Vehicle-mounted display screen color calibration method and system based on ambient luminosity

Abstract

The invention discloses a vehicle-mounted display screen color calibration method and system based on ambient luminosity. The method comprises the following steps: acquiring ambient light characteristic data; performing filtering processing and dimension reduction processing on the environment light characteristic data, performing comparative analysis on the environment light characteristic data and a light characteristic database to obtain environment light classification and environment light characteristics, inputting the environment light classification and the environment light characteristics into a screen color parameter adjustment model, and outputting initial display parameters; environment light influence data is obtained through analysis according to the initial display parameters and the spectral distribution data, color parameters in the initial display parameters are corrected to obtain corrected display parameters, the corrected display parameters are applied to a screen display module, and adjusted screen display effect data are obtained; and comparing the screen display effect data with the visual consistency standard, and when the screen display effect data does not reach the visual consistency standard, re-correcting the color parameter to obtain a final display parameter, and applying the final display parameter to the screen display module. The method can improve the screen display effect.

Description

Technical Field

[0001] The present invention relates to the field of display control technologies, and particularly to a method and system for calibrating the color of an in-vehicle display screen based on ambient light intensity. Background Art

[0002] Currently, in the context of the rapid development of intelligent display technologies, in-vehicle display screens, as the core carriers of human-machine interaction, need to adapt to complex and changing driving environments. The dynamic relationship between the current ambient light and the screen color presentation has become a key difficulty in technological breakthroughs. The real-time fluctuations in ambient light characteristics (such as brightness, color temperature, spectral distribution) directly affect the user's perception of the screen color, and the interaction between light and display parameters is not a simple linear relationship, involving the coupled effects of multi-dimensional spectral components. Especially in the in-vehicle scenario, the lighting conditions may change frequently with the alternation of day and night, weather changes, or passing through tunnels, etc., making it difficult for traditional static calibration methods to maintain color consistency. Therefore, how to establish an accurate mapping between light characteristics and screen color parameters under changing ambient light conditions and achieve dynamic adjustment through quantifiable values has become a key issue in improving the display effect.

[0003] In an existing technology, the system collects ambient brightness data in real time through a single light sensor, and maps the original brightness value to a screen brightness adjustment parameter through a linear function. The specific operation process is as follows: First, the sensor obtains the ambient brightness value at a fixed sampling frequency (such as once per second), compares the collected original data with a preset brightness threshold range (such as 200 - 800 lux), and triggers the adjustment mechanism if it exceeds the threshold range. Subsequently, the system calls a static brightness mapping table, and according to the interval segment where the current brightness value is located (such as the low brightness segment, the medium brightness segment, the high brightness segment), directly outputs the corresponding brightness adjustment coefficient according to a preset linear proportional relationship (such as for every 100 lux increase in ambient brightness, the screen brightness increases by 10%). During the adjustment process, the system only performs a one-way adjustment on the supply voltage of the screen backlight module, and changes the LED backlight brightness through PWM dimming technology. This process does not introduce a spectral analysis module, nor does it perform collaborative calculations on other display parameters such as color temperature and contrast. Finally, the adjusted screen brightness parameter is solidified and output through a display driver chip until the next ambient brightness data triggers a new adjustment cycle.

[0004] The existing technology only relies on brightness data and unidirectionally adjusts screen parameters through a linear model, and cannot analyze multi-dimensional light characteristics such as spectra and color temperature. The static mapping table ignores the non-linear perception of the human eye and the spectral superposition effect, resulting in parameter adjustment lagging behind environmental changes and unable to achieve real-time adaptation of the screen display effect to ambient light changes. In summary, the existing technology has the problem of poor screen display effect. Summary of the Invention

[0005] The present invention provides a method and system for calibrating the color of an in-vehicle display based on ambient light intensity to improve the screen display effect.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a method for calibrating the color of an in-vehicle display based on ambient light intensity, including: Collecting ambient light characteristic data, where the ambient light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; Performing filtering processing and dimensionality reduction processing on the ambient light characteristic data, and comparing and analyzing the processed data with a pre-established light characteristic database to obtain ambient light classification and ambient light characteristics; Inputting the ambient light classification and the ambient light characteristics into a preset screen color parameter adjustment model to output initial display parameters; Analyzing the ambient light influence data based on the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the ambient light influence data to obtain corrected display parameters; Applying the corrected display parameters to a screen display module and obtaining adjusted screen display effect data; Comparing the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-correcting the color parameters to obtain final display parameters, and applying the final display parameters to the screen display module.

[0007] In an optional implementation manner, the performing filtering processing and dimensionality reduction processing on the ambient light characteristic data, and comparing and analyzing the processed data with a pre-established light characteristic database to obtain ambient light classification and ambient light characteristics includes: Performing filtering processing on the ambient light characteristic data by using a median filtering algorithm to obtain filtered data; Performing dimensionality reduction processing on the filtered data by using a principal component analysis method to obtain principal component characteristics; According to the principal component characteristics, combining the K-nearest neighbor classification algorithm to match with a pre-established light characteristic database to obtain ambient light classification and ambient light characteristics.

[0008] In an optional implementation manner, the process of establishing the light characteristic database includes: Obtaining historical ambient light characteristic data and corresponding historical ambient light classification and historical ambient light characteristics; Removing noise from the historical ambient light characteristic data by using a median filtering algorithm and extracting principal component characteristics by using a principal component analysis algorithm, and outputting a dimensionality-reduced feature dataset; The K-means clustering algorithm is used to cluster the feature data set, and the clustering quality is evaluated by the silhouette coefficient. After aligning the clustering result with the historical ambient light classification and the historical ambient light features, the link relationship data between the historical ambient light characteristics data, the historical ambient light classification, and the historical ambient light features is obtained; The link relationship data is stored in a relational database to obtain a light characteristic database.

[0009] In an alternative embodiment, the training process of the screen color parameter adjustment model includes: Obtain the historical ambient light classification and historical ambient light features, input them into the input layer of a preliminarily constructed neural network model for training, and obtain the predicted display parameters output by the output layer of the neural network model; Substitute the predicted display parameters and the pre-stored actual display parameters into the loss function to calculate the loss value; Calculate the gradient of the output of the output layer of the neural network model according to the loss value, and forward the gradient layer by layer through the chain rule to calculate the gradient of each layer of parameters to obtain gradient data; Update the parameters of each layer of the neural network model according to the gradient data and the preset learning rate; Iteratively update the parameters of each layer until when the training times of the neural network model are greater than the preset times, or when the loss value data of the neural network model is less than the preset loss threshold, it is determined that the training is completed, and a screen color parameter adjustment model is obtained.

[0010] In an alternative embodiment, the obtaining the environmental light influence data according to the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the environmental light influence data to obtain corrected display parameters includes: Use the non-negative matrix factorization algorithm to decompose the interaction between the spectral distribution data and the initial display parameters, and extract the weight coefficients of the spectral bands on the initial display parameters to obtain environmental light influence data; Use the CIEDE2000 algorithm to quantify the deviation value between the color parameters in the initial display parameters and the standard color. When the deviation value exceeds the preset deviation threshold, preliminarily correct the color parameters according to the environmental light influence data and the preset mapping table to obtain preliminarily corrected color parameters; Perform layer-by-layer calibration on the preliminarily corrected color parameters through the gamma correction algorithm to obtain corrected display parameters.

[0011] In an alternative embodiment, the applying the corrected display parameters to the screen display module and obtaining the adjusted screen display effect data includes: According to the calibrated display parameters, dynamically adjust the screen display module through the PID algorithm, and obtain the screen display parameters; Calculate the similarity between the screen display parameters and the pre-stored standard image data using the structural similarity index, and use the similarity, the screen display parameters, and the environmental light characteristic data as the screen display effect data.

[0012] In an optional implementation manner, comparing the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, recalibrate the color parameters to obtain the final display parameters, and apply the final display parameters to the screen display module, including: Classify the screen display effect data through the K-means clustering algorithm to obtain the display effect categories; Compare according to the display effect category and the screen display effect data with a preset visual consistency standard. When the display effect category and the screen display effect data reach the range of the visual consistency standard, there is no need to recalibrate the display parameters of the screen; When the display effect category and the screen display effect data do not reach the range of the visual consistency standard, re-use the non-negative matrix factorization algorithm to extract the weight coefficients of the spectral bands for the calibrated display parameters, recalibrate the display parameters of the screen to obtain the final display parameters, and apply them to the screen display module.

[0013] In a second aspect, the present invention provides a vehicle-mounted display color calibration system based on ambient light intensity, including: A data acquisition module for collecting environmental light characteristic data, where the environmental light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; A data processing module for filtering and dimensionality reduction processing of the environmental light characteristic data, and comparing and analyzing with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics; An initial parameter determination module for inputting the environmental light classification and the environmental light characteristics into a preset screen color parameter adjustment model and outputting initial display parameters; A calibration parameter module for obtaining environmental light influence data based on the initial display parameters and the spectral distribution data, and calibrating the color parameters in the initial display parameters according to the environmental light influence data to obtain calibrated display parameters; A display effect analysis module for applying the calibrated display parameters to the screen display module and obtaining the adjusted screen display effect data; An execution module is used to compare the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, the color parameters are recalibrated to obtain final display parameters, and the final display parameters are applied to the screen display module. In a third aspect, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for calibrating the color of a vehicle-mounted display screen based on ambient light intensity as described in any one of the above is implemented.

[0014] In a fourth aspect, the present invention also provides a computer-readable storage medium. The computer-readable storage medium includes a stored computer program. Wherein, when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for calibrating the color of a vehicle-mounted display screen based on ambient light intensity as described in any one of the above.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) By combining the median filtering algorithm with the principal component analysis and the K-nearest neighbor classification algorithm, the problems of computational redundancy and insufficient classification accuracy in the processing of high-dimensional light data by traditional methods are solved, the efficiency and accuracy of ambient light classification are improved, and the fast adaptation ability of the vehicle-mounted display screen in different lighting scenarios is ensured.

[0016] (2) Based on the neural network model, the model parameters are updated by backpropagation through the chain rule, overcoming the defect that the static parameter adjustment model cannot dynamically respond to the complex changes of ambient light, and realizing the real-time accurate prediction and dynamic calibration of the screen color parameters.

[0017] (3) The CIEDE2000 color difference algorithm is used to quantify the color deviation, and the gamma correction and PID control algorithms are combined to adjust the brightness of the RGB channels layer by layer, solving the problem of color temperature deviation caused by insufficient consideration of the spectral interaction effect in traditional brightness adjustment, and enhancing the stability and visual consistency of the displayed color in a dynamic environment.

[0018] (4) Through the structural similarity index and K-means clustering, and integrating the long-term trend analysis of the light characteristic database, the adaptive iterative correction of the screen display effect is realized, ensuring the long-term stability of the color parameters and the consistency of the user visual experience under complex lighting conditions. Description of the Drawings

[0019] Figure 1 is a schematic flowchart of the method for calibrating the color of a vehicle-mounted display screen based on ambient light intensity provided by the first embodiment of the present invention; Figure 2 is a schematic structural diagram of the system for calibrating the color of a vehicle-mounted display screen based on ambient light intensity provided by the second embodiment of the present invention. Specific implementation mode

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Refer to Figure 1 , the first embodiment of the present invention provides a method for calibrating the color of a vehicle-mounted display based on ambient light intensity, including the following steps: S11, collecting ambient light characteristic data, where the ambient light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; S12, performing filtering processing and dimensionality reduction processing on the ambient light characteristic data, and comparing and analyzing it with a pre-established light characteristic database to obtain ambient light classification and ambient light characteristics; S13, inputting the ambient light classification and the ambient light characteristics into a preset screen color parameter adjustment model to output initial display parameters; S14, analyzing the ambient light influence data based on the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the ambient light influence data to obtain corrected display parameters; S15, applying the corrected display parameters to the screen display module and obtaining adjusted screen display effect data; S16, comparing the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-correcting the color parameters to obtain final display parameters, and applying the final display parameters to the screen display module.

[0022] In step S11, ambient light characteristic data is collected, where the ambient light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data.

[0023] Specifically, the light data inside the cockpit is obtained in real time through an in-vehicle multi-channel optical sensor group, which includes a high-precision photometer, a spectral color temperature probe, and a micro spectrometer. The photometer captures the ambient light brightness at a preset sampling period (e.g., 100 milliseconds), with a measurement range covering the full dynamic range (e.g., the low illuminance to strong light interval). The light intensity signal is converted into a brightness value through an analog-to-digital converter. The spectral color temperature probe is based on a standard colorimetric system and uses a dual-light-path spectral splitting structure to synchronously measure the spectral power distribution in a specific wavelength range (e.g., the visible light band). The color temperature characteristic value represented by chromaticity coordinates is output through a built-in calculation module and a color temperature calibration value is generated through a conversion module. The micro spectrometer decomposes the incident light into multiple spectral channels (e.g., 256 channels) through a grating spectral splitting element. Each channel covers a fixed wavelength band width, and the photoelectric sensor array records the relative radiation intensity of each band to generate a spectral distribution vector of multi-dimensional values. During the data acquisition process, the sensor group packages the light brightness value, the color temperature calibration value, and the spectral distribution vector into a structured data frame through a bus protocol and transmits it to the main control unit. The light brightness threshold is set based on the human eye visual characteristic curve, and a threshold interval (e.g., a low threshold segment and a high threshold segment) is generated through algorithm analysis of the driver's pupil dynamic response data. The normalization process of the spectral distribution vector uses sensor calibration parameters to achieve spectral response correction by loading a reference reflectance curve. The structured light characteristic data set output in this step provides the original input for the subsequent filtering and dimensionality reduction module to ensure the complete capture of the multi-dimensional characteristics of the ambient light.

[0024] In step S12, the ambient light characteristic data is subjected to filtering processing and dimensionality reduction processing, and compared and analyzed with a pre-established light characteristic database to obtain the ambient light classification and ambient light characteristics.

[0025] In a specific embodiment, the process of subjecting the ambient light characteristic data to filtering processing and dimensionality reduction processing, and comparing and analyzing with a pre-established light characteristic database to obtain the ambient light classification and ambient light characteristics includes: The median filtering algorithm is used to perform filtering processing on the ambient light characteristic data to obtain filtered data; The principal component analysis method is used to perform dimensionality reduction processing on the filtered data to obtain principal component characteristics; Based on the principal component characteristics, combined with the K-nearest neighbor classification algorithm, it is matched with a pre-established light characteristic database to obtain the ambient light classification and ambient light characteristics.

[0026] In a specific embodiment, the process of establishing the light characteristic database includes: Obtain historical ambient light characteristic data and the corresponding historical ambient light classification and historical ambient light characteristics; Remove noise from the historical environmental light characteristic data through the median filtering algorithm and extract the principal component features through the principal component analysis algorithm, and output the feature dataset after dimensionality reduction; Use the K-means clustering algorithm to cluster the feature dataset, evaluate the clustering quality through the silhouette coefficient, and after aligning the clustering result with the historical environmental light classification and the historical environmental light features, obtain the link relationship data between the historical environmental light characteristic data, the historical environmental light classification, and the historical environmental light features; Store the link relationship data in a relational database to obtain a light characteristic database.

[0027] Specifically, input the collected environmental light characteristic data into the median filtering module, and use the sliding window mechanism to suppress noise for each data channel. The length of the filtering window is dynamically adjusted according to the sensor sampling frequency (for example, a short window is used for high-frequency sampling), and the pulse noise is eliminated by replacing the median value of the data within the window point by point, and the filtered brightness correction value, color temperature smoothing value, and spectral stationary vector are output. In the dimensionality reduction processing stage, the filtered data is constructed into a multi-dimensional matrix, the covariance matrix between each dimension is calculated, the eigenvectors are extracted through eigenvalue decomposition, and the first N principal components with a cumulative contribution rate exceeding a preset threshold (for example, the proportion of the cumulative variance of the principal components determined by algorithm iteration) are selected after sorting by eigenvalue size to generate a low-dimensional principal component feature vector.

[0028] During the construction of the light characteristic database, use the K-means clustering algorithm to cluster the principal component features after dimensionality reduction. First, select the initial clustering center points from the historical feature dataset through the maximum-minimum distance method to ensure that the initial centroids are as scattered as possible. Subsequently, iteratively calculate the Euclidean distance (i.e., the straight-line distance between two points in multi-dimensional space) from each data point to each centroid, and reassign the data point to the category of the centroid with the closest distance. This process continuously updates the centroid positions until the position changes of all centroids are less than a preset convergence threshold, that is, the centroids tend to be stable.

[0029] The clustering quality is evaluated through the silhouette coefficient, which reflects the tightness of the data points within the class and the separation degree between classes. For each data point, calculate the average distance to other points in the same class (intra-class tightness), and the average distance to all points in the nearest other class (inter-class separation degree). The silhouette coefficient is obtained by dividing the difference between these two distances by the larger of the two values, and the value range is between -1 and 1. When the average silhouette coefficient of all data points exceeds a preset threshold (such as the lowest acceptable value determined through historical data verification), the clustering result is determined to be valid.

[0030] After clustering is completed, map and align the generated clustering labels (such as "Category 1", "Category 2") with the existing environmental light classification labels (such as "strong sunlight", "tunnel lighting") in the historical data. The specific method is as follows: Analyze the distribution of historical classification labels in each clustering cluster, and use the historical label with the highest proportion as the final classification label for this clustering cluster. For example, if 70% of the data in a clustering cluster corresponds to the historical label "cloudy day", then map this cluster to the "cloudy day" type, form an index relationship table between the environmental light type code and the principal component feature vector, and finally store it in a relational database (such as SQLite database).

[0031] In the real-time comparison and analysis stage, calculate the similarity between the principal component feature vector obtained by real-time acquisition and dimensionality reduction and the historical feature set in the database. Here, the Euclidean distance is used as the similarity metric standard, calculate the distance value between the real-time feature vector and each historical feature vector in the database, and the smaller the distance, the higher the similarity. Traverse the database through the K-nearest neighbor algorithm, select the top K historical samples with the smallest distance (such as K = 5), count the majority voting results of the environmental light classification labels corresponding to these samples, and output the environmental light classification code with the highest matching degree (such as "tunnel lighting") and its associated spectral response feature vector.

[0032] This processing flow improves the data processing efficiency through noise suppression and dimensionality compression. At the same time, the classification system established based on historical data clustering enhances the accuracy of light feature matching, providing reliable environmental feature inputs for subsequent color parameter adjustment.

[0033] In step S13, input the environmental light classification and the environmental light features into a preset screen color parameter adjustment model to output initial display parameters.

[0034] In a specific implementation manner, the training process of the screen color parameter adjustment model includes: Obtain historical environmental light classification and historical environmental light features, input them into the input layer of a preliminarily constructed neural network model for training, and obtain the predicted display parameters output by the output layer of the neural network model; Substitute the predicted display parameters and the pre-stored actual display parameters into the loss function to calculate the loss value; Calculate the gradient output by the output layer of the neural network model according to the loss value, and pass the gradient forward layer by layer through the chain rule to calculate the gradient of each layer of parameters to obtain gradient data; Update the parameters of each layer of the neural network model according to the gradient data and the preset learning rate; Iteratively update the parameters of each layer repeatedly until when the training times of the neural network model are greater than the preset times, or when the loss value data of the neural network model is less than the preset loss threshold, it is determined that the training is completed, and the screen color parameter adjustment model is obtained.

[0035] Specifically, the specific implementation process of inputting the ambient light classification and features into the screen color parameter adjustment model is as follows: The input layer of the model receives the combined features formed by splicing the ambient light classification and the ambient light features, where the ambient light features include the low-dimensional feature components after principal component analysis. The neural network model adopts a multi-layer feedforward structure, the activation function of the hidden layer selects the rectified linear unit (e.g., ReLU), and the output layer uses the Sigmoid function to map the output value to a preset parameter interval. The historical data set in the training stage includes a display parameter matrix with labeled true values, which is measured by a professional color calibration device in a controlled lighting environment and includes the screen brightness calibration value, the RGB three-channel gain coefficients, and the gamma correction curve parameters. The loss function uses the mean square error to calculate the deviation between the model prediction value and the true value. By calculating the prediction errors of each node in the output layer, the partial derivatives of the weight matrix are calculated layer by layer using the backpropagation algorithm. The optimizer uses the adaptive moment estimation algorithm (e.g., Adam), and the initial learning rate is dynamically adjusted according to the convergence of the loss curve. When the loss decrease rate of consecutive iteration cycles is lower than the set threshold (e.g., the critical value of the loss change rate monitored by the early stopping method), the learning rate decay mechanism is triggered. The training termination condition is set as a double index of the maximum number of iterations and the validation set loss, where the loss threshold is determined by the error tolerance when selecting the optimal model through cross-validation. The trained model can output an initial display parameter set including the target brightness adjustment amount, the color gamut compensation coefficient, and the color space conversion matrix according to the real-time input light feature vector. This model captures the complex relationship between the light characteristics and the display parameters through a non-linear mapping relationship, overcomes the defect of insufficient modeling of the spectral superposition effect by traditional linear models, and provides high-precision initial values for subsequent parameter correction.

[0036] In step S14, the ambient light influence data is analyzed based on the initial display parameters and the spectral distribution data, and the color parameters in the initial display parameters are corrected according to the ambient light influence data to obtain the corrected display parameters.

[0037] In a specific implementation manner, the analyzing the ambient light influence data based on the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the ambient light influence data to obtain the corrected display parameters includes: The non-negative matrix factorization algorithm is used to decompose the interaction influence between the spectral distribution data and the initial display parameters, and the weight coefficients of the spectral bands on the initial display parameters are extracted to obtain the ambient light influence data; Quantify the deviation value between the color parameters in the initial display parameters and the standard color using the CIEDE2000 algorithm. When the deviation value exceeds the preset deviation threshold, perform a preliminary correction on the color parameters according to the environmental light influence data and the preset mapping table to obtain the preliminarily corrected color parameters; Perform layer-by-layer calibration on the preliminarily corrected color parameters through the gamma correction algorithm to obtain the corrected display parameters.

[0038] Specifically, the spectral distribution data is constructed into an association matrix composed of a multi-band intensity vector (such as the intensity values of spectral channels discretized within the visible light range) and the RGB channel gain values and brightness adjustment amounts in the initial display parameters. The non-negative matrix factorization algorithm is used to extract the basis vectors from this matrix. Through iterative optimization, the original matrix is decomposed into a basis matrix representing the weights of spectral bands and a coefficient matrix reflecting the influence of parameters, and the corrected weight coefficients of each spectral band on the RGB channels are output (such as the compensation factor of the red band for the blue channel). The environmental light influence data is stored in the form of a weight coefficient matrix, which is used to identify the interference intensity of specific spectral components on the display parameters. In the color deviation evaluation stage, the target color coordinates (including the lightness value "target lightness", chromaticity components "target chromaticity a" and "target chromaticity b") in the initial display parameters and the standard color gamut reference values (such as the "reference lightness", "reference chromaticity a" and "reference chromaticity b" of the sRGB color gamut reference point) are input into the color difference calculation module. First, calculate the lightness difference between the target and the reference, that is, subtract the "reference lightness" from the "target lightness" to obtain the lightness difference. Then, based on the chromaticity components of the target and the reference, calculate the chromaticity values of both: the target chromaticity value is the square root of the sum of the squares of "target chromaticity a" and "target chromaticity b", and the reference chromaticity value is the square root of the sum of the squares of "reference chromaticity a" and "reference chromaticity b". The difference between the two is the chromaticity difference. This chromaticity difference needs to be further corrected by the chromaticity adjustment factor to eliminate the non-linear influence of ambient light on chromaticity perception. For the calculation of the hue difference, first calculate the hue angles of both through the chromaticity components a and b of the target and the reference. If the difference in hue angles between the target and the reference exceeds 180 degrees, it is adjusted by adding 360 degrees to ensure that the angle difference is within a reasonable range. Subsequently, combined with the corrected chromaticity difference and the preset hue rotation parameter, the angle difference is non-linearly corrected to obtain the final hue difference. The calculation of the comprehensive deviation value needs to balance the sensitivity differences of the human eye to changes in lightness, chromaticity, and hue. Specifically, square the lightness difference after dividing it by the lightness weight coefficient, square the corrected chromaticity difference after dividing it by the chromaticity weight coefficient, square the corrected hue difference after dividing it by the hue weight coefficient, and then take the square root of the sum of these three items to obtain the comprehensive color difference value. The above-mentioned weight coefficients and rotation parameters are preset according to international standards to reflect the perception characteristics of the human eye to different color components. If the comprehensive color difference value exceeds the preset deviation threshold (such as the critical value set based on the experiment of the color difference perceptible by the human eye), it is determined that the current color parameters need to be corrected, and the subsequent color compensation process is triggered. In the preliminary correction stage, the preset color temperature-gain mapping table is called, and according to the compensation factors corresponding to the high-influence bands in the weight coefficient matrix, the gain coefficients of the RGB channels are adjusted proportionally (such as enhancing the compensation value of the red channel that is more affected by the blue light band).In the gamma correction stage, a segmented curve fitting method is adopted. The preliminarily corrected RGB values are divided into multiple brightness intervals (such as low brightness segment, medium brightness segment, high brightness segment) according to the characteristics of the display panel. Different gamma values are respectively applied to perform non-linear mapping on the gray levels of each interval to eliminate the distortion of the panel response curve caused by ambient light. The finally output corrected display parameters include optimized brightness control instructions, RGB gain coefficient matrix and gamma correction curve parameters to ensure the visual consistency of the displayed colors in a complex spectral environment. This correction process combines spectral weight analysis and color difference quantization to break through the limitation of traditional single brightness adjustment and achieve the collaborative optimization of multi-dimensional color parameters.

[0039] In step S15, the corrected display parameters are applied to the screen display module, and the adjusted screen display effect data is obtained.

[0040] In a specific implementation manner, the applying the corrected display parameters to the screen display module and obtaining the adjusted screen display effect data includes: According to the corrected display parameters, the screen display module is dynamically adjusted through the PID algorithm, and the screen display parameters are obtained; The structural similarity index is used to calculate the similarity between the screen display parameters and the pre-stored standard image data, and the similarity, the screen display parameters and the environmental light characteristic data are used as the screen display effect data.

[0041] Specifically, the target brightness value, RGB gain matrix, and gamma curve parameters in the calibration display parameters are written into the screen control register through the display driver interface. The PID control module receives the feedback value of the current screen actual brightness sensor, calculates the instantaneous error, error integral, and differential change rate with the target brightness value, and outputs a pulse width modulation signal to dynamically adjust the duty cycle of the backlight drive circuit, so that the screen brightness gradually converges to the target value. The screen display parameter real-time acquisition module synchronously captures the adjusted RGB channel voltage values, white balance coordinates, and the current gamma lookup table content to form a real-time display parameter vector. The structural similarity index calculation unit loads the pre-stored reference image data (e.g., digital copy of the international standard color card), decomposes the frame buffer data actually displayed on the screen and the reference image into multiple local windows, calculates the brightness mean, contrast covariance, and structural covariance of the pixels in each window respectively, and generates an overall similarity score through weighted fusion. The real-time spectral distribution vector in the ambient light characteristic data, the similarity score, and the real-time display parameter vector together constitute the screen display effect data set, which is stored as a multi-dimensional data record aligned with the time stamp. The proportional coefficient, integral time constant, and differential gain parameters in the PID control are tuned by the step response test method to ensure that the system response speed and stability meet the preset requirements. The similarity score threshold is set according to the human eye visual perception experiment. When the score is lower than the acceptable threshold, the parameter recalibration process is triggered. This step ensures the accurate implementation of the display parameters through closed-loop control, combines objective quality assessment and environmental data correlation analysis to form a complete display effect optimization verification chain.

[0042] In step S16, compare the screen display effect data with the preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-calibrate the color parameters to obtain the final display parameters, and apply the final display parameters to the screen display module.

[0043] In a specific implementation manner, the comparing the screen display effect data with the preset visual consistency standard, when the screen display effect data does not meet the visual consistency standard, re-calibrating the color parameters to obtain the final display parameters, and applying the final display parameters to the screen display module includes: Classify the screen display effect data through the K-means clustering algorithm to obtain the display effect category; Compare according to the display effect category and the screen display effect data with the preset visual consistency standard. When the display effect category and the screen display effect data are within the range of the visual consistency standard, there is no need to re-calibrate the display parameters of the screen; When the display effect category and the screen display effect data do not reach the range of the visual consistency standard, the non - negative matrix factorization algorithm is re - used to extract the weight coefficients of the spectral bands for correcting the display parameters, and the display parameters of the screen are corrected again to obtain the final display parameters, which are then applied to the screen display module.

[0044] Specifically, the screen display effect data set includes a structural similarity index scoring vector, a real - time RGB gain matrix, and an environmental spectral distribution vector. The visual consistency standard library stores reference data groups calibrated by experts (such as: the threshold interval of display parameters certified under standard lighting conditions), including the gamut coverage benchmark value, the allowable fluctuation range of brightness uniformity, and the chromatic aberration tolerance parameter. When initializing the K - means clustering algorithm, the maximum - minimum distance method is used to select the initial centroid points, and the category division is carried out by calculating the Euclidean distance from each data point to the centroid, and the centroid position is iteratively updated until the intra - class variance change rate is lower than the convergence threshold (such as: the critical value of the variance change rate adapted by the algorithm). The display effect category labels output by clustering (such as: qualified category, to - be - optimized category) are matched with the category mapping table in the standard library. When the median similarity score corresponding to the category label is lower than the visually perceptible threshold preset in the standard library (such as: the lower limit of visual acceptability determined by a large - scale subjective evaluation experiment), it is determined that the current display effect does not meet the standard.

[0045] In the re - calibration stage, the spectral distribution vector and the current display parameter matrix are called, and the non - negative matrix factorization algorithm is used for secondary decomposition. During the decomposition process, the display parameter basis matrix is fixed, and only the spectral weight matrix is updated to enhance the compensation coefficient of specific bands (such as: the bands sensitive to color temperature offset). The calibration parameter generation module adjusts the gain ratio of the RGB channels according to the updated weight matrix (such as: increasing the compensation value of the blue channel that is more affected by ambient red light), and combines the piece - wise slope correction of the gamma correction curve to generate the final display parameter set. The final parameters are subjected to a write - protection check by the check module of the display driver interface to ensure that the parameter values are within the hardware safety range (such as: the backlight voltage does not exceed the maximum rated value), and the screen control register is immediately refreshed after passing the check. This step realizes the iterative optimization of the display effect through a closed - loop verification mechanism, ensures the maintenance of color reproduction accuracy in a dynamic lighting environment, and its dual - calibration strategy effectively overcomes the problem of residual environmental interference that may exist in a single adjustment, achieving persistent stability of the visual consistency standard.

[0046] Refer to Figure 2 , the second embodiment of the present invention provides a vehicle - mounted display color calibration system based on ambient light intensity, including: A data acquisition module, configured to acquire ambient light characteristic data, where the ambient light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; A data processing module, configured to perform filtering processing and dimensionality reduction processing on the environmental light characteristic data, and perform comparison and analysis with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics; An initial parameter determination module, configured to input the environmental light classification and the environmental light characteristics into a preset screen color parameter adjustment model, and output initial display parameters; A correction parameter module, configured to analyze the environmental light influence data according to the initial display parameters and the spectral distribution data, and correct the color parameters in the initial display parameters according to the environmental light influence data to obtain corrected display parameters; A display effect analysis module, configured to apply the corrected display parameters to a screen display module, and obtain adjusted screen display effect data; An execution module, configured to compare the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-correct the color parameters to obtain final display parameters, and apply the final display parameters to the screen display module.

[0047] It should be noted that a vehicle-mounted display color calibration device based on ambient light intensity provided in an embodiment of the present invention is used to execute all process steps of a vehicle-mounted display color calibration method based on ambient light intensity in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated herein.

[0048] An 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 vehicle-mounted display color calibration program based on ambient light intensity. When the processor executes the computer program, the steps in each embodiment of the above vehicle-mounted display color calibration method based on ambient light intensity are implemented, such as Figure 1 Step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in each device embodiment above are implemented, such as a vehicle-mounted display color calibration module based on ambient light intensity.

[0049] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0050] The electronic device may be a computing device such as a desktop computer, notebook, handheld computer, and smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may also include input / output devices, network access devices, a bus, etc.

[0051] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device using various interfaces and lines.

[0052] The memory may be used to store the computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area may store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0053] Among them, if the modules / units integrated in the electronic device are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included 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, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0054] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached 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 can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative work.

[0055] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for calibrating the color of an in-vehicle display based on ambient light intensity, characterized in that, Including: Collecting environmental light characteristic data, where the environmental light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; Performing filtering processing and dimensionality reduction processing on the environmental light characteristic data, and comparing and analyzing it with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics; Inputting the environmental light classification and the environmental light characteristics into a preset screen color parameter adjustment model to output initial display parameters; Analyzing the environmental light influence data based on the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the environmental light influence data to obtain corrected display parameters; Applying the corrected display parameters to the screen display module and obtaining adjusted screen display effect data; Comparing the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-correcting the color parameters to obtain final display parameters, and applying the final display parameters to the screen display module.

2. The method for calibrating the color of an in-vehicle display based on ambient light intensity according to claim 1, wherein The performing filtering processing and dimensionality reduction processing on the environmental light characteristic data, and comparing and analyzing it with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics includes: Performing filtering processing on the environmental light characteristic data using a median filtering algorithm to obtain filtered data; Performing dimensionality reduction processing on the filtered data using a principal component analysis method to obtain principal component features; According to the principal component features, combining the K-nearest neighbor classification algorithm to match with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics.

3. The method for calibrating the color of a vehicle-mounted display based on ambient light intensity according to claim 1, wherein The establishment process of the light characteristic database includes: Obtaining historical environmental light characteristic data and corresponding historical environmental light classification and historical environmental light characteristics; Removing noise from the historical environmental light characteristic data through a median filtering algorithm and extracting principal component features through a principal component analysis algorithm, and outputting a dimensionality-reduced feature dataset; Using a K-means clustering algorithm to cluster the feature dataset, evaluating the clustering quality through a silhouette coefficient, and aligning the clustering result with the historical environmental light classification and the historical environmental light characteristics to obtain link relationship data of the historical environmental light characteristic data, the historical environmental light classification, and the historical environmental light characteristics; Storing the link relationship data in a relational database to obtain a light characteristic database.

4. The method for calibrating the color of an in-vehicle display screen based on ambient light intensity according to claim 1, characterized in that, The training process of the screen color parameter adjustment model includes: Obtaining historical environmental light classification and historical environmental light characteristics, inputting them into the input layer of a preliminarily constructed neural network model for training, and obtaining predicted display parameters output by the output layer of the neural network model; Substituting the predicted display parameters and pre-stored actual display parameters into a loss function to calculate a loss value; Calculating the gradient output by the output layer of the neural network model according to the loss value, and propagating the gradient layer by layer forward through the chain rule to calculate the gradient of each layer of parameters to obtain gradient data; Updating the parameters of each layer of the neural network model according to the gradient data and a preset learning rate. Iteratively update the parameters of each layer until it is determined that the training is completed and a screen color parameter adjustment model is obtained when the number of training times of the neural network model is greater than a preset number of times, or when the loss value data of the neural network model is less than a preset loss threshold.

5. The method for calibrating the color of an in-vehicle display based on ambient light intensity according to claim 1, characterized in that, The method for analyzing the environmental light influence data based on the initial display parameters and the spectral distribution data, and correcting the color parameters in the initial display parameters according to the environmental light influence data to obtain corrected display parameters includes: Using the non-negative matrix factorization algorithm to decompose the interaction influence between the spectral distribution data and the initial display parameters, and extracting the weight coefficients of the spectral bands on the initial display parameters to obtain environmental light influence data; Using the CIEDE2000 algorithm to quantify the deviation value between the color parameters in the initial display parameters and the standard color. When the deviation value exceeds a preset deviation threshold, the color parameters are preliminarily corrected according to the environmental light influence data and a preset mapping table to obtain preliminarily corrected color parameters; According to the preliminarily corrected color parameters, layer-by-layer calibration is performed through the gamma correction algorithm to obtain corrected display parameters.

6. The method for calibrating the color of an in-vehicle display based on ambient light intensity according to claim 1, wherein The method for applying the corrected display parameters to the screen display module and obtaining the adjusted screen display effect data includes: Dynamically adjusting the screen display module according to the corrected display parameters through the PID algorithm and obtaining the screen display parameters; Calculating the similarity between the screen display parameters and the pre-stored standard image data using the structural similarity index, and using the similarity, the screen display parameters, and the environmental light characteristic data as the screen display effect data.

7. The method for calibrating the color of an in-vehicle display based on ambient light according to claim 1, wherein The method for comparing the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, re-correcting the color parameters to obtain the final display parameters, and applying the final display parameters to the screen display module includes: Classifying the screen display effect data through the K-means clustering algorithm to obtain the display effect category; Comparing the display effect category and the screen display effect data with a preset visual consistency standard. When the display effect category and the screen display effect data are within the range of the visual consistency standard, there is no need to re-correct the display parameters of the screen; When the display effect category and the screen display effect data are not within the range of the visual consistency standard, re-using the non-negative matrix factorization algorithm to extract the weight coefficients of the spectral bands on the corrected display parameters, re-correcting the display parameters of the screen to obtain the final display parameters, and applying them to the screen display module.

8. An in-vehicle display color calibration system based on ambient light intensity, characterized in that, Including: A data acquisition module for acquiring environmental light characteristic data, where the environmental light characteristic data includes light brightness, ambient light color temperature, and spectral distribution data; A data processing module for performing filtering processing and dimensionality reduction processing on the environmental light characteristic data, and comparing and analyzing it with a pre-established light characteristic database to obtain environmental light classification and environmental light characteristics; An initial parameter determination module, configured to input the classified ambient light and the ambient light characteristics into a preset screen color parameter adjustment model, and output initial display parameters; A calibration parameter module, configured to analyze ambient light influence data based on the initial display parameters and the spectral distribution data, and calibrate the color parameters in the initial display parameters according to the ambient light influence data to obtain calibrated display parameters; A display effect analysis module, configured to apply the calibrated display parameters to a screen display module and obtain adjusted screen display effect data; An execution module, configured to compare the screen display effect data with a preset visual consistency standard. When the screen display effect data does not meet the visual consistency standard, recalibrate the color parameters to obtain final display parameters, and apply the final display parameters to the screen display module.

9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for calibrating the color of an in-vehicle display screen based on ambient light intensity as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for calibrating the color of an in-vehicle display screen based on ambient light intensity as described in any one of claims 1 to 7.

Images