A color adjustment method and system of a display screen based on environmental information

Abstract

The present application relates to the technical field of display screen color adjustment, and discloses a display screen color adjustment method and system based on environmental information, the method comprising obtaining display content data, environmental color data, environmental light data and display screen type of the display screen; calculating color perception data according to the environmental color data and the display screen type, and performing data processing thereon to obtain chroma feedback data; establishing the relationship between the environmental light data and user preference by a support vector machine method to obtain user evaluation data; inputting the chroma feedback data and the user evaluation data into a color correction model to obtain color adjustment data; classifying the display content data and calculating the type weight index by a class Cai bird optimization algorithm to obtain the weight index; and generating color adjustment instructions according to the color adjustment data and the weight index to adaptively adjust the display content. The method solves the problems of high light source environment requirement and strong subjectivity of display screen color adjustment.

Description

Technical Field

[0001] This invention relates to the field of display screen color adjustment technology, and in particular to a color adjustment method and system for a display screen based on environmental information. Background Technology

[0002] With continuous technological advancements and rising living standards, consumers' demands for electronic devices such as mobile phones, tablets, televisions, and monitors are gradually increasing. The screens of these devices are evolving towards larger sizes, thinner designs, higher brightness, finer resolution, and more vivid colors to meet users' growing needs for visual experience. However, the development of display technology faces numerous challenges. When ambient light conditions change, screen colors can exhibit varying degrees of distortion, leading to unstable display effects and impacting user experience. This color difference phenomenon is not only affected by changes in external lighting conditions but also limited by the material properties and light-emitting principles of the display itself. The screen's color performance will differ under different environments, directly affecting the quality of the viewing experience. Therefore, to ensure a consistent and high-quality visual experience under various lighting conditions, researching and developing advanced color calibration algorithms and technologies is particularly important.

[0003] Traditional methods for adjusting display colors first utilize light sensors to acquire ambient brightness and color temperature data. Based on this data, the ambient reflectance is calculated. Next, an ambient correction matrix is ​​generated based on the calculated reflectance, determining the appropriate adjustment parameters for each color under the current environment. This matrix is ​​then applied to the display's color management system, adjusting the gain values ​​corresponding to each color coordinate to compensate for the effects of ambient light, ensuring accurate color reproduction under various lighting conditions. Finally, the final color output parameters are adjusted according to the viewing distance to ensure the display correctly shows colors under specific ambient lighting conditions.

[0004] However, traditional color adjustment methods rely on precise ambient light measurements and human judgment, and are primarily applicable to standard RGB lighting environments. This necessitates reconstructing the environmental reflectivity model under non-standard lighting conditions, increasing complexity and error. Therefore, existing technologies suffer from high requirements for the lighting environment and strong subjectivity in display color adjustment. Summary of the Invention

[0005] This invention provides a color adjustment method and system for a display screen based on environmental information, in order to solve the problems of high requirements for the light source environment and strong subjectivity in display screen color adjustment.

[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a color adjustment method for a display screen based on environmental information, comprising:

[0007] Acquire display content data, display environment data, and display type of the display screen, where the display environment data includes ambient color data and ambient light data;

[0008] Calculate color perception data based on the ambient color data and the display screen type;

[0009] The color perception data is quantized, smoothed, clustered, and subjected to multinomial regression analysis to obtain color feedback data.

[0010] The relationship between the ambient light data and user preferences is established using the support vector machine method to obtain user evaluation data;

[0011] The chromaticity feedback data and the user evaluation data are input into a pre-trained color correction model to obtain color adjustment data;

[0012] The displayed content data is classified using the Kingfisher optimization algorithm, and a type weight index is calculated based on the classification results and the user evaluation data to obtain the weight index.

[0013] Color adjustment instructions are generated based on the color adjustment data and the weight index, and the displayed content is adaptively adjusted according to the color adjustment instructions.

[0014] In one optional implementation, calculating color perception data based on the ambient color data and the display screen type includes:

[0015] Perform gamma correction on the environmental color data to obtain corrected color data;

[0016] Select the elements in the corresponding independent space transformation matrix according to the display screen type;

[0017] The corrected color data is then subjected to independent spatial transformation to obtain independent color data;

[0018] The independent color data is converted to a uniform color space to obtain color perception data.

[0019] Gamma correction is performed using the following formula:

[0020] C′=C γ

[0021] Independent space transformations are performed using the following formula:

[0022]

[0023] The following formula is used to perform a perceptual uniform color space conversion:

[0024] L *=116f(Y / Y) n )-16

[0025] a * =500[f(X / X) n )-f(Y / Y n )]

[0026] b * =200[f(Y / Y) n )-f(Z / Z n )]

[0027]

[0028] Where C represents ambient color data, γ represents the preset gamma value, C′ represents the corrected color data, R represents the red data in the corrected color data, G represents the green data in the corrected color data, B represents the blue data in the corrected color data, and m 11 m 12 m 13 m 21 m 22 m 23 m 31 m 32 m 33 This represents the elements in the preset independent space transformation matrix, where X, Y, and Z represent independent color data. n Y n Z n L represents the tristimulus value of the preset reference white point. * a * b * represents color perception data, and t represents the result of dividing the value of the substituted independent color data by its corresponding reference white point.

[0029] In one optional implementation, the step of quantizing, smoothing, clustering, and performing multinomial regression analysis on the color perception data to obtain chromaticity feedback data includes:

[0030] The color perception data is quantized using a uniform quantization method to obtain quantized data.

[0031] The quantized data is smoothed by using a Gaussian filter to obtain smoothed data.

[0032] The smoothed data was clustered using the K-means clustering algorithm to obtain clustered analysis data;

[0033] The clustering data was analyzed using multinomial regression analysis to obtain color feedback data.

[0034] Quantification is performed using the following formula:

[0035]

[0036] The Gaussian filter is shown below:

[0037]

[0038] Where F represents L in color perception data * or a * or b * One of the data points, F min F represents the minimum value in the color perception data. max The value of p represents the maximum value in the color perception data, N represents the total number of color perception data points, and p represents the quantized data. i+j Let p' represent the (i+j)th quantized data, σ and w represent the preset filter parameters, and p'' i This indicates smoothed data.

[0039] In one optional implementation, the step of establishing the relationship between the ambient light data and user preferences using a support vector machine method to obtain user evaluation data includes:

[0040] The ambient light data is standardized to obtain standardized data;

[0041] The standardized data is input into a pre-trained support vector machine model to calculate a decision function, and the output of the decision function is used as the decision data.

[0042] The decision data is mapped to a pre-stored evaluation interval to obtain user evaluation data;

[0043] The decision function is as follows:

[0044]

[0045] Where, f(e) new ) represents the decision function, and SV represents the number of pre-stored historical ambient light data samples. b represents the optimal solution of the Lagrange multipliers after training is complete. * The term y represents the paranoia associated with completed training. i K(e) represents the weight coefficients after training is complete. i e new ) represents the preset kernel function, e i Represents the i-th data point in the historical ambient light data, e new This represents ambient light data.

[0046] In one optional implementation, the training process of the color correction model includes:

[0047] The pre-stored historical color feedback data and pre-stored historical user evaluation data are input into the input layer of the initial color correction model for training, and the predicted value data output by the output layer of the initial color correction model is obtained.

[0048] Substitute the predicted value data and the pre-stored actual target value into the loss function to calculate the loss value data.

[0049] The gradient of the output layer of the initial color correction model is calculated based on the loss value data, and the gradient is passed forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer and obtain the gradient data.

[0050] Based on the gradient data and the preset learning rate, update the parameters of each layer of the initial color correction model;

[0051] The parameters of each layer are iteratively updated until the number of training iterations of the initial color correction model is greater than a preset number of iterations, or the loss value of the initial color correction model is less than a preset loss threshold. At this point, the training is considered complete, and the color correction model is obtained.

[0052] In one optional implementation, the step of classifying the displayed content data using the kingfisher optimization algorithm and calculating a type weight index based on the classification results and the user review data to obtain the weight index includes:

[0053] The display content data is classified using the kingfisher optimization algorithm to obtain the display content type.

[0054] The displayed content type and the user evaluation data are input into a pre-trained content type weight adjustment model to obtain a weight index.

[0055] In one optional implementation, the step of generating a color adjustment instruction based on the color perception data, the color adjustment data, and the weighting index, and adaptively adjusting the displayed content according to the color adjustment instruction, includes:

[0056] Calculate the color difference between the color adjustment data to obtain color difference data;

[0057] Calculate the weighting coefficient of the display content type based on the color difference data;

[0058] Generate color adjustment instructions based on the weighting coefficients and the weighting index;

[0059] The display screen is adjusted according to the color adjustment instructions and the color adjustment data, including gamma value adjustment, white balance adjustment, RGB gain adjustment, and color adjustment.

[0060] The color difference data is calculated using the following formula:

[0061]

[0062] The weighting coefficients are calculated using the following formula:

[0063]

[0064] Generate color adjustment instructions using the following formula:

[0065]

[0066] Where, ΔL * ik , Δa * ij Δb * ik ΔE represents the difference between the i-th and j-th color perception data points. ij w represents the color difference between the i-th color perception data and the j-th color perception data. ij W represents the weighting coefficient. i The weighting index is represented by k, the preset slope is represented by θ, and the preset threshold parameter is represented by A. i This indicates a color adjustment command.

[0067] Secondly, the present invention provides a color adjustment system for a display screen based on environmental information, comprising:

[0068] The data acquisition module is used to acquire the display content data, display environment data, and display type of the display screen, wherein the display environment data includes ambient color data and ambient light data;

[0069] A color perception calculation module is used to calculate color perception data based on the ambient color data and the display screen type.

[0070] The chromaticity feedback analysis module is used to perform quantization, smoothing adjustment, cluster analysis, and multinomial regression analysis on the color perception data to obtain chromaticity feedback data.

[0071] The user evaluation analysis module is used to establish the relationship between the ambient light data and user preferences using the support vector machine method to obtain user evaluation data.

[0072] The color correction module is used to input the chromaticity feedback data and the user evaluation data into a pre-trained color correction model to obtain color adjustment data;

[0073] The weight index calculation module is used to classify the displayed content data using the kingfisher optimization algorithm, and calculate the type weight index based on the classification results and the user evaluation data to obtain the weight index;

[0074] The color adjustment module is used to generate color adjustment instructions based on the color adjustment data and the weight index, and to adaptively adjust the displayed content according to the color adjustment instructions.

[0075] Thirdly, 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, wherein the processor, when executing the computer program, implements the color adjustment method for a display screen based on environmental information as described in any one of the preceding claims.

[0076] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the color adjustment method for a display screen based on environmental information as described above.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] This invention discloses a color adjustment method for a display screen based on environmental information. The method includes acquiring display content data, display environment data, and display screen type, wherein the display environment data includes ambient color data and ambient light data; calculating color perception data based on the ambient color data and the display screen type; performing quantization, smoothing adjustment, cluster analysis, and multinomial regression analysis on the color perception data to obtain chromaticity feedback data; establishing the relationship between the ambient light data and user preferences using a support vector machine method to obtain user evaluation data; inputting the chromaticity feedback data and the user evaluation data into a pre-trained color correction model to obtain color adjustment data; classifying the display content data using a kingfisher optimization algorithm, and calculating a type weight index based on the classification results and the user evaluation data to obtain a weight index; generating color adjustment instructions based on the color adjustment data and the weight index, and adaptively adjusting the display content according to the color adjustment instructions. This method acquires display content data, display environment data, and display screen type, calculates color perception data, and obtains chromaticity feedback data through analysis. Simultaneously, it uses a support vector machine to correlate ambient light with user preferences to derive user evaluation data. These data are input into the color correction model to generate color adjustment data. The kingfisher optimization algorithm is then used to classify the displayed content and calculate weight indices. Finally, based on the adjustment scheme and weight generation instructions, adaptive adjustment of the displayed content is achieved. This method reduces the requirement for specific light sources, lowers subjectivity, and improves the consistency and accuracy of color adjustment. Attached Figure Description

[0079] Figure 1 This is a schematic flowchart of a color adjustment method for a display screen based on environmental information provided in the first embodiment of the present invention;

[0080] Figure 2 This is a schematic diagram of a color adjustment system for a display screen based on environmental information, provided in the second embodiment of the present invention. Detailed Implementation

[0081] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0082] Reference Figure 1 The first embodiment of the present invention provides a color adjustment method for a display screen based on environmental information, comprising the following steps:

[0083] S11, Obtain display content data, display environment data, and display type of the display screen, wherein the display environment data includes ambient color data and ambient light data;

[0084] S12, calculate color perception data based on the ambient color data and the display screen type;

[0085] S13, perform quantization, smoothing adjustment, cluster analysis and multinomial regression analysis on the color perception data to obtain color feedback data;

[0086] S14, The relationship between the ambient light data and user preferences is established using the support vector machine method to obtain user evaluation data;

[0087] S15, input the chromaticity feedback data and the user evaluation data into the pre-trained color correction model to obtain color adjustment data;

[0088] S16, the displayed content data is classified using the Kingfisher optimization algorithm, and the type weight index is calculated based on the classification results and the user evaluation data to obtain the weight index;

[0089] S17, Generate a color adjustment instruction based on the color adjustment data and the weight index, and adaptively adjust the displayed content according to the color adjustment instruction.

[0090] In step S11, the display content data, display environment data, and display type of the display screen are obtained, wherein the display environment data includes ambient color data and ambient light data.

[0091] Display content data consists of the information currently presented on the screen, including but not limited to the color, brightness, and contrast characteristics of images, videos, and text elements. Display environment data is collected through integrated or external sensors that sense the color and light intensity of the surrounding environment and convert these physical quantities into digital signals. Ambient color data reflects the color distribution in the environment, while ambient light data records the intensity and color temperature characteristics of ambient light. Display type information can be directly obtained from product data, including the display's technical parameters and physical properties, such as screen size, resolution, panel type (e.g., LCD, OLED), and light-emitting principle.

[0092] In step S12, color perception data is calculated based on the ambient color data and the display screen type.

[0093] In one specific implementation, calculating color perception data based on the ambient color data and the display screen type includes:

[0094] Perform gamma correction on the environmental color data to obtain corrected color data;

[0095] Select the elements in the corresponding independent space transformation matrix according to the display screen type;

[0096] The corrected color data is then subjected to independent spatial transformation to obtain independent color data;

[0097] The independent color data is converted to a uniform color space to obtain color perception data.

[0098] Gamma correction is performed using the following formula:

[0099] C′=C γ

[0100] Independent space transformations are performed using the following formula:

[0101]

[0102] The following formula is used to perform a perceptual uniform color space conversion:

[0103] L * =116f(Y / Y) n )-16

[0104] a * =500[f(X / X) n )-f(Y / Y n )]

[0105] b * =200[f(Y / Y) n )-f(Z / Z n )]

[0106]

[0107] Where C represents ambient color data, γ represents the preset gamma value, C′ represents the corrected color data, R represents the red data in the corrected color data, G represents the green data in the corrected color data, B represents the blue data in the corrected color data, and m 11 m 12 m 13 m 21 m 22 m 23 m 31 m 32 m 33 This represents the elements in the preset independent space transformation matrix, where X, Y, and Z represent independent color data. n Y n Z nL represents the tristimulus value of the preset reference white point. * a * b * This represents color perception data, where t represents the value of the input independent color data (X, Y, Z) divided by its corresponding reference white point (X). n Y n Z n The result after that.

[0108] Specifically, firstly, gamma correction is applied to the ambient color data to address the non-linear relationship between display output and human visual perception, ensuring color consistency and accuracy across different display devices. Gamma correction adjusts the brightness distribution of the image by applying a power function, making midtones brighter and closer to the actual perception of the human eye. Specifically, for a given ambient color data C, its gamma-corrected value C′ can be expressed by the formula C′=C γ The calculation shows that γ is a preset gamma value, which is a parameter of a non-linear transformation and is related to the display type.

[0109] Next, based on the obtained display type information, select an element from an independent color space conversion matrix that matches it. Each display, due to its unique physical properties and technical parameters (e.g., LCD or OLED panel), requires a specific color space conversion matrix for color conversion. This matrix consists of multiple preset coefficients used to convert the red (R), green (G), and blue (B) data in the RGB color model to their corresponding values ​​in the XYZ color space. The specific conversion formula is as follows:

[0110]

[0111] Where X, Y, and Z represent independent color data, m 11 m 12 m 13 m 21 m 22 m 23 m 31 m 32 m 33 This represents the element selected from a predefined spatial transformation matrix.

[0112] Subsequently, to further process the obtained independent color data, it is converted to a perceptually uniform color space, namely the CIELAB color space. This conversion aims to create a color representation that better conforms to the human visual system, where similarly located color points appear closer together. The conversion formula is as follows:

[0113] L * =116f(Y / Y1)-16

[0114] a * =500[f(X / X) n )-f(Y / Y n )]

[0115] b * =200[f(Y / Y) n )-f(Z / Z n )]

[0116] Among them, X n Y n Z n L represents the tristimulus value of the preset reference white point. * a * b * Represents color perception data; the function f(t) is defined as:

[0117]

[0118] In the formula, t represents the value of the substituted independent color data (X, Y, Z) divided by its corresponding reference white point (X). n Y n Z n The result after performing the above series of transformations on the color data. The final color perception data (L) is obtained. * a * b * It can better reflect the true perception of color by the human visual system, thereby enabling accurate calculation of color perception data.

[0119] The function f(t) performs a non-linear mapping on independent color data to more accurately reflect changes in brightness and chromaticity in human visual perception. When the value of t is less than or equal to a threshold... The linear component is used to avoid large errors in low-brightness areas, while when the t value is greater than the threshold... When the brightness changes, the cube root operation is used to approximate the human eye's response to changes in brightness.

[0120] Substituting the function f(t) into the perceptual uniform color space conversion formula yields the following specific formula:

[0121]

[0122] In step S13, the color perception data is quantized, smoothed, clustered, and subjected to multinomial regression analysis to obtain color feedback data.

[0123] In one specific implementation, the step of quantizing, smoothing, clustering, and performing multinomial regression analysis on the color perception data to obtain chromaticity feedback data includes:

[0124] The color perception data is quantized using a uniform quantization method to obtain quantized data.

[0125] The quantized data is smoothed by using a Gaussian filter to obtain smoothed data.

[0126] The smoothed data was clustered using the K-means clustering algorithm to obtain clustered analysis data;

[0127] The clustering data was analyzed using multinomial regression analysis to obtain color feedback data.

[0128] Quantification is performed using the following formula:

[0129]

[0130] The Gaussian filter is shown below:

[0131]

[0132] Where F represents L in color perception data * or a * or b * One of the data points, F min F represents the minimum value in the color perception data. max The value of p represents the maximum value in the color perception data, N represents the total number of color perception data points, and p represents the quantized data. i+j Let p' represent the (i+j)th quantized data, σ and w represent the preset filter parameters, and p'' i This indicates smoothed data.

[0133] First, the color perception data is quantized using a uniform quantization method. This process aims to reduce the complexity of color representation while maintaining sufficient accuracy to capture key color features. Specifically, for a given set of color perception data, its quantized value can be expressed using the formula... The calculation yields the result. This formula maps the raw color perception data to a discrete numerical range, thereby achieving effective compression of color information.

[0134] Next, to further improve the quality of the quantized data and reduce the impact of noise, a Gaussian filter is used to smooth the quantized data. A Gaussian filter is a linear low-pass filter that effectively suppresses noise that follows a normal distribution while preserving the main features of the image. For each quantized data point, its smoothed value can be calculated using the following formula:

[0135]

[0136] Where σ and w are preset filter parameters that determine the size and shape of the filter window; p i+j Let represent the (i+j)th quantized data point. This formula achieves a smoothing effect by applying a weighted average to each data point and its neighborhood. The weights are determined by a Gaussian function, with the center point having the largest weight, which gradually decreases as the distance increases.

[0137] Next, in the process of clustering smoothed data using the K-means clustering algorithm, the first step is to determine the number of clusters K to be formed. The choice of this parameter can be determined based on the specific application scenario or through techniques such as the Elbow method. Once K is determined, the algorithm will randomly select K data points as the initial cluster centers.

[0138] Next, the algorithm enters the iterative optimization phase, the core steps of which include:

[0139] Sample allocation: Calculate p for each data point i Calculate the distance to each cluster center and assign it to the cluster to which the nearest cluster center belongs. The distance metric used here is typically Euclidean distance, but other metrics can be chosen depending on the specific circumstances.

[0140] Update cluster center: For each cluster, recalculate the mean of all data points within that cluster, and use this as the new cluster center location. This means that the cluster center will gradually "move" towards the data points within the cluster to better represent the data distribution characteristics of that cluster.

[0141] Check convergence conditions: Repeat the above two steps until a specific termination condition is met. Common termination conditions include no longer significant changes in cluster centers, reaching a preset maximum number of iterations, or changes in cluster members being less than a preset threshold.

[0142] After clustering is completed, the results provide structured input for subsequent analysis. To further understand the relationships between different clusters and the trends in color variation within clusters, multinomial regression analysis can be applied based on the clustering results. Multinomial regression models allow for the capture of more complex nonlinear relationships between independent and dependent variables, and are particularly suitable for describing data patterns that exhibit curved rather than linear trends.

[0143] The specific formation process of the polynomial model:

[0144] Suppose x is one or more independent variables (e.g., brightness, hue, etc.), and y is the corresponding dependent variable. To build a multinomial regression model, the order n of the polynomial must first be determined, which depends on the observed data trends and the desired fitting accuracy. While a higher order can provide a better fit, it can also lead to overfitting, where the model too closely follows noise or outliers in the training data.

[0145] Once the order n of the polynomial is determined, the general form of the polynomial regression model can be defined as follows:

[0146] y = β0 + β1x + β2x 2 +...+β n x n +∈

[0147] Where, β k (k = 0, 1, ..., n) are the coefficients to be estimated, and ∈ represents the error term, reflecting the data variability that the model fails to explain. A common method for solving these unknown coefficients is the least squares method, which attempts to minimize the predicted values. The sum of squared differences between the observed value y and the actual observed value y. Specifically, the residual sum of squares (RSS) can be calculated using the following formula:

[0148]

[0149] Where m is the sample size. The coefficient β to be estimated is adjusted... k The value of minimizes RSS, thus yielding an optimal set of polynomial coefficients.

[0150] In this way, the multinomial regression model can not only reveal the color change patterns within clusters, but also provide a scientific basis for dynamically adjusting the content displayed on the screen, thereby ensuring a consistent and realistic visual experience in various environments.

[0151] In step S14, the relationship between the ambient light data and user preferences is established using the support vector machine method to obtain user evaluation data.

[0152] In one specific implementation, establishing the relationship between the ambient light data and user preferences using a support vector machine method to obtain user evaluation data includes:

[0153] The ambient light data is standardized to obtain standardized data;

[0154] The standardized data is input into a pre-trained support vector machine model to calculate a decision function, and the output of the decision function is used as the decision data.

[0155] The decision data is mapped to a pre-stored evaluation interval to obtain user evaluation data;

[0156] The decision function is as follows:

[0157]

[0158] Where, f(e)new ) represents the decision function, and SV represents the number of pre-stored historical ambient light data samples. b represents the optimal solution of the Lagrange multipliers after training is complete. * The term y represents the paranoia associated with completed training. i K(e) represents the weight coefficients after training is complete. i e new ) represents the preset kernel function, e i Represents the i-th data point in the historical ambient light data, e new This represents ambient light data. The output of the decision function is the decision data.

[0159] Specifically, the first step is to preprocess the raw ambient light data to ensure it is suitable for input into a pre-trained support vector machine model for computation. In this stage, the task is to perform standardization on the ambient light data to obtain standardized data. Standardization converts data of different dimensions into dimensionless pure numerical values ​​for easier comparison and weighting. The z-score standardization method is commonly used, which ensures that the processed data conforms to a standard normal distribution, i.e., a mean of 0 and a standard deviation of 1.

[0160] After data standardization, the next step is to input the standardized ambient light data as new samples into a pre-trained support vector machine (SVM) model to calculate the decision function. The core of SVM lies in finding the maximum margin hyperplane. For non-linearly separable data, a kernel function maps the data to a higher-dimensional space, making it linearly separable. During this process, the model constructs the optimal classification boundary based on the support vectors in the training set and determines the Lagrange multipliers. and bias term b * And the category label y corresponding to each support vector. i For the new ambient light data e new Decision function f(e) new The expression for ) is as follows:

[0161]

[0162] Once the decision function f(e) is obtained newThe next step after processing the output of the decision function, i.e., the decision data, is to map this decision data to a pre-stored evaluation interval to obtain the final user evaluation data. This step involves converting continuous decision values ​​into discrete evaluation levels or scores, ensuring that the output has intuitive meaning and directly reflects the user's preference level. To achieve this, a mapping rule can be defined, such as linear mapping or piecewise linear mapping, and the corresponding conversion formula can be set according to the output range of the decision function and the expected evaluation interval. If the evaluation interval is set to [0,1] or [-1,1], the decision values ​​can be adapted to this interval by adjusting the mapping parameters. In addition, a threshold segmentation method can be considered to divide the decision values ​​into several intervals, each interval corresponding to a different evaluation level, thereby realizing the transformation from continuous values ​​to discrete evaluations.

[0163] The training process of the Support Vector Machine (SVM) model is an iterative optimization process aimed at finding the hyperplane that best distinguishes different classes of data. To achieve this, the SVM algorithm needs to solve a convex quadratic programming problem, which aims to maximize the classification margin while minimizing the classification error. The following are the specific steps involved in training the SVM model in establishing the relationship between ambient light data and user preferences.

[0164] First, the dataset must be prepared and preprocessed before training begins. This includes collecting ambient light data and standardizing or normalizing it to ensure all features have similar numerical ranges, thus avoiding some features from having an undue impact on the model due to their large magnitude. Furthermore, the dataset needs to be divided into training and test sets to evaluate model performance after training.

[0165] Next, choosing an appropriate kernel function is a crucial step. For linearly separable datasets, a simple linear kernel can be chosen; while for non-linearly separable datasets, multinomial kernels, radial basis function (RBF) kernels, or other types of kernel functions can be used to elevate the data to a higher-dimensional space, making it linearly separable. The choice of kernel function directly affects the model's capabilities and generalization performance.

[0166] Then, we move on to the core training phase. In this phase, SVM attempts to find the optimal hyperplane by solving an objective function of the following form:

[0167]

[0168] Where w is the weight vector, b is the bias term, C is the regularization parameter, which controls the degree of penalty for misclassified samples, and ξ iis a slack variable, allowing for a certain degree of misclassification, and v is the number of training samples. This objective function not only considers maximizing the classification margin (by ∥w∥) 2 (represented), and also considered reducing classification errors (by ξ) i (Represented). To ensure the existence and uniqueness of the solution, the constraints under the KKT conditions must also be satisfied.

[0169] To solve the above optimization problem, a common approach is to use the Lagrange multiplier method, introducing the Lagrange multiplier α, and transforming the primal problem into a dual problem. The dual problem is the maximization of α, and its expression is as follows:

[0170]

[0171] Here, K(x) i ,x j W(α) is the selected kernel function used to measure the similarity between two samples. The optimal kernel function can be obtained by taking the derivative of W(α) and setting it equal to zero. Thus, the optimal w is obtained. * and b * .

[0172] Once the optimal solution is found... * and b * This allows us to determine the final support vectors. These support vectors are the data points closest to the decision boundary, and they determine the position of the hyperplane. It's important to note that the SVM model relies only on these support vectors for prediction, not all training samples. This means that removing non-support vectors will not change the model's outcome.

[0173] After training is complete, the next step is to validate and fine-tune the model. This typically involves testing the model's performance on independent validation sets and adjusting the model parameters based on the results. The goal is to ensure that the model is neither overfitting nor underfitting, and that it can generalize well to unknown data.

[0174] In summary, by standardizing ambient light data, calculating decision functions using a pre-trained support vector machine model, and mapping the decision data to pre-stored evaluation intervals, a connection between ambient light data and user preferences can be effectively established, ultimately generating evaluation data that accurately reflects user preferences. This approach not only improves the reliability and accuracy of the analysis results but also provides a solid foundation for personalized services.

[0175] In step S15, the chromaticity feedback data and the user evaluation data are input into the pre-trained color correction model to obtain color adjustment data.

[0176] In one specific implementation, the training process of the color correction model includes:

[0177] The pre-stored historical color feedback data and pre-stored historical user evaluation data are input into the input layer of the initial color correction model for training, and the predicted value data output by the output layer of the initial color correction model is obtained.

[0178] Substitute the predicted value data and the pre-stored actual target value into the loss function to calculate the loss value data.

[0179] The gradient of the output layer of the initial color correction model is calculated based on the loss value data, and the gradient is passed forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer and obtain the gradient data.

[0180] Based on the gradient data and the preset learning rate, update the parameters of each layer of the initial color correction model;

[0181] The parameters of each layer are iteratively updated until the number of training iterations of the initial color correction model is greater than a preset number of iterations, or the loss value of the initial color correction model is less than a preset loss threshold. At this point, the training is considered complete, and the color correction model is obtained.

[0182] Specifically, the training of the color correction model begins with the preparation phase, which involves collecting and preprocessing historical color feedback data and user evaluation data. This data, after preprocessing operations such as cleaning, normalization, or standardization, is organized into a format suitable for neural network input. Subsequently, these datasets are divided into training and validation sets. The training set guides the learning of model parameters, while the validation set is used to evaluate the model's generalization ability and avoid overfitting.

[0183] During the training phase, the initial color correction model receives input from the training set—historical color feedback data and user evaluation data—which is passed through the input layer to the hidden layer for feature extraction and transformation. Finally, the output layer generates predicted data.

[0184] To measure the difference between the predicted value and the actual target value, a loss function is introduced. This function calculates the distance between the two, and the loss value reflects the current predictive performance of the model. A smaller loss value means that the prediction is closer to the true result.

[0185] Next, the gradient of the output layer is calculated based on the loss value data; that is, the derivative of the loss function with respect to the model parameters is taken. This step is to determine how to adjust the parameters to reduce the loss value. For multi-layer neural networks, this process requires using the chain rule to propagate the gradient of the output layer forward layer by layer until it reaches the input layer, thereby calculating the gradient of the parameters in each layer. In this way, the gradient information of each layer's parameters relative to the loss function can be obtained.

[0186] After obtaining the gradient information, the next step is to update the parameters of each layer of the model according to the preset learning rate. The update rule follows the gradient descent algorithm. Choosing an appropriate learning rate is crucial, as too large a learning rate may lead to over-updating of parameters, while too small a learning rate will slow down convergence. Furthermore, strategies such as momentum and adaptive learning rates can be combined to further improve optimization efficiency.

[0187] The entire training process is an iterative process, with each iteration consisting of four main steps: forward propagation, loss calculation, backpropagation, and parameter update. As the number of iterations increases, the model gradually learns to extract useful features from the input data and make more accurate predictions. Training terminates when either of the following conditions is met: first, the number of training iterations exceeds the preset maximum number; second, the model's loss value has fallen below the preset loss threshold, indicating that the model has reached a satisfactory level of accuracy.

[0188] After obtaining the trained color correction model, the chromaticity feedback data and user evaluation data need to be formatted into a form suitable for the input of the color correction model to ensure data consistency and compatibility.

[0189] Once data preprocessing is complete, the next step is to prepare the model input. At this stage, chromaticity feedback data and user evaluation data are combined to form a multidimensional feature vector, which serves as the model's input. This multidimensional feature vector contains all the necessary information, allowing the color correction model to predict new color adjustment parameters based on historical learning experience.

[0190] When input data enters the color correction model, the model performs forward propagation based on its internally learned parameters. During this process, each neuron receives information from the previous layer and applies an activation function for non-linear mapping, ultimately producing a set of values ​​representing color adjustment suggestions from the output layer. This series of values ​​constitutes the color adjustment data.

[0191] In summary, by inputting chromaticity feedback data and user evaluation data into a pre-trained color correction model, not only can effective acquisition of color adjustment data be achieved, but a deep understanding of the color correction model's training process also ensures that the model can provide high-quality color correction services in complex real-world scenarios. This process demonstrates the powerful capabilities and flexibility of machine learning technology in solving practical problems.

[0192] In step S16, the displayed content data is classified using the Kingfisher optimization algorithm, and the type weight index is calculated based on the classification results and the user evaluation data to obtain the weight index.

[0193] In one specific implementation, the step of classifying the displayed content data using the kingfisher optimization algorithm and calculating a type weight index based on the classification results and the user review data to obtain the weight index includes:

[0194] The display content data is classified using the kingfisher optimization algorithm to obtain the display content type.

[0195] The displayed content type and the user evaluation data are input into a pre-trained content type weight adjustment model to obtain a weight index.

[0196] First, to achieve effective classification of displayed content data, the Kingfisher Optimization Algorithm needs to be applied. The Kingfisher Optimization Algorithm is a novel metaheuristic algorithm inspired by the unique hunting behavior and symbiotic relationships of the kingfisher in nature. This algorithm is constructed around three different stages: perching / hovering prey (exploration / diversification), diving to find prey (development / intensive farming), and cultivating symbiotic relationships. For the task of classifying displayed content data, the Kingfisher Optimization Algorithm initiates the search process by initializing the population, i.e., randomly generating a set of initial solutions from the search space as the first attempt. Each individual represents a possible classification scheme, and the quality of these schemes is evaluated by a fitness function. As the number of iterations increases, the Kingfisher Optimization Algorithm continuously updates the individual positions, simulating the foraging activities of the kingfisher in its natural environment, in order to find the optimal or near-optimal classification scheme.

[0197] Specifically, in the initialization phase, the kingfisher optimization algorithm randomly generates an initial population based on the formula Xi;j = LB + rand × (UB - LB), where Xi;j represents the position of the i-th individual in the j-th dimension, rand is a random value between 0 and 1, and UB and LB are the upper and lower bounds of the search range, respectively. Subsequently, in the exploration phase, the algorithm updates the position of the search population based on the kingfisher's perching and hovering behavior, using the formula Xi(t+1) = Xi(t) + α × T, where α = 2 × randn(1,dim) - 1, randn is used to generate a random number tensor, dim specifies the size of the tensor in the second dimension, and T is dynamically determined based on the current strategy. In the development phase, the kingfisher optimization algorithm mimics the kingfisher's fast and accurate diving ability, using the formula Xi(t+1) = Best p osition+α×(Xi(t)-Best p (osition), where Best p The osiion is the best location discovered to date.

[0198] After classification, the next task is to input the display content types along with user review data into a pre-trained content type weighting model. This model aims to learn the relationship between different types of display content and user satisfaction, and calculate the corresponding weight index for each type. The pre-trained model has been thoroughly trained on a large amount of historical data, enabling it to capture the potential connections between display content features and user feedback. When new display content types and user review data are input, the model makes predictions based on its internal parameters and outputs the corresponding weight index.

[0199] To ensure the accuracy of the weight index calculation, the content type weight adjustment model was trained using a supervised learning method. During training, the model received historical display content types and their corresponding user rating data as input, and the actual type weight index as the target output. By minimizing the difference between the predicted and actual values, the model gradually adjusted its parameters until a satisfactory performance level was achieved.

[0200] In summary, the process of classifying displayed content data using the Kingfisher optimization algorithm and calculating type weight indices based on user review data is a comprehensive solution that integrates advanced optimization techniques and machine learning models. This method not only improves classification accuracy but also provides a scientific basis for subsequent decision-making by quantifying the importance of different types of displayed content.

[0201] In step S17, a color adjustment instruction is generated based on the color adjustment data and the weight index, and the displayed content is adaptively adjusted according to the color adjustment instruction.

[0202] In one specific implementation, the step of generating a color adjustment instruction based on the color perception data, the color adjustment data, and the weighting index, and adaptively adjusting the displayed content according to the color adjustment instruction, includes:

[0203] Calculate the color difference between the color adjustment data to obtain color difference data;

[0204] Calculate the weighting coefficient of the display content type based on the color difference data;

[0205] Generate color adjustment instructions based on the weighting coefficients and the weighting index;

[0206] The display screen is adjusted according to the color adjustment instructions and the color adjustment data, including gamma value adjustment, white balance adjustment, RGB gain adjustment, and color adjustment.

[0207] The color difference data is calculated using the following formula:

[0208]

[0209] The weighting coefficients are calculated using the following formula:

[0210]

[0211] Generate color adjustment instructions using the following formula:

[0212]

[0213] Where, ΔL * ij , Δa * ij Δb * ij ΔE represents the difference between the i-th and j-th color perception data points. ij w represents the color difference between the i-th color perception data and the j-th color perception data. ij W represents the weighting coefficient. i The weighting index is represented by k, the preset slope is represented by θ, and the preset threshold parameter is represented by A. i This indicates a color adjustment command.

[0214] First, to quantify the differences between color adjustment data, color difference data needs to be calculated. Color difference is an important indicator for measuring the perceptual difference between two colors, and the color difference formula in the CIE 1976Lab color space is used here. Specifically, for any pair of color perception data, the color difference between them is calculated using the following formula:

[0215]

[0216] This formula uses Euclidean distance to evaluate the difference between two sets of color perception data, providing an objective and quantitative measure of color similarity.

[0217] Next, based on the color difference data obtained above, the weighting coefficient for the displayed content type is further calculated. The weighting coefficient reflects the degree of influence of the color difference between a pair of color perception data on the final color adjustment decision, and its calculation formula is as follows:

[0218]

[0219] Here, k is a preset slope parameter used to control the steepness of the function curve; while θ is a threshold parameter that determines the starting point for the change of the weight coefficients. When ΔE ij When it is greater than θ, the weighting coefficient w ij The color difference will gradually increase, and vice versa. This mechanism ensures that only data with significant color differences will have a major impact on subsequent color adjustments.

[0220] Subsequently, combining the previously calculated weighting coefficients and weighting indices, color adjustment instructions are generated for each type of display content. These instructions comprehensively consider the importance of different types of display content and their inter-content color differences to guide specific color adjustment operations. The specific formula is:

[0221]

[0222] In this way, color adjustment instructions not only take into account the characteristics of individual displayed content, but also ensure the consistency and harmony of the overall visual experience.

[0223] Finally, based on the generated color adjustment instructions and data, a series of specific color adjustment measures are executed, including gamma adjustment, white balance adjustment, and RGB gain adjustment. These adjustments directly affect the physical characteristics of the display screen, such as the brightness response curve and color balance, thereby changing the actual presentation of the displayed content. For example, gamma adjustment can correct the image brightness distribution, making details in dark areas more clearly visible; white balance adjustment helps eliminate color temperature shifts caused by changes in light source, ensuring faithful color reproduction; and RGB gain adjustment allows independent control of the proportions of the three primary colors—red, green, and blue—optimizing color saturation and contrast.

[0224] In summary, through in-depth analysis of color adjustment data and by utilizing methods such as color difference calculation, weighting coefficient determination, and color adjustment parameter setting, precise adaptive adjustment of the displayed content was achieved. This process not only improved display quality but also enhanced the consistency and comfort of the user experience.

[0225] The following describes the working process of this invention using a common scenario as an example. For specific embodiments of this invention, please refer to... Figure 1 A color adjustment method for a display screen based on environmental information includes the following steps:

[0226] Assume the environment is a well-lit office where a monitor displays professional photographs. To ensure accurate colors and optimal visual effects, a color adjustment system begins operation. First, the system receives data from an ambient light sensor, including the current ambient light intensity and color temperature. Simultaneously, the system also acquires the user's color preference settings.

[0227] Next, the color adjustment system analyzed the type of the currently displayed content and identified it as a series of professional photographs requiring high color fidelity. Based on this, the system assessed the gap between the existing display settings and the ideal state, and determined the necessary color adjustment parameters.

[0228] The color adjustment system then calculates the color difference between different color perception data and adjusts the weighting coefficients for the displayed content type based on this. This process ensures that the color adjustment instructions fully reflect the impact of ambient light conditions and user preferences on the final display effect.

[0229] Subsequently, the color adjustment system generates color adjustment instructions based on the calculated weighting coefficients and pre-set weighting indices. These instructions comprehensively consider various factors to achieve the best visual experience. Following the color adjustment instructions, the system adjusts the gamma value of the display screen, making the brightness distribution of the image more uniform and reasonable; it adjusts the white balance to ensure faithful color reproduction under different light sources; and it adjusts the RGB gain to optimize color saturation and contrast, ensuring that every photo displays the most realistic colors.

[0230] After the above adjustments were completed, the professional photographs displayed on the screen exhibited outstanding color performance, with vibrant and natural colors, rich details, and distinct layers, perfectly meeting the photographer's creative intent and the viewer's viewing needs. Throughout the process, the color adjustment system, through precise environmental information analysis and intelligent color adjustment, achieved optimal presentation of the displayed content in specific environments, solving the problems of high requirements for light source environment and strong subjectivity in display screen color adjustment.

[0231] Reference Figure 2 The second embodiment of the present invention provides a color adjustment system for a display screen based on environmental information, comprising:

[0232] The data acquisition module is used to acquire the display content data, display environment data, and display type of the display screen, wherein the display environment data includes ambient color data and ambient light data;

[0233] A color perception calculation module is used to calculate color perception data based on the ambient color data and the display screen type.

[0234] The chromaticity feedback analysis module is used to perform quantization, smoothing adjustment, cluster analysis, and multinomial regression analysis on the color perception data to obtain chromaticity feedback data.

[0235] The user evaluation analysis module is used to establish the relationship between the ambient light data and user preferences using the support vector machine method to obtain user evaluation data.

[0236] The color correction module is used to input the chromaticity feedback data and the user evaluation data into a pre-trained color correction model to obtain color adjustment data;

[0237] The weight index calculation module is used to classify the displayed content data using the kingfisher optimization algorithm, and calculate the type weight index based on the classification results and the user evaluation data to obtain the weight index;

[0238] The color adjustment module is used to generate color adjustment instructions based on the color adjustment data and the weight index, and to adaptively adjust the displayed content according to the color adjustment instructions.

[0239] It should be noted that the color adjustment device for a display screen based on environmental information provided in this embodiment of the invention is used to execute all the process steps of the color adjustment method for a display screen based on environmental information in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0240] This invention also 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 color adjustment method program for a display screen based on environmental information. When the processor executes the computer program, it implements the steps in the various embodiments of the color adjustment method program for a display screen based on environmental information described above, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the color adjustment method module for a display screen based on environmental information.

[0241] For example, the computer program may be divided into one or more modules / units, which 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 a specific function, which describe the execution process of the computer program in the electronic device.

[0242] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0243] The processor can be a Central Processing Unit (CPU), or 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. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0244] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0245] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed 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 electrical carrier signals and telecommunication signals.

[0246] 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 separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0247] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A color adjustment method for a display screen based on environmental information, characterized in that, Executed by the controller, including: Acquire display content data, display environment data, and display type of the display screen, where the display environment data includes ambient color data and ambient light data; Calculate color perception data based on the ambient color data and the display screen type; The color perception data is quantized, smoothed, clustered, and subjected to multinomial regression analysis to obtain color feedback data. The relationship between the ambient light data and user preferences is established using the support vector machine method to obtain user evaluation data; The chromaticity feedback data and the user evaluation data are input into a pre-trained color correction model to obtain color adjustment data; The displayed content data is classified using the Kingfisher optimization algorithm, and a type weight index is calculated based on the classification results and the user evaluation data to obtain the weight index. A color adjustment instruction is generated based on the color adjustment data and the weight index, and the displayed content is adaptively adjusted according to the color adjustment instruction. The step of calculating color perception data based on the ambient color data and the display screen type includes: Perform gamma correction on the environmental color data to obtain corrected color data; Select the elements in the corresponding independent space transformation matrix according to the display screen type; The corrected color data is then subjected to independent spatial transformation to obtain independent color data; The independent color data is converted to a uniform color space to obtain color perception data. Gamma correction is performed using the following formula: ； Independent space transformations are performed using the following formula: ； The following formula is used to perform a perceptual uniform color space conversion: ； ； in, Represents environmental color data. This indicates the preset gamma value. Indicates the corrected color data. This represents the red data in the corrected color data. This indicates the green data in the corrected color data. This indicates the blue data in the corrected color data. , , , , , , , , This represents the elements in the predefined independent space transformation matrix. , , Represents independent color data. , , This represents the tristimulus value of the preset reference white point. , , Represents color perception data, This represents the result of dividing the value of the substituted independent color data by its corresponding reference white point.

2. The color adjustment method for a display screen based on environmental information according to claim 1, characterized in that, The process of quantizing, smoothing, clustering, and performing multinomial regression analysis on the color perception data to obtain color feedback data includes: The color perception data is quantized using a uniform quantization method to obtain quantized data. The quantized data is smoothed by using a Gaussian filter to obtain smoothed data. The smoothed data was clustered using the K-means clustering algorithm to obtain clustered analysis data; The clustering data was analyzed using multinomial regression analysis to obtain color feedback data. Quantification is performed using the following formula: ； The Gaussian filter is shown below: ； in, Representing color perception data or or One of the data, This represents the minimum value in the color perception data. This represents the maximum value in the color perception data. This represents the total number of data points related to color perception. Represents quantified data. Indicates the first A quantitative data point, and This represents the preset filter parameters. This indicates smoothed data.

3. The color adjustment method for a display screen based on environmental information according to claim 1, characterized in that, The step of establishing the relationship between the ambient light data and user preferences using the support vector machine method to obtain user evaluation data includes: The ambient light data is standardized to obtain standardized data; The standardized data is input into a pre-trained support vector machine model to calculate a decision function, and the output of the decision function is used as the decision data. The decision data is mapped to a pre-stored evaluation interval to obtain user evaluation data; The decision function is as follows: ； in, Represents the decision function. This indicates the number of samples of pre-stored historical ambient light data. This represents the optimal solution of the Lagrange multipliers after training is complete. The obsessive items indicating training completion The weight coefficients represent the weights after training is complete. This represents the preset kernel function. The first part representing historical ambient light data One data point, This represents ambient light data.

4. The color adjustment method for a display screen based on environmental information according to claim 1, characterized in that, The training process of the color correction model includes: The pre-stored historical color feedback data and pre-stored historical user evaluation data are input into the input layer of the initial color correction model for training, and the predicted value data output by the output layer of the initial color correction model is obtained. Substitute the predicted value data and the pre-stored actual target value into the loss function to calculate the loss value data. The gradient of the output layer of the initial color correction model is calculated based on the loss value data, and the gradient is passed forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer and obtain the gradient data. Based on the gradient data and the preset learning rate, update the parameters of each layer of the initial color correction model; The parameters of each layer are iteratively updated until the number of training iterations of the initial color correction model is greater than a preset number of iterations, or the loss value of the initial color correction model is less than a preset loss threshold. At this point, the training is considered complete, and the color correction model is obtained.

5. The color adjustment method for a display screen based on environmental information according to claim 1, characterized in that, The process involves classifying the displayed content data using the kingfisher optimization algorithm, and calculating a type weight index based on the classification results and user review data. The resulting weight index includes: The display content data is classified using the kingfisher optimization algorithm to obtain the display content type. The displayed content type and the user evaluation data are input into a pre-trained content type weight adjustment model to obtain a weight index.

6. The color adjustment method for a display screen based on environmental information according to claim 5, characterized in that, The step of generating color adjustment instructions based on the color perception data, the color adjustment data, and the weight index, and adaptively adjusting the displayed content according to the color adjustment instructions, includes: Calculate the color difference between the color adjustment data to obtain color difference data; Calculate the weighting coefficient of the display content type based on the color difference data; Generate color adjustment instructions based on the weighting coefficients and the weighting index; The display screen is adjusted according to the color adjustment instructions and the color adjustment data, including gamma value adjustment, white balance adjustment, RGB gain adjustment, and color adjustment. The color difference data is calculated using the following formula: ； The weighting coefficients are calculated using the following formula: ； Generate color adjustment instructions using the following formula: ； in, , , Indicates the first The first color perception data and the second The difference between the color perception data, Indicates the first The first color perception data and the second Color difference of individual color perception data Indicates the weighting coefficient. Indicates the weighting index. This indicates the preset slope. This represents the preset threshold parameter. This indicates a color adjustment command.

7. A color adjustment system for a display screen based on environmental information, characterized in that, A color adjustment method for a display screen based on environmental information as described in any one of claims 1 to 6, comprising: The data acquisition module is used to acquire the display content data, display environment data, and display type of the display screen, wherein the display environment data includes ambient color data and ambient light data; A color perception calculation module is used to calculate color perception data based on the ambient color data and the display screen type. The chromaticity feedback analysis module is used to perform quantization, smoothing adjustment, cluster analysis, and multinomial regression analysis on the color perception data to obtain chromaticity feedback data. The user evaluation analysis module is used to establish the relationship between the ambient light data and user preferences using the support vector machine method to obtain user evaluation data. The color correction module is used to input the chromaticity feedback data and the user evaluation data into a pre-trained color correction model to obtain color adjustment data; The weight index calculation module is used to classify the displayed content data using the kingfisher optimization algorithm, and calculate the type weight index based on the classification results and the user evaluation data to obtain the weight index; The color adjustment module is used to generate color adjustment instructions based on the color adjustment data and the weight index, and to adaptively adjust the displayed content according to the color adjustment instructions.

8. An electronic device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the color adjustment method for a display screen based on environmental information as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the color adjustment method for a display screen based on environmental information as described in any one of claims 1 to 6.

Images