambient light depth compensation model includes: Building a convolutional neural network model based on the brightness response and chromaticity response of the LED display screen under different ambient lighting conditions and the brightness and chromaticity data of the LED display screen under an ideal environment, and training the convolutional neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the convolutional neural network model is less than the preset loss threshold, the training is determined to be completed, and the trained convolutional neural network model is used as the ambient light depth compensation model.

[0008] In an optional implementation, the training process of the display correction optimization model includes: Building a non-local neural network model based on historical ambient light characteristic parameters, historical LED display original brightness and chromaticity data, ideal display effect standard parameters and historical correction instruction data, and training the non-local neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the non-local neural network model is less than the preset loss threshold, the training is determined to be completed, and the non-local neural network model that has completed the training is used as a display correction optimization model.

[0009] In an optional implementation, before acquiring the ambient light data, the process includes: When the environment meets the preset trigger conditions under complex environmental conditions, new sensors are started or the detection range of existing sensors is adjusted.

[0010] In an optional implementation, the raw luminance and chromaticity data and the ambient light normalization data are input into a pre-trained ambient light depth compensation model to obtain a luminance and chromaticity correction coefficient matrix, including: The brightness and chromaticity correction coefficient matrix is ​​obtained by the following formula: , , , in, is the brightness correction factor, To control the coefficient of the proportion of the original brightness in the overall brightness adjustment, is the equilibrium constant, and Respectively represent the original brightness and chromaticity data of the LED display. and Respectively represent the brightness and chromaticity data after ambient light standardization, is the chromaticity adjustment coefficient, is the chroma adjustment gain coefficient, is a small constant, Represents the brightness and chromaticity correction coefficient matrix.

[0011] In an optional implementation, the step of inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction includes: The correction instruction includes a current magnitude adjustment value and a PWM magnitude adjustment value; The correction instruction is obtained by the following formula: , ; in, is the current adjustment value, is the PWM size adjustment value, is the brightness and chromaticity correction coefficient matrix, is the original brightness and chromaticity data matrix, ,in, and Respectively represent the original brightness and chromaticity data of the LED display. and are the nonlinear correction coefficients for brightness and chromaticity, and are the brightness and chromaticity offsets, and They are the ambient light brightness correction factor and the ambient light chromaticity correction factor respectively.

[0012] In an optional implementation, after sending the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen, the method further includes: Continuously monitor the ambient light sensor and the brightness and color sensor to obtain new ambient light data and currently displayed brightness and color data; When the new ambient light data changes by more than a preset stability threshold, the new ambient light data and the currently displayed brightness and chromaticity data are input into an ambient light depth compensation model and a display correction optimization model to generate a new correction instruction; When the new ambient light data change is less than the preset stability threshold, the current display state is maintained and no new correction instruction is generated.

[0013] In a second aspect, the present invention provides a brightness and chromaticity correction system for an LED display screen, comprising: A data acquisition module is used to acquire ambient light data and original brightness and chromaticity data of the LED display screen; A data processing module, used for performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; A coefficient generation module, used for inputting the original brightness and chromaticity data and the ambient light normalization data into a pre-trained ambient light depth compensation model to obtain a brightness and chromaticity correction coefficient matrix; An instruction generation module, used for inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; The display correction module is used to send the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

[0014] In a third aspect, the present invention further provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the brightness and chromaticity correction method of the LED display screen described in any one of the above is implemented.

[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the brightness and chromaticity correction methods for the LED display screen described above.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention obtains ambient light data and the original brightness and chromaticity data of the LED display screen, performs preprocessing to generate ambient light standardized data, and uses the pre-trained ambient light depth compensation model and display correction optimization model to obtain the brightness and chromaticity correction coefficient matrix and correction instructions respectively, thereby realizing automatic adjustment of the brightness and chromaticity of the LED display screen. The convolutional neural network and non-local neural network model adopted are trained based on a large amount of historical data, and have good generalization ability and prediction accuracy under different ambient light conditions, solving the problem that the traditional single compensation coefficient cannot cope with complex lighting conditions. The present invention can not only adjust the brightness and chromaticity of the display screen in real time and dynamically, improve the stability and consistency of the display effect, but also protect the display quality from the influence of the external environment by continuously monitoring the changes in ambient light and updating the correction instructions in time, thereby improving the user viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic flow chart of a brightness and chromaticity correction method for an LED display screen provided by a first embodiment of the present invention; Figure 2 It is a schematic diagram of the structure of a brightness and chromaticity correction system for an LED display screen provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] Reference Figure 1 The first embodiment of the present invention provides a method for calibrating brightness and chromaticity of an LED display screen, comprising the following steps: S11, obtaining ambient light data and original brightness and chromaticity data of the LED display screen; S12, performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; S13, inputting the original luminance and chromaticity data and the ambient light normalization data into a pre-trained ambient light depth compensation model to obtain a luminance and chromaticity correction coefficient matrix; S14, inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; S15, sending the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

[0020] In step S11, the main purpose is to collect the lighting information of the current environment of the display screen and the original display status data of the display screen itself, to ensure that the subsequent correction model can perform correction calculations according to accurate environmental conditions and display status, thereby achieving more accurate brightness and chromaticity adjustment.

[0021] In the specific implementation process, the acquisition of ambient light data can be achieved through an integrated light sensor. The sensor can detect the ambient light intensity and color temperature around the display in real time. Ambient light data includes the intensity value and color temperature information of the light, which can help the correction model analyze the potential impact of external light on the display effect of the display. For example, in a high-brightness environment, the display needs to increase the brightness appropriately to improve visibility; in an environment with a lower color temperature, the display may need to fine-tune the color to ensure color accuracy. Therefore, by collecting accurate ambient light data, the display can dynamically adjust the display effect under different environmental conditions.

[0022] Obtaining the original brightness and color data of the LED display is another key step. This data is read directly from the display control system and can reflect the current brightness and color status of each pixel. The original brightness data represents the luminous intensity of the display under the current driving conditions, measured in lumens (lm) or relative brightness values; while the color data describes the color characteristics of each pixel, such as RGB color values ​​or CIEXYZ color space coordinates. These raw data can help identify whether the display has problems such as uneven brightness and color deviation during the display process, and provide basic input for the correction model.

[0023] It should be noted that the acquisition process of ambient light data and the original brightness and chromaticity data of the LED display screen will be affected by external conditions, hardware performance, and sampling accuracy. Therefore, in order to ensure the accuracy and consistency of the data, a variety of optimization measures will be taken in the data collection process. For example, multi-point sampling can be used to obtain ambient light data to ensure that the light intensity at different positions and angles can be fully perceived. In addition, the ambient light sensor may need to be calibrated and calibrated to eliminate the inherent deviation of the sensor and ensure that the acquired ambient light data can truly reflect the characteristics of the external ambient light.

[0024] On the other hand, the original brightness and color data of the display screen is usually directly read through the internal monitoring system to ensure the real-time and accuracy of the data. In order to further improve the acquisition accuracy of the original brightness and color data of the LED display screen, the method of regional measurement or pixel-by-pixel acquisition can be used to obtain more detailed brightness and color distribution information. For some application scenarios with high color requirements, the internal light metering device of the display screen can also be added to obtain more comprehensive display status information by combining it with the ambient light sensor.

[0025] In practical applications, these acquired ambient light data and raw brightness and color data will be input into the subsequent ambient light depth compensation model and display correction optimization model, providing necessary input for the entire correction system. Ambient light data provides a reference background for display correction, while raw brightness and color data provides the correction model with the current status information of the display screen. Together, the two constitute the key basis for correction adjustment. In most cases, this data collection is performed periodically to ensure that the display screen can respond and adjust in time when the ambient light changes, so as to always maintain appropriate brightness and color output.

[0026] In a preferred implementation, before obtaining the ambient light data, the process includes: When the environment meets the preset trigger conditions under complex environmental conditions, new sensors are started or the detection range of existing sensors is adjusted.

[0027] It should be noted that this design is to be able to respond more effectively to changes in the external environment under specific lighting conditions (such as strong light, low light, multiple light sources crossing, drastic changes in color temperature, etc.), so that the system has higher adaptability and data collection accuracy. The core goal of this pre-operation is to ensure the authenticity and accuracy of ambient light data, because under complex lighting conditions, ordinary ambient light sensors may cause deviations in collected data due to limitations in detection range, sensitivity or resolution, and newly activated sensors or adjusted detection ranges can make up for these deficiencies and ensure the comprehensiveness and accuracy of ambient light data.

[0028] Specifically, when the system detects an ambient light trigger condition, it will start additional sensors according to the preset strategy. For example, when the external light intensity exceeds a certain threshold (such as strong sunlight or direct light), the system will automatically turn on the high-intensity ambient light sensor, which has higher sensitivity and a wider dynamic range, so that it can still accurately capture light intensity and color temperature changes in strong light environments. On the other hand, when the system detects that the ambient light is weak or there is a complex combination of light sources (such as the presence of multiple light sources with different color temperatures at the same time), it will start a dedicated low-light environment sensor or color temperature compensation sensor to ensure that the ambient light characteristics can be accurately collected in low-brightness environments or complex light source combinations. This multi-sensor coordination method can improve the resolution of ambient light data and enhance the system's adaptability to complex lighting conditions.

[0029] It should be noted that the system does not enable additional sensors or adjust the detection range of sensors in all cases, but decides whether to perform these operations based on preset trigger conditions to balance power consumption and acquisition accuracy. In specific implementations, these trigger conditions are usually implemented by monitoring and determining factors such as ambient light intensity and color temperature change rate. For example, if the system detects that the light intensity fluctuation exceeds a certain threshold or the color temperature changes significantly in a short period of time (such as a color temperature change of more than 300K), this operation will be triggered. This trigger-based mechanism can dynamically allocate sensor resources when necessary, making the system more intelligent in complex environments while also avoiding unnecessary energy consumption.

[0030] In addition, when adjusting the detection range of existing sensors, the system will take measures such as adjusting gain, changing sampling frequency or resolution to enable existing sensors to adapt to current environmental conditions. For example, when the lighting environment changes dramatically, the system will automatically increase the sampling frequency of the sensor to capture rapid lighting changes in a timely manner. For the adjustment of the detection range, the system may switch the sensitivity mode of the sensor under different environmental conditions, such as reducing the gain in a strong light environment to avoid sensor overload, and increasing the gain in a low light environment to improve signal quality. In addition, if it is detected that the light source in the environment has a large color temperature difference, the system will enable the multi-channel acquisition mode to collect data from three different spectral channels of red, green, and blue in order to more accurately restore the color temperature characteristics of the current ambient light. These measures together constitute an adaptive mechanism for ambient light data acquisition to adapt to various complex lighting scenarios and ensure the accuracy and comprehensiveness of ambient light data.

[0031] Through this preferred implementation, the system can collect ambient light data more accurately under complex environmental conditions, so that the subsequent display correction model can obtain more realistic input data. In other words, no matter what kind of lighting environment the display screen is in, the system can respond to changes in external lighting in a timely manner and make corresponding data collection adjustments to provide reliable environmental data support for the final brightness and chromaticity correction. This method is suitable for various application scenarios, especially outdoor displays, commercial displays, and other occasions that need to frequently respond to changes in ambient light, thereby ensuring that the display effect of the display device under different lighting conditions remains consistent and stable.

[0032] In step S12, the main purpose is to convert the original ambient light data into standardized data that can be directly used for display correction, so as to improve the data processing efficiency and the adaptability of the model to external environmental changes. In practical applications, the original collection results of ambient light data may be affected by various factors such as external noise, sampling errors, and changes in the position of the light source. The preprocessing step improves the accuracy and consistency of the data through a series of data cleaning, normalization, filtering and other technologies.

[0033] Specifically, the preprocessing operation of ambient light data includes multiple stages, the first of which is noise filtering. In complex environments, ambient light sensors are affected by factors such as reflected light and light interference from other devices, resulting in noise in the collected ambient light data. These noises interfere with the judgment of the actual ambient light characteristics, which in turn affects the accuracy of the correction. Therefore, in the first step of data preprocessing, the system eliminates noise from the ambient light data through filters (such as mean filtering, Gaussian filtering, etc.). In this way, random high-frequency noise in the data is removed, ensuring that the data used in subsequent steps has better quality.

[0034] Next is the data normalization operation. The output of the ambient light sensor varies due to factors such as hardware configuration and light source intensity differences, so the data needs to be normalized. The purpose of normalization is to convert data of different magnitudes into values ​​within a standard range. For example, mapping the ambient light intensity to a range of 0 to 1 ensures that the system can obtain a uniform output value under different acquisition conditions. The normalization process can use methods such as linear normalization or logarithmic normalization. This operation not only facilitates subsequent processing and model calculations, but also ensures data standardization so that light data at different times and locations can be compared and analyzed at the same magnitude.

[0035] On this basis, the ambient light data is also analyzed for color temperature and spectrum. The original ambient light data usually contains light intensity and color temperature information, and the latter is an important parameter that affects the display effect, because different color temperatures will give people different visual perceptions of warm and cold tones. Therefore, during the data preprocessing process, the system will calculate its color temperature characteristics based on the spectral distribution of the ambient light. The color temperature data is directly measured by the color temperature sensor and standardized to a preset color temperature range. In this way, the standardized color temperature data can be used as the input of the display correction model, so that the correction system can automatically adjust the color performance of the display according to the color temperature of the environment, ensuring the natural and consistent display effect.

[0036] After processing basic data such as light intensity and color temperature, dynamic changes will be smoothed. In practical applications, the intensity of ambient light often fluctuates with time or changes in environmental conditions. Especially in outdoor scenes, cloud cover, changes in sunlight angle, etc. will cause frequent changes in ambient light data. If these fluctuating data are directly input into the correction model, the display may frequently adjust the brightness and chromaticity, affecting the user experience. To this end, during the preprocessing of ambient light data, the system will use smoothing algorithms (such as sliding average, exponential smoothing, etc.) to smoothly process the data for dynamic changes. The smoothed data is more stable, which helps reduce unnecessary display adjustments and makes the display effect smoother and more natural.

[0037] In step S13, the core purpose is to accurately adjust the brightness and chromaticity of the display screen under different ambient light conditions through the depth compensation model to meet the user's demand for display effects. The role of the depth compensation model is to compensate for the deviation of brightness and chromaticity based on the current environmental conditions, thereby generating a set of correction coefficient matrices for subsequent application to the automatic correction of the brightness and chromaticity of the display screen.

[0038] First, the input of step S13 includes two parts of data: one is the original brightness and chromaticity data of the LED display screen, and the other is the pre-processed ambient light standardized data. The original brightness and chromaticity data comes directly from the detection information of the display screen, and records the brightness and chromaticity information of each pixel or lamp bead in the current display state of the screen. These data reflect the original display effect of the screen without correction. The ambient light standardized data is the standardized processing result completed in step S12, which characterizes the external light intensity and color temperature information. Since ambient light directly affects the visual performance of the display screen, the ambient light data is used as the input of the compensation model, which can help the model give adaptive brightness and chromaticity adjustment schemes under different lighting conditions.

[0039] The pre-trained ambient light depth compensation model is one of the key components of the present invention. It is trained based on a large amount of historical data to ensure that it can generate a correction coefficient matrix with high generalization and accuracy under different ambient light conditions.

[0040] In this step, after being processed by the depth compensation model, the output correction coefficient matrix is ​​a set of multidimensional parameter matrices. Each element in the matrix represents the brightness or chromaticity adjustment coefficient that the corresponding pixel should perform in the current environment, thereby achieving accurate compensation for the original brightness and chromaticity. Specifically, the brightness and chromaticity correction coefficients in the matrix can be divided into two parts: one is the basic gain coefficient, and the other is the nonlinear adjustment coefficient. The basic gain coefficient is used to make an overall increase or decrease in the brightness or chromaticity of the display screen, while the nonlinear adjustment coefficient is to make subtle adjustments to the local chromaticity or brightness deviation of the display screen. Through the combination of these two parts of the coefficients, the correction matrix can achieve global and local optimization of the display effect, thereby improving the visual consistency of the display effect.

[0041] It should be noted that the correction coefficient matrix in step S13 not only provides a parameter basis for adaptive adjustment of the current display screen, but also has good dynamic response capabilities. Due to the frequent changes in ambient light, such as changes in the brightness of sunlight, the switching of indoor lighting, etc., the real-time calculation capability of the depth compensation model can generate a new correction coefficient matrix in time according to the changes in ambient light data. In this way, under complex lighting conditions, the display screen can obtain and apply the latest correction instructions in real time, so that its brightness and chromaticity performance are always maintained at an ideal level.

[0042] In practical applications, the correction coefficient matrix generated by step S13 will directly participate in the subsequent correction instruction generation and display adjustment process, providing accurate and stable adjustment parameters for the final display correction. This correction coefficient matrix generated based on the depth compensation model can not only improve the display effect, but also reduce the complexity and uncertainty of manual adjustment in traditional correction technology.

[0043] In a preferred implementation, the training process of the ambient light depth compensation model includes: Building a convolutional neural network model based on the brightness response and chromaticity response of the LED display screen under different ambient lighting conditions and the brightness and chromaticity data of the LED display screen under an ideal environment, and training the convolutional neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the convolutional neural network model is less than the preset loss threshold, the training is determined to be completed, and the trained convolutional neural network model is used as the ambient light depth compensation model.

[0044] It should be noted that the training process of the ambient light depth compensation model relies on a convolutional neural network (CNN), which aims to generate correction parameters suitable for different lighting scenes by simulating the brightness and chromaticity responses of LED displays under various ambient light conditions. The basic idea of ​​this process is to use the powerful feature extraction and recognition capabilities of convolutional neural networks to automatically analyze the impact of ambient light on the brightness and chromaticity of the display, thereby ensuring that the model can maintain accurate display correction effects under variable lighting conditions. The convolutional neural network is selected as the core model mainly because CNN has unique advantages in spatial feature extraction, nonlinear relationship modeling, and model generalization performance, which can meet the technical requirements of the ambient light depth compensation model.

[0045] First of all, convolutional neural networks are highly adaptable and effective in visual feature extraction. In the actual application scenarios of LED displays, the changes in ambient light are often very complex, involving multiple factors such as different intensities, color temperatures, and directions of light sources, and the combination of these factors will have a multi-level impact on the brightness and chromaticity performance of the display. In order to address this complexity, the convolutional layers in the convolutional neural network can gradually extract the brightness and chromaticity features of the display under different lighting conditions, including basic information such as edges, contours, and textures. By superimposing multiple convolutional layers, the model can gradually combine low-level features into higher-level features, capture the nonlinear relationship between lighting conditions and display effects, and thus generate more accurate brightness and chromaticity adjustment parameters.

[0046] Secondly, the convolutional neural network is highly adjustable and can continuously optimize the parameters of the model according to the feedback of the loss function during the training process, so that the output of the model gradually approaches the ideal effect. In this implementation, the training data of the model includes two parts: one is the brightness and chromaticity response data of the LED display under different ambient light conditions, and the other is the brightness and chromaticity data of the display under an ideal standard environment. The goal of the training process is to let the model learn the impact of different lighting conditions on the brightness and chromaticity of the display, so that it can output appropriate correction parameters when given ambient light data input. The loss function of the convolutional neural network plays a key role in this process. It reflects the gap between the correction effect currently output by the model and the ideal correction effect. By repeatedly adjusting the model weights, the parameters of the convolutional neural network are gradually optimized, so that it can generate accurate brightness and chromaticity correction coefficient matrices under various ambient light conditions.

[0047] Another important reason for choosing convolutional neural networks is their strong nonlinear modeling capabilities. The brightness and chromaticity responses of LED displays under different lighting conditions are highly nonlinear. For example, an increase in light intensity does not linearly increase the brightness of the display, but may have the opposite effect under certain conditions. Convolutional neural networks can effectively handle this complex nonlinear relationship through nonlinear activation functions. Activation functions such as ReLU can introduce nonlinear computing capabilities into the model, enabling the model to adapt to nonlinear changes in the impact of light on the display effect. Compared with traditional linear regression or simple polynomial regression models, convolutional neural networks have more advantages in dealing with nonlinear problems and can generate more stable and accurate correction coefficient matrices.

[0048] In addition, the convolutional neural network has good generalization ability. Even when the ambient light conditions change significantly, the model can still provide effective correction parameters. This feature comes from the convolutional neural network's extraction of local features and weight sharing mechanism. By adding various typical lighting conditions such as strong light, low light, and uneven light sources to the training data, the model can learn a wider range of features and form adaptability to changes in ambient light.

[0049] Two methods are set for the termination conditions of model training: one is that when the number of training times reaches the preset maximum number of training times, the model is considered to have been sufficiently optimized and training is stopped; the other is that when the loss function value of the convolutional neural network is lower than the preset threshold, it means that the output of the model is close enough to the ideal effect and training can be stopped. The setting of these two methods can prevent the model from overfitting during training and ensure the rational use of training resources. The former is to limit the maximum cycle of model training to prevent excessive training time from wasting computing resources; the latter is to set the loss threshold and terminate the training in time when the model error is reduced to the expected range to ensure the generalization performance of the model.

[0050] In a preferred implementation, the original luminance and chromaticity data and the ambient light normalization data are input into a pre-trained ambient light depth compensation model to obtain a luminance and chromaticity correction coefficient matrix, including: The brightness and chromaticity correction coefficient matrix is ​​obtained by the following formula: , , ; in, is the brightness correction factor, To control the coefficient of the proportion of the original brightness in the overall brightness adjustment, is the equilibrium constant, and Respectively represent the original brightness and chromaticity data of the LED display. and Respectively represent the brightness and chromaticity data after ambient light standardization, is the chromaticity adjustment coefficient, is the chroma adjustment gain coefficient, is a small constant, Represents the brightness and chromaticity correction coefficient matrix.

[0051] It should be noted that the brightness correction factor The calculation method is based on the original brightness data of the display Brightness data normalized with ambient light In the formula, Indicates the brightness of the current ambient light, and It is the original brightness information of the LED screen when it is not calibrated. By combining these two factors, it is possible to dynamically adjust to reflect the amount of adjustment required for the display brightness under the current ambient light. In the formula, it is used as a control coefficient to control the proportion of the original brightness in the overall brightness adjustment. This means that The size of The degree of influence on the correction factor, this control factor It is learned by the CNN model during the training process. It reflects the adjustment requirements of the ambient light on the display brightness, making the final brightness adjustment more accurate. Further provides a baseline amount to prevent The fluctuation caused by the lower value also helps the numerical stability during the correction process. The parameters included in the calculation formula are , , and . and Represent the standardized chromaticity data of the current ambient light and the original chromaticity data of the display, respectively. is a chromaticity gain coefficient, which is used to control the weight of ambient light chromaticity in the correction. Through this formula, The changes will affect The value of , thus dynamically compensating the display color when the ambient color changes. In addition, The value is a small constant used to prevent the denominator from having a zero value, which would cause unstable calculations. This ensures that the correction calculation remains stable even when the original chromaticity data is low. In this way, the chromaticity correction of the display can still have an accurate adjustment effect in low chromaticity scenes. Correction coefficients for brightness and chromaticity and Finally, it is combined into a diagonal matrix , which is the brightness and chromaticity correction coefficient matrix. This matrix contains two independent adjustment coefficients for brightness and chromaticity, so it can accurately correct the brightness and chromaticity of the display screen respectively. In the specific correction process, It is applied to the original brightness and color data of the display screen, and through matrix transformation, the corrected brightness and color data are generated. Such a matrix design not only improves the calculation efficiency, but also makes the correction process more direct and efficient. The various parameters in the above formula (such as the control coefficient and the chroma gain coefficient etc.) are optimized and determined during the training phase of the depth compensation model. During the training process, the model gradually adjusts these parameters through a large amount of brightness and chromaticity response data under different ambient light conditions so that the model can adapt to the changing ambient light conditions. The process of training with a convolutional neural network gives the model a strong generalization ability, which can effectively correct the brightness and chromaticity under different ambient light conditions. In summary, this process can not only adapt to changes in ambient light through the parameterized design of the ambient light compensation model and CNN training, but also ensure the stability and consistency of the display effect under different lighting conditions. This solution performs real-time correction of the brightness and chromaticity of the LED display under different ambient light conditions, effectively improving the user's viewing experience and visual comfort.

[0052] In step S14, the brightness and color correction coefficient matrix and the original brightness and color data are input into the trained display correction optimization model to generate correction instructions suitable for the current environment. The core of this process is to use the preprocessed data and the correction model to generate specific instructions to optimize the display effect. The correction instructions include the current adjustment value and the pulse width modulation (PWM) adjustment value. These two parameters can effectively control the brightness and color of the LED display, thereby ensuring that the visual effect of the display is always in line with expectations under different lighting conditions.

[0053] The display correction optimization model is based on a large amount of training data and can generate accurate correction instructions. This model converts the input brightness and color correction coefficient matrix and raw brightness and color data into actual hardware adjustment parameters by learning the relationship between ambient light changes and display requirements. Specifically, the brightness and color correction coefficient matrix contains compensation parameters for adjusting brightness and color according to ambient light conditions, while the raw brightness and color data reflects the brightness and color state of the LED display in an uncorrected state. These data are combined in the model to generate an adjustment scheme for current and PWM values ​​to ensure that the display presents an ideal display effect under the current ambient light conditions.

[0054] It should be noted that the principle of this correction instruction design is based on the following key factors. First, the current adjustment value directly affects the driving strength of the LED pixel, thereby adjusting the overall brightness. On LED displays, by precisely controlling the current, the brightness can be increased while avoiding the increased energy consumption and display fatigue caused by excessive brightness. In addition, the size of the current adjustment value is based on the model's analysis of the input brightness and ambient light data to ensure adaptability. For example, in a high-light environment, the model may generate a larger current adjustment value to increase the screen brightness so that the image content is clearly visible; in a low-light environment, the current will be reduced accordingly to avoid excessive light pollution.

[0055] Secondly, the PWM size adjustment value controls the light-emitting cycle of the LED lamp beads. The principle of PWM control is to affect the brightness and chromaticity by changing the frequency and time ratio of the LED on, which is especially suitable for subtle brightness adjustment and color correction. The PWM control value mainly affects the grayscale level of the LED, so that accurate color adjustment can be achieved at a more subtle level. For example, in scenes that require higher saturation or more accurate color performance, PWM adjustment can appropriately increase the chromaticity to make the display effect more vivid. In addition, PWM control is more flexible, so subtle brightness adjustments can be made without changing the current intensity. This is especially important in specific ambient light changes, which can maintain energy saving effects and avoid frequent hardware losses.

[0056] During the training process of the display correction optimization model, a large amount of historical ambient light data, the original brightness and color data of the LED display, the ideal display effect standards and historical correction instructions were included in the training set to improve the generalization and adaptability of the model. The ambient light data includes the changes in light intensity and color temperature under various conditions such as sunny days, cloudy days, and nights; the original brightness and color data of the display record the screen performance in the uncorrected state. These data provide the model with multi-dimensional training information, so that it can quickly generate correction instructions that meet the current environment in practical applications.

[0057] The generation process of the correction command aims to maximize the user's visual experience. The model comprehensively considers the adjustment effects of current and PWM through internal nonlinear analysis. For example, in a strong light environment outdoors, the correction command may require an increase in the current value and a high-frequency, large-ratio adjustment on PWM to ensure a high-brightness display effect; in an indoor or low-light environment, a relatively low-current, high-ratio PWM command may be generated to ensure a comfortable viewing experience and maintain reasonable control of energy consumption.

[0058] It should be noted that the current and PWM value adjustment algorithm built into the display correction optimization model is not only based on the single brightness adjustment requirement, but also takes into account color calibration and energy management. The current adjustment can control the overall brightness, and the PWM adjustment can optimize the display quality while maintaining color accuracy. The final generated correction instructions can be directly executed in actual applications, thereby realizing intelligent and automated control of the brightness and color of the LED display.

[0059] In a preferred implementation, the training process of the display correction optimization model includes: Building a non-local neural network model based on historical ambient light characteristic parameters, historical LED display original brightness and chromaticity data, ideal display effect standard parameters and historical correction instruction data, and training the non-local neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the non-local neural network model is less than the preset loss threshold, the training is determined to be completed, and the non-local neural network model that has completed the training is used as a display correction optimization model.

[0060] It should be noted that during the training process, the ambient light feature parameters reflect the changes under different lighting conditions, such as the intensity and color temperature of the light source. These parameters can help the model identify specific ambient light features and generate corresponding correction instructions based on this. Since ambient light varies greatly in different time periods and scenes, these parameters provide key contextual information for the model to ensure that the correction model can make reasonable adjustments under various lighting conditions.

[0061] At the same time, the original brightness and chromaticity data of historical LED displays are also used for model training. These data describe the brightness and chromaticity output of the display in an uncorrected state. By learning these data, the model can grasp the original response characteristics of the display itself under different ambient light conditions, providing a basis for subsequent brightness and chromaticity adjustments. Combined with the standard parameters of ideal display effects, the model can learn and refer to specific ideal display standards, and optimize the display with this as the goal, so that the final correction effect can meet the expected visual standards.

[0062] In this process, historical correction instruction data is also used to guide the model to generate optimization instructions. These instruction data record the correction schemes under different conditions in the past, including specific adjustment values ​​for current intensity and PWM parameters. By analyzing these data, the model can further optimize the correction effect under similar environmental conditions. For example, when the model detects lighting conditions similar to a historical environment, it can make corresponding adjustments based on the corresponding correction instruction data, thereby improving the stability of the display effect.

[0063] In the process of training the non-local neural network model, the model parameters are constantly adjusted with the goal of minimizing the error of the correction result. Specifically, the training process will stop when the loss function value gradually converges or reaches the preset number of training times. The loss function is used to measure the gap between the correction effect and the ideal display effect. If the value of the loss function is less than the preset loss threshold, it means that the correction instructions generated by the model are very close to the ideal effect, and the training is completed. Training will also stop when the preset maximum number of training times is reached to ensure a balance between training efficiency and effect.

[0064] It should be noted that the main reason for using the non-local neural network model is that it can efficiently capture the global characteristics of display brightness and chromaticity changes, and provide sophisticated correction solutions under multiple ambient light characteristics. The advantage of the non-local neural network model lies in its ability to capture long-distance feature dependencies, especially when different environmental conditions have complex effects on the display effect. Finally, the trained non-local neural network model is used as a display correction optimization model, and real-time data is used in practical applications to generate correction instructions to achieve automatic and accurate brightness and chromaticity adjustment.

[0065] In a preferred implementation, the step of inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction includes: The correction instruction includes a current magnitude adjustment value and a PWM magnitude adjustment value; The correction instruction is obtained by the following formula: , ; in, is the current adjustment value, is the PWM size adjustment value, is the brightness and chromaticity correction coefficient matrix, is the original brightness and chromaticity data matrix, ,in, and Respectively represent the original brightness and chromaticity data of the LED display. and are the nonlinear correction coefficients for brightness and chromaticity, and are the brightness and chromaticity offsets, and They are the ambient light brightness correction factor and the ambient light chromaticity correction factor respectively.

[0066] It should be noted that the brightness correction factor The calculation method is based on the original brightness data of the display Brightness data normalized with ambient light In the formula, Indicates the brightness of the current ambient light, and It is the original brightness information of the LED screen when it is not calibrated. By combining these two factors, it is possible to dynamically adjust to reflect the amount of adjustment required for the display brightness under the current ambient light. It is used as a control coefficient in the formula to control the proportion of the original brightness in the overall brightness adjustment. The size of The degree of influence on the correction factor, this control factor It is learned by the non-local neural network model during the training process. It reflects the adjustment requirements of the ambient light on the display brightness, making the final brightness adjustment more accurate. Further provides a baseline amount to prevent The fluctuation caused by the lower value also helps the numerical stability during the correction process. The parameters included in the calculation formula are , , and . and Represent the standardized chromaticity data of the current ambient light and the original chromaticity data of the display, respectively. is a chromaticity gain coefficient, which is used to control the weight of ambient light chromaticity in the correction. Through this formula, The changes will affect The value of , thus dynamically compensating the display color when the ambient color changes. In addition, The value is a small constant used to prevent the denominator from having a zero value, which would cause unstable calculations. This ensures that the correction calculation remains stable even when the original chromaticity data is low. In this way, the chromaticity correction of the display can still have an accurate adjustment effect in low chromaticity scenes. Correction coefficients for brightness and chromaticity and Finally, it is combined into a diagonal matrix , which is the brightness and chromaticity correction coefficient matrix. This matrix contains two independent adjustment coefficients for brightness and chromaticity, so it can accurately correct the brightness and chromaticity of the display screen respectively. In the specific correction process, It is applied to the original brightness and color data of the display screen, and through matrix transformation, the corrected brightness and color data are generated. Such a matrix design not only improves the calculation efficiency, but also makes the correction process more direct and efficient. The various parameters in the above formula (such as the control coefficient and the chroma gain coefficient etc.) are optimized and determined during the training phase of the depth compensation model. During the training process, the model gradually adjusts these parameters through a large amount of brightness and chromaticity response data under different ambient light conditions so that the model can adapt to the changing ambient light conditions. The process of training with a convolutional neural network gives the model a strong generalization ability, which can effectively correct brightness and chromaticity under different ambient light conditions. In summary, this process can not only adapt to changes in ambient light through the parameterized design of the ambient light compensation model and the training of non-local neural networks, but also ensure the stability and consistency of the display effect under different lighting conditions.

[0067] In step S15, the generated correction instructions are sent to the LED display screen for real-time brightness and chromaticity adjustment so that the display effect of the display screen adapts to the current ambient light conditions. These correction instructions consist of current size adjustment values ​​and PWM (pulse width modulation) size adjustment values. By adjusting the power supply current and duty cycle of the LED display screen, the brightness and chromaticity output of the display screen can be accurately controlled, thereby achieving adaptive correction of the ambient light. It should be noted that this step not only involves one-time sending and adjustment, but also includes a set of continuous monitoring and feedback adjustment mechanisms to ensure that the display effect can be maintained within the ideal viewing range even when the ambient light changes greatly.

[0068] In a preferred implementation, after sending the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen, the method further includes: Continuously monitor the ambient light sensor and the brightness and color sensor to obtain new ambient light data and currently displayed brightness and color data; When the new ambient light data changes by more than a preset stability threshold, the new ambient light data and the currently displayed brightness and chromaticity data are input into an ambient light depth compensation model and a display correction optimization model to generate a new correction instruction; When the new ambient light data change is less than the preset stability threshold, the current display state is maintained and no new correction instruction is generated.

[0069] It should be noted that after the correction command is sent to the LED display to complete the brightness and chromaticity adjustment, the system will continue to enable the real-time monitoring operation of the ambient light sensor and the brightness and chromaticity sensor. Continue to obtain the light intensity and color temperature information in the current environment, as well as the brightness and chromaticity data actually output on the display. By using this data as feedback input, the system can effectively monitor the matching degree between the display correction effect and the ambient light conditions.

[0070] The system determines whether further adjustment is needed based on the data changes fed back by the sensor. When the new ambient light data changes exceed the preset stability threshold, this indicates that the ambient light conditions have changed significantly. For example, when switching from a bright outdoor environment to a darker indoor environment, or in a scene where the light changes significantly between morning and evening, the LED display may need to readjust the output brightness and chromaticity to adapt to new visual needs. At this time, the system inputs the new ambient light data and the currently displayed brightness and chromaticity data into the ambient light depth compensation model and the display correction optimization model, and regenerates the correction instructions. These new correction instructions will again adjust the display effect of the LED display by adjusting the current and PWM signal size to ensure that the final output can quickly adapt to changes in the environment, thereby providing users with a consistent and clear visual experience.

[0071] On the other hand, when the new ambient light data change is less than the preset stability threshold, it means that the ambient light conditions are in a relatively stable state, and the current display correction is sufficient to meet the display requirements. In this case, the system will not generate new correction instructions, but maintain the current display state. This strategy not only ensures the stability of the display effect, but also avoids the frequent sending of adjustment instructions, reduces the occupation of system resources and the dynamic fluctuation of the display, and helps to extend the service life of the device.

[0072] It should be noted that the preset stability threshold is determined by a large amount of experimental data and the characteristics of ambient light. The setting of this threshold balances the response speed of the LED display and the degree of influence of ambient light changes, thereby achieving the best balance between stability and response speed.

[0073] Through this combination of continuous monitoring and on-demand adjustment, the LED display can always maintain a high degree of sensitivity to changes in external ambient light, thereby dynamically adjusting brightness and chromaticity, so that the audience's visual experience can be maintained at a relatively consistent level under different ambient light conditions. This process realizes the intelligent adjustment function of the device, allowing the display to maintain the best display effect indoors and outdoors, in environments with changing light and dark, and will not cause problems such as too bright or too dark display effects due to environmental changes.

[0074] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0075] In one implementation, the present invention can be applied to LED display screens of large outdoor billboards, especially in commercial districts where day and night frequently alternate. Such display screens need to respond quickly to changes in ambient light to ensure that the best display effect is maintained whether in direct sunlight during the day or at night when the lights are dim. By integrating a highly sensitive ambient light sensor, the system can monitor changes in ambient light in real time and adjust the brightness and chromaticity of the display screen accordingly. For example, at sunset, the system will automatically reduce the brightness of the display screen to reduce light pollution, while optimizing the chromaticity so that the advertising content remains bright and not dazzling even at night, providing pedestrians with a comfortable visual experience.

[0076] In another implementation, the present invention is also applicable to indoor scenes, such as LED display screens in large shopping malls or exhibition halls. In these occasions, since the lighting conditions may change with time and activities, the display effect of the display screen also needs to be adjusted accordingly. The present invention can sense these subtle changes in real time through built-in brightness and chromaticity sensors and ambient light sensors. For example, when holding special events, the on-site lighting may become softer or the colors may be richer. At this time, the system will automatically adjust the color temperature of the display screen to match the ambient atmosphere, thereby enhancing the atmosphere of the event and improving the visitor experience. In addition, for those display contents that have extremely high color requirements, such as art exhibitions, the present invention can ensure the color reproduction of the exhibits, allowing visitors to see the most realistic works of art.

[0077] The following describes the working process of the present invention using a common scenario as an example. Figure 2 , which is Figure 1 Schematic diagram of the working scenario of the method.

[0078] In modern smart shopping malls, the stability of brightness and color directly affects the visual experience of customers. As the ambient light inside and outside the mall changes, especially under natural light at different time periods, the performance of the display may be significantly affected. In order to solve this problem, an intelligent adjustment solution based on ambient light depth compensation and display correction optimization model is adopted.

[0079] Suppose there is a large screen displaying advertisements at the entrance of a shopping mall. During the day, the natural light outside is strong, while at night, the internal lighting of the mall dominates the brightness and color output of the display. In this process, the display needs to be able to adjust its display effect in real time to cope with changes in ambient light.

[0080] When the light outside the mall is strong, the brightness of the LED display may appear dim, affecting the clarity and attractiveness of the advertisement. At this time, the ambient light sensor detects the change in external light and transmits the new ambient light data to the system. At the same time, the brightness and color sensor also captures the current brightness and color information of the display in real time. By inputting this data into the pre-trained ambient light depth compensation model, the system will output a set of compensation coefficients to adjust the brightness and color accordingly based on the current ambient light.

[0081] The system then combines these compensation coefficients with the original brightness and color data to generate new correction instructions through the display correction optimization model. These correction instructions include adjusting the value of the LED display current and the duty cycle of the PWM signal to adjust the brightness and color of the display. This process not only ensures that the display can adapt to the ambient light conditions, but also avoids the need for human intervention or frequent adjustments. The correction instructions are sent to the LED display in real time to quickly adjust the display effect.

[0082] In summary, in this process, the system can not only adapt to large changes in ambient light (such as the transition from day to night), but also display stably under smaller changes. When the external ambient light changes slightly and remains stable, the system will maintain the existing display effect according to the set stability threshold and no longer make unnecessary corrections, thus avoiding energy waste or fluctuations in display effects caused by over-adjustment.

[0083] Therefore, in this smart shopping mall environment, the LED display can maintain ideal visual performance under various complex ambient light conditions, attract customers' attention and improve advertising effects. Whether it is during the day when the external light is strong or in a darker indoor environment, the system can automatically adjust to ensure clear and stable display effects, improving the communication efficiency of advertising in the mall and the user experience.

[0084] Reference Figure 2 The second embodiment of the present invention provides a brightness and chromaticity correction system for an LED display screen, comprising: A data acquisition module is used to acquire ambient light data and original brightness and chromaticity data of the LED display screen; A data processing module, used for performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; A coefficient generation module, used for inputting the original brightness and chromaticity data and the ambient light normalization data into a pre-trained ambient light depth compensation model to obtain a brightness and chromaticity correction coefficient matrix; An instruction generation module, used for inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; The display correction module is used to send the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

[0085] It should be noted that the brightness and chromaticity correction system of an LED display screen provided in an embodiment of the present invention is used to execute all the process steps of the brightness and chromaticity correction method of an LED display screen in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be repeated here.

[0086] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a brightness and chromaticity correction method program for an LED display screen. When the processor executes the computer program, the steps in the brightness and chromaticity correction method embodiments of the LED display screen are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the display correction module.

[0087] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0088] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0089] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0090] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0091] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0092] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0093] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calibrating brightness and chromaticity of an LED display screen, characterized in that: include: Obtain ambient light data and original brightness and chromaticity data of LED display screen; Performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; Inputting the original luminance and chromaticity data and the ambient light normalized data into a pre-trained ambient light depth compensation model to obtain a luminance and chromaticity correction coefficient matrix; Inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; The correction instruction is sent to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

2. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: The training process of the ambient light depth compensation model includes: Building a convolutional neural network model based on the brightness response and chromaticity response of the LED display screen under different ambient lighting conditions and the brightness and chromaticity data of the LED display screen under an ideal environment, and training the convolutional neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the convolutional neural network model is less than the preset loss threshold, the training is determined to be completed, and the trained convolutional neural network model is used as the ambient light depth compensation model.

3. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: The training process of the display correction optimization model includes: Building a non-local neural network model based on historical ambient light characteristic parameters, historical LED display original brightness and chromaticity data, ideal display effect standard parameters and historical correction instruction data, and training the non-local neural network model; When the number of training times is greater than or equal to the preset maximum number of training times, or when the loss function value of the non-local neural network model is less than the preset loss threshold, the training is determined to be completed, and the non-local neural network model that has completed the training is used as a display correction optimization model.

4. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: Before obtaining the ambient light data, the method includes: When the environment meets the preset trigger conditions under complex environmental conditions, new sensors are started or the detection range of existing sensors is adjusted.

5. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: The original brightness and chromaticity data and the ambient light normalized data are input into a pre-trained ambient light depth compensation model to obtain a brightness and chromaticity correction coefficient matrix, including: The brightness and chromaticity correction coefficient matrix is ​​obtained by the following formula: ， ， ； in, is the brightness correction factor, To control the coefficient of the proportion of the original brightness in the overall brightness adjustment, is the equilibrium constant, and Respectively represent the original brightness and chromaticity data of the LED display. and Respectively represent the brightness and chromaticity data after ambient light standardization, is the chromaticity adjustment coefficient, is the chroma adjustment gain coefficient, is a small constant, Represents the brightness and chromaticity correction coefficient matrix.

6. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: The step of inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction includes: The correction instruction includes a current magnitude adjustment value and a PWM magnitude adjustment value; The correction instruction is obtained by the following formula: ， ； in, is the current adjustment value, is the PWM size adjustment value, is the brightness and chromaticity correction coefficient matrix, is the original brightness and chromaticity data matrix, ,in, and Respectively represent the original brightness and chromaticity data of the LED display. and are the nonlinear correction coefficients for brightness and chromaticity, and are the brightness and chromaticity offsets, and They are the ambient light brightness correction factor and the ambient light chromaticity correction factor respectively.

7. The brightness and chromaticity correction method of an LED display screen according to claim 1, characterized in that: After sending the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen, the method further includes: Continuously monitor the ambient light sensor and the brightness and color sensor to obtain new ambient light data and currently displayed brightness and color data; When the new ambient light data changes by more than a preset stability threshold, the new ambient light data and the currently displayed brightness and chromaticity data are input into an ambient light depth compensation model and a display correction optimization model to generate a new correction instruction; When the new ambient light data change is less than the preset stability threshold, the current display state is maintained and no new correction instruction is generated.

8. A brightness and chromaticity correction system for an LED display screen, characterized in that: include: A data acquisition module is used to acquire ambient light data and original brightness and chromaticity data of the LED display screen; A data processing module, used for performing a preprocessing operation on the ambient light data to obtain ambient light standardized data; A coefficient generation module, used for inputting the original brightness and chromaticity data and the ambient light normalization data into a pre-trained ambient light depth compensation model to obtain a brightness and chromaticity correction coefficient matrix; An instruction generation module, used for inputting the brightness and chromaticity correction coefficient matrix and the original brightness and chromaticity data into a pre-trained display correction optimization model to obtain a correction instruction; The display correction module is used to send the correction instruction to the LED display screen to adjust the brightness and chromaticity of the LED display screen.

9. An electronic device, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the brightness and chromaticity correction method of the LED display screen as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the brightness and chromaticity correction method for the LED display screen as described in any one of claims 1 to 7.

Images