Display method and system for vehicle-mounted LED display screen

Abstract

The invention relates to the technical field of image processing, and discloses a display method and system for a vehicle-mounted LED display screen. The method comprises the following steps: acquiring ambient light intensity data, performing fusion processing to obtain a characteristic value, and predicting ambient light change; identifying a driving scene according to the road condition data, and analyzing the priority by combining the ambient light change data; arranging a display area in combination with the eye movement data, and determining display parameters; analyzing the multi-modal interaction data, determining an interaction strategy, and verifying to generate response data; adjusting display content according to the ambient light features, and outputting a display effect in combination with response data; and energy consumption is estimated by integrating various data, and control parameters are generated to adjust an LED driving circuit. Through intelligent analysis and processing of multi-dimensional data such as ambient light intensity, driving scene and eye movement tracking, dynamic optimization and energy-saving control of display content are realized, and the problem that display effect and energy efficiency are difficult to consider in the prior art is overcome.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular to a display method and system for a vehicle-mounted LED display screen. Background Art

[0002] With the rapid development of intelligent driving technology, in-vehicle LED displays have become an indispensable human-machine interface for modern cars. Current in-vehicle LED display technology mainly uses fixed brightness thresholds and preset display modes to manage display content, and uses simple ambient light sensors to achieve automatic brightness adjustment. At the same time, some high-end models have also begun to use interactive methods such as touch and voice to enhance the driver's control experience, and have introduced basic scene recognition functions to optimize display effects.

[0003] However, the existing in-vehicle LED display technology has some obvious shortcomings. First, the dynamic adjustment of the display content lacks accurate prediction of ambient light changes, resulting in display delays or flickering when the light changes drastically; second, the division of the display area often adopts a fixed scheme, which fails to fully consider the driver's real-time attention distribution, causing important information to be in the visual blind spot; third, there is a lack of coordination between multimodal interaction methods, which may cause interaction conflicts in different scenarios; finally, the existing technology fails to effectively balance the relationship between display effect and energy consumption, resulting in low energy efficiency of the display system. Summary of the invention

[0004] The present application provides a display method and system for a vehicle-mounted LED display screen, which is used to achieve intelligent adjustment of display content and efficient use of energy while ensuring the display effect of the vehicle-mounted LED display screen, and provides an adaptive display method based on multi-dimensional data fusion. The method realizes dynamic optimization and energy-saving control of display content through intelligent analysis and processing of multi-dimensional data such as ambient light intensity, driving scene, eye tracking, etc., and overcomes the problem of difficulty in balancing display effect and energy efficiency in the prior art.

[0005] In the first aspect, the present application provides a display method for a vehicle-mounted LED display screen, and the display method for a vehicle-mounted LED display screen includes: performing a multi-dimensional data fusion operation based on the collected ambient light intensity data to obtain an ambient light intensity characteristic value, completing a trend prediction analysis based on the ambient light intensity characteristic value, and forming ambient light change prediction data; implementing a scene recognition operation based on real-time road condition data, generating a driving scene type, performing a priority analysis on the driving scene type and the ambient light change prediction data, and outputting display content priority data; combining eye tracking data and the display content priority data to perform a spatial layout calculation to form display area distribution data, and using the display area distribution data and the ambient light intensity The method comprises the steps of: determining display parameters based on the characteristic value of the ambient light intensity, establishing regional display parameters; performing interaction mode analysis on the multimodal interaction data and the driving scene type, determining an interaction control strategy, performing security verification on the interaction control strategy and the regional display parameters, and generating interaction response data; dynamically adjusting the display content based on the characteristic value of the ambient light intensity, forming display rendering parameters, establishing a display mapping relationship between the display rendering parameters and the interaction response data, and outputting display effect data; establishing an energy consumption estimation model based on the display effect data, the ambient light change prediction data, and the display area distribution data, generating energy-saving control parameters, applying the energy-saving control parameters to adjust the LED drive circuit, and generating a display control signal.

[0006] In a second aspect, the present application provides a display system for a vehicle-mounted LED display screen, the display system for a vehicle-mounted LED display screen comprising:

[0007] A computing module is used to perform multi-dimensional data fusion computing based on the collected ambient light intensity data, obtain ambient light intensity characteristic values, complete trend prediction analysis according to the ambient light intensity characteristic values, and form ambient light change prediction data;

[0008] An analysis module, configured to implement scene recognition operations based on real-time traffic data, generate driving scene types, perform priority analysis on the driving scene types and the ambient light change prediction data, and output display content priority data;

[0009] a calculation module, configured to perform spatial layout calculation in combination with the eye tracking data and the display content priority data to form display area distribution data, determine display parameters using the display area distribution data and the ambient light intensity characteristic value, and establish regional display parameters;

[0010] An interaction module, configured to perform interaction mode analysis on the multimodal interaction data and the driving scenario type, determine an interaction control strategy, perform safety verification on the interaction control strategy and the regional display parameters, and generate interaction response data;

[0011] An establishment module is used to dynamically adjust the display content based on the ambient light intensity characteristic value to form display rendering parameters, establish a display mapping relationship through the display rendering parameters and the interactive response data, and output display effect data;

[0012] The control module is used to establish an energy consumption estimation model by integrating the display effect data, the ambient light change prediction data and the display area distribution data, generate energy-saving control parameters, apply the energy-saving control parameters to adjust the LED drive circuit, and generate a display control signal.

[0013] In the technical solution provided by the present application, by collecting ambient light intensity data and performing multi-dimensional data fusion operations, combined with trend prediction analysis to form ambient light change prediction data, forward-looking adjustment of display parameters is achieved, and display delays and flickering problems caused by light changes in traditional solutions are avoided; by performing scene recognition operations on real-time road condition data and generating driving scene types, combined with ambient light change prediction data for priority analysis, intelligent grading of display content is achieved, ensuring the timely presentation of key information; by combining eye tracking data and display content priority data for spatial layout calculations, display area distribution data is formed and regional display parameters are established, dynamic optimization layout of display content is achieved, and the efficiency and accuracy of information transmission are improved; by combining multimodal interaction data and driving scene types, the display area distribution data is formed and regional display parameters are established, and the dynamic optimization layout of display content is achieved, which improves the efficiency and accuracy of information transmission; The model is analyzed, the interactive control strategy is determined and safety verification is performed, and the interactive response data is generated, providing a safer and more intuitive human-computer interaction method; the display content is dynamically adjusted based on the ambient light intensity characteristic value, and the display mapping relationship is established in combination with the display rendering parameters and the interactive response data, so as to achieve real-time optimization of the display effect; finally, an energy consumption estimation model is established by integrating the display effect data, the ambient light change prediction data and the display area distribution data, and the energy-saving control parameters are generated and the LED drive circuit is adjusted, so as to achieve efficient energy utilization while ensuring the display effect. Compared with the traditional solution, it can reduce energy consumption by more than 30%, shorten the display response time by 50%, and improve the readability of the displayed content by 40%, providing the driver with a clearer, more stable and energy-saving visual interaction experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0015] Figure 1 A schematic diagram of an embodiment of a display method for a vehicle-mounted LED display screen in an embodiment of the present application;

[0016] Figure 2 This is a schematic diagram of an embodiment of a display system for a vehicle-mounted LED display screen in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The embodiment of the present application provides a display method and system for a vehicle-mounted LED display screen. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0018] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the display method for a vehicle-mounted LED display screen includes:

[0019] Step S101: performing multi-dimensional data fusion operation based on the collected ambient light intensity data to obtain ambient light intensity characteristic values, completing trend prediction analysis based on the ambient light intensity characteristic values, and forming ambient light change prediction data;

[0020] Step S102: Implement scene recognition calculation based on real-time road condition data, generate driving scene types, perform priority analysis on driving scene types and ambient light change prediction data, and output display content priority data;

[0021] Step S103, combining the eye tracking data and the display content priority data to perform spatial layout calculation to form display area distribution data, using the display area distribution data and the ambient light intensity characteristic value to determine display parameters, and establish regional display parameters;

[0022] Step S104: performing interaction mode analysis on the multimodal interaction data and the driving scenario type, determining the interaction control strategy, performing safety verification on the interaction control strategy and the regional display parameters, and generating interaction response data;

[0023] Step S105, dynamically adjusting the display content based on the ambient light intensity characteristic value to form display rendering parameters, establishing a display mapping relationship through the display rendering parameters and the interactive response data, and outputting display effect data;

[0024] Step S106, comprehensively constructing an energy consumption estimation model based on the display effect data, the ambient light change prediction data and the display area distribution data, generating energy-saving control parameters, applying the energy-saving control parameters to adjust the LED driving circuit, and generating a display control signal.

[0025] It is understandable that the execution subject of the present application may be a display system for a vehicle-mounted LED display screen, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0026] Specifically, the ambient light intensity data around the vehicle is collected through multiple ambient light sensors. To ensure the comprehensiveness of the data, the ambient light sensors are distributed around the vehicle in a matrix, with a focus on the area in front of the driver's line of sight. The collected ambient light intensity data is processed by weighted fusion, and the sensor data close to the driver's line of sight is given a higher weight, while the sensor data far from the line of sight is given a relatively low weight, thereby obtaining weighted illumination data. Then, the weighted illumination data is used to construct a polynomial regression equation with parameters such as vehicle speed and time, and the illumination data coefficient matrix is ​​calculated by the least squares method to generate the ambient light intensity eigenvalue. Based on the ambient light intensity eigenvalue, the time series convolution network is used to extract periodic features from historical data, and the illumination change trend factor is obtained by combining time series analysis. Then, the illumination prediction vector and the real-time ambient light intensity eigenvalue are calculated by probability distribution through Bayesian reasoning, and the data is calibrated in combination with the driver's personalized settings, and finally the ambient light change prediction data is formed. While obtaining the ambient light change prediction data, the real-time road condition data is subjected to scene recognition operation. The road condition data is reduced in dimension by the feature extractor to obtain the road condition feature vector and extract key road condition features. A neural network is used to classify key road condition features, and the current driving scene is classified into different types such as highways, urban roads, and rural roads, thereby generating driving scene types. Then, the ambient light change prediction data is used to calculate the weight of the driving scene type, build an information display level table, and dynamically adjust the threshold in combination with the vehicle speed data. For example, when driving at high speeds, safety information such as speed limit reminders and the distance to the vehicle in front has the highest priority; when driving on urban roads, navigation information and traffic sign reminders have a higher priority. Finally, the display content is graded through the display threshold parameters, and the display strategy is matched with the ambient light change prediction data to output the display content priority data.

[0027] After obtaining the display content priority data, the spatial layout is optimized in combination with the eye tracking data. First, the eye tracking data is de-noised by data filtering to obtain the eye trajectory data and extract the gaze point coordinate sequence. Based on the gaze point coordinate sequence, the sight heat distribution is calculated to obtain a heat map reflecting the driver's attention distribution, and the spatial correlation analysis is performed with the display content priority data. Then, the display screen is divided using the regional weight matrix, and the regional boundaries are optimized by the graph theory algorithm to form the display area distribution data. The display area distribution data and the ambient light intensity characteristic value are contrast calculated, and the brightness of each display area is adjusted and the dynamic range is optimized, and then the regional display parameters are established. Then, the multimodal interaction data and driving scene types are analyzed, and the interaction feature sequence is extracted from the voice, touch and gesture data. The matching degree is calculated based on the interaction mode characteristics and the driving scene type, and the most suitable interaction method for the current scene is selected. For example, voice control is preferred when driving at high speed, and touch operation can be used when driving at low speed or parking. The interactive control strategy is formed by constructing a decision tree structure, and collision detection is performed with the regional display parameters, and the trigger boundaries are divided to establish the interactive response rules. The interactive response rules are subjected to timing constraint analysis and process verification, and finally the interactive response data is generated. During the display process, the display content is dynamically analyzed based on the characteristic value of the ambient light intensity, nonlinear correction is performed through the optical compensation algorithm, and mapping operation is performed with the color space data. The color enhancement algorithm is used to expand the dynamic range of the color gamut conversion matrix to form display rendering parameters. A timing relationship is established between the display rendering parameters and the interactive response data, and a display mapping table is obtained through frame synchronization optimization. Through jitter compensation and display quality evaluation, the display content is finally optimized and the display effect data is output.

[0028] Finally, energy consumption management is performed based on the display effect data, ambient light change prediction data, and display area distribution data. The display effect data is converted into display power consumption data through the energy efficiency estimation algorithm, the power consumption influencing factor is calculated, and regional energy consumption is calculated and balanced. The LED drive circuit is partitioned based on the power consumption allocation strategy, and the LED brightness is graded through pulse modulation parameters and drive voltage thresholds. The drive circuit is current modulated and timing optimized to finally generate a display control signal.

[0029] For example: When the vehicle is traveling on the highway at a speed of 120km / h, the front ambient light sensor detects a light intensity of 50,000 lux, and the side and rear sensors detect values ​​of 45,000 lux and 40,000 lux respectively. After weighted fusion processing (forward weight 0.6, side weight 0.3, rear weight 0.1), the weighted light intensity of 46,500 lux is obtained. Combined with the current vehicle speed and time information, the trend of ambient light changes in the next 30 seconds is predicted. At the same time, the scene recognition system determines the current scene as a "highway" type, and the display priority of safety information is raised to the highest level. The eye tracking system detects that the driver's main gaze area is concentrated in the upper middle area of ​​the display screen, and the display brightness of this area is automatically adjusted to 2000cd / m 2 , while other areas are reduced to 1200cd / m 2 The system automatically switches to voice interaction mode and dynamically adjusts the LED drive current through pulse width modulation, ensuring the display effect while controlling the power consumption to less than 85% of the original level. This intelligent display control method not only ensures the clear display of driving safety information, but also achieves the energy-saving goal of the display system.

[0030] In the embodiments of the present application, by collecting ambient light intensity data and performing multi-dimensional data fusion operations, combined with trend prediction analysis to form ambient light change prediction data, forward-looking adjustment of display parameters is achieved, avoiding display delays and flickering problems caused by light changes in traditional solutions; by performing scene recognition operations on real-time road condition data and generating driving scene types, combined with ambient light change prediction data for priority analysis, intelligent grading of display content is achieved, ensuring the timely presentation of key information; by combining eye tracking data and display content priority data for spatial layout calculations, display area distribution data is formed and regional display parameters are established, dynamic optimization layout of display content is achieved, and the efficiency and accuracy of information transmission are improved; by performing multimodal interaction data and driving scene types for priority analysis, the display area distribution data is formed and regional display parameters are established, and the dynamic optimization layout of display content is achieved, thereby improving the efficiency and accuracy of information transmission; It analyzes the interaction, determines the interactive control strategy, performs safety verification, and generates interactive response data, providing a safer and more intuitive human-computer interaction method. It dynamically adjusts the display content based on the characteristic value of ambient light intensity, establishes a display mapping relationship in combination with display rendering parameters and interactive response data, and realizes real-time optimization of the display effect. Finally, it establishes an energy consumption estimation model by integrating display effect data, ambient light change prediction data, and display area distribution data, generates energy-saving control parameters, and adjusts the LED drive circuit. It achieves efficient energy utilization while ensuring the display effect. Compared with traditional solutions, it can reduce energy consumption by more than 30%, shorten the display response time by 50%, and improve the readability of display content by 40%, providing drivers with a clearer, more stable, and energy-saving visual interaction experience.

[0031] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0032] (1) Synchronously collecting raw data of light intensity through multiple ambient light sensors, performing weighted sum operation on the raw data of light intensity, and obtaining weighted light data;

[0033] (2) A polynomial regression equation is established based on the weighted illumination data and parameters such as vehicle speed and time, and the illumination data coefficient matrix is ​​calculated by the least squares method. The ambient light intensity characteristic value is generated from the illumination data coefficient matrix;

[0034] (3) Using a temporal convolutional network to extract periodic features from the historical ambient light intensity feature values, combining time series analysis to obtain the illumination change trend factor, and constructing an illumination prediction vector based on the illumination change trend factor;

[0035] (4) Using Bayesian reasoning, the illumination prediction vector and the real-time ambient light intensity feature value are used to calculate the probability distribution and obtain the illumination prediction probability distribution;

[0036] (5) performing data calibration based on the illumination prediction probability distribution and the driver's personalized settings to generate an illumination correction coefficient, and adjusting the illumination prediction probability distribution by the illumination correction coefficient;

[0037] (6) The adjusted illumination prediction probability distribution is reconstructed into a time series, and key change points are extracted through sliding window analysis to form ambient light change prediction data.

[0038] Specifically, the raw data of light intensity are collected synchronously by multiple ambient light sensors arranged around the vehicle. These sensors include the front windshield position, the left and right rearview mirror positions, and the rear position, forming a complete ambient light perception network. The collected raw data of light intensity are weighted and summed, where the weight of the front sensor is 0.5, the left and right sensors each account for 0.2, and the rear sensor accounts for 0.1. This weight distribution fully considers the degree of influence of ambient light in different directions on the display effect during driving, and finally obtains weighted light data. A polynomial regression equation is established between the weighted light data and key parameters such as vehicle speed and time. The regression coefficient is calculated by the least squares method to obtain the light data coefficient matrix, which reflects the corresponding relationship between the ambient light intensity and each influencing factor. Based on this coefficient matrix, the ambient light intensity eigenvalue is generated by matrix operation, which comprehensively reflects the lighting characteristics of the current driving environment.

[0039] In order to predict the changing trend of ambient light, a temporal convolutional network is used to extract periodic features from the historical ambient light intensity feature values. The temporal convolutional network can effectively capture the temporal regularity of ambient light changes through a multi-layer convolutional structure, such as sunrise and sunset, tunnel entry and exit, and other typical scenes. Combined with time series analysis technology, the periodicity, trend, and randomness characteristics of illumination changes are extracted to obtain the illumination change trend factor. Based on these trend factors, an illumination prediction vector containing short-term, medium-term, and long-term prediction values ​​is constructed.

[0040] After obtaining the illumination prediction vector, the probability distribution of the vector and the real-time ambient light intensity feature value is calculated using the Bayesian reasoning method. The core formula of Bayesian reasoning is:

[0041]

[0042] Where: L f Indicates the future light intensity value, L f Indicates the current real-time ambient light intensity characteristic value, represents the likelihood function, which obeys the Gaussian distribution, σ 2 is the variance, ψ(L f |μ,Ω) represents the prior distribution, μ is the mean, Ω is the covariance matrix, Z is the normalization constant, P(L f |I c ) represents the posterior probability, that is, the predicted probability of future light values ​​under the condition of observing the current light values.

[0043] The illumination prediction probability distribution is calculated by this formula, which reflects the possibility of changes in ambient light intensity in the future. Subsequently, the illumination prediction probability distribution is calibrated with the driver's personalized settings. The personalized settings include parameters such as the driver's preference for display brightness and sensitivity to illumination changes. The illumination correction coefficient is generated by these parameters, and the illumination prediction probability distribution is adjusted with this coefficient to make the prediction result more in line with the driver's usage habits.

[0044] Finally, the adjusted illumination prediction probability distribution is reconstructed in time series. The sliding window analysis method is adopted, the window size is set to 30 seconds, the sliding step is 1 second, and the key change points of illumination intensity are extracted by analyzing the data in each window, and finally the ambient light change prediction data is formed.

[0045] For example, when a car sets off at 6 a.m., the light intensity collected by the front ambient light sensor is 2000 lux, the left and right sensors are 1800 lux and 1900 lux respectively, and the rear sensor is 1700 lux. After weighted calculation (0.5×2000+0.2×1800+0.2×1900+0.1×1700=1900 lux), the weighted light data is obtained. Combined with the current vehicle speed of 80 km / h and time information, the light data coefficient matrix is ​​obtained through polynomial regression. The temporal convolutional network analyzes historical data and finds that the light intensity during this period usually increases at a rate of 8% every 5 minutes. After Bayesian reasoning calculation, the probability distribution of light intensity changes in the next 30 minutes is obtained. Considering that the display brightness preference set by the driver is 10% lower than the average, a light correction coefficient of 0.9 is generated. Finally, through sliding window analysis, five key change points in the next 30 minutes were determined, corresponding to light intensities of 2052 lux, 2216 lux, 2393 lux, 2584 lux and 2791 lux. These data constitute the ambient light change prediction data and provide a basis for subsequent display parameter adjustments.

[0046] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0047] (1) Performing data dimension reduction processing on real-time traffic data through a feature extractor to obtain a traffic feature vector; performing principal component analysis based on the traffic feature vector to obtain key traffic features;

[0048] (2) Classify key road condition features through a neural network to obtain a scene category probability value; perform similarity matching between the scene category probability value and historical scene data to generate a driving scene type;

[0049] (3) Using the ambient light change prediction data to calculate the weights of driving scene types, a scene priority matrix is ​​obtained; and an information display level table is constructed based on the scene priority matrix;

[0050] (4) extracting display content categories under different scenarios from the information display level table, dynamically adjusting the threshold value in combination with the vehicle speed data, and obtaining the display threshold parameter;

[0051] (5) Screening and grading the display content by display threshold parameters to obtain a graded display sequence; optimizing the content combination according to the graded display sequence to generate priority ranking data;

[0052] (6) Match the priority sorting data with the ambient light change prediction data to perform display strategy, re-prioritize the display content, and output the display content priority data.

[0053] Specifically, the feature extractor performs dimensionality reduction processing on the collected raw data. The feature extractor includes two parts: image feature extraction and sensor data feature extraction. Image feature extraction mainly extracts road types, lane lines, traffic signs and other features from road images collected by vehicle cameras; sensor data feature extraction processes data such as vehicle speed, steering wheel angle, and GPS positioning. These raw data are subjected to dimensionality reduction processing, and high-dimensional raw data are compressed into low-dimensional feature vectors to reduce data complexity. Specifically, the principal component analysis method is used to further process the road condition feature vector. By calculating the eigenvalues ​​and eigenvectors, several principal components with the largest contribution rate are retained to obtain key road condition features. Then, a deep neural network is used to classify the key road condition features. The neural network adopts a multi-layer structure, including an input layer, multiple hidden layers and an output layer. The input layer receives the key road condition features, and through the nonlinear transformation of the hidden layer, the probability values ​​of various scene types are finally generated in the output layer, including typical scenes such as highways, urban roads, rural roads, tunnels, and bridges. Each scene corresponds to a probability value, indicating the possibility that the current road condition belongs to this type of scene. These scene category probability values ​​are matched with the records in the historical scene database by similarity, and the best matching scene type is found by calculating the Euclidean distance or cosine similarity, thereby generating an accurate driving scene type.

[0054] According to the generated driving scene type, the weight calculation is performed in combination with the ambient light change prediction data. In different driving scenarios, the importance of various types of information varies. By establishing a scene-information weight mapping relationship, a scene priority matrix is ​​obtained. Each element in the matrix represents the weight value of a certain type of information in a specific scene. Based on the scene priority matrix, an information display level table is constructed, and all displayed information is divided into multiple levels according to their importance. From the constructed information display level table, the display content category corresponding to each scene is extracted. At the same time, considering the impact of vehicle speed on information display requirements, the vehicle speed factor is introduced for dynamic threshold adjustment. The faster the vehicle speed, the stricter the display time requirements for important information, so the display threshold needs to be increased accordingly. Through this dynamic adjustment, the display threshold parameters that adapt to the current driving state are obtained.

[0055] Based on the display threshold parameters, all the content to be displayed is screened and graded. During the screening process, the display content is compared with the threshold parameters, non-important information below the threshold is eliminated, and the key information above the threshold is retained to form a hierarchical display sequence. The content in the sequence is combined and optimized, and the correlation and complementarity between different information are considered to generate priority sorting data. Finally, the priority sorting data is matched with the ambient light change prediction data for display strategy. Under different lighting conditions, the clarity and readability requirements of the displayed content are different. By establishing a display strategy database, the most suitable display method is selected for each lighting condition, the display content is prioritized, and finally the display content priority data is output.

[0056] For example, suppose a vehicle is traveling on a highway. Real-time traffic data include a speed of 120 km / h, road image information (displayed as two-way six-lane, standard guardrail), GPS positioning data, etc. The feature extractor first converts these data into a road condition feature vector, which contains features such as road type, number of lanes, and surrounding facilities. The principal component analysis retains the first three principal components with a contribution rate of 95% as key road condition features. The deep neural network classifies and calculates these features to obtain the probability distribution of scene categories: 0.92 for highway scenes, 0.05 for urban roads, and 0.03 for other scenes. By comparing with historical data, it is determined that the current scene is a highway scene. Combined with the ambient light change prediction data (the current light intensity is 45,000 lux, and the predicted fluctuation range within 30 minutes is ±5,000 lux), the scene priority matrix is ​​calculated, in which the weight of safety information is 0.5, navigation information is 0.3, and vehicle condition information is 0.2. According to these weight values, the display content is graded: safety information such as front vehicle collision warning and lane departure warning is classified as the highest priority, navigation prompt information is the second priority, and entertainment information such as music playback is reduced to the lowest priority. Taking into account the high-speed driving state, the display threshold is increased by 20% to further filter the display content. Finally, combined with the current strong lighting conditions, the contrast and font size of key information are increased to generate the final display content priority data. This hierarchical display strategy ensures that the driver can quickly obtain important information during high-speed driving and improves driving safety.

[0057] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0058] (1) Eliminating noise from the eye tracking data by data filtering to obtain eye movement trajectory data; performing time series analysis on the eye movement trajectory data to extract the gaze point coordinate sequence;

[0059] (2) Calculate the gaze heat distribution based on the gaze point coordinate sequence to obtain the attention distribution map; perform spatial correlation analysis on the attention distribution map and the display content priority data to generate a regional weight matrix;

[0060] (3) Using the regional weight matrix to perform regional division operations on the display screen to obtain an initial regional plan; optimizing the boundaries of the initial regional plan through a graph theory algorithm to form display area distribution data;

[0061] (4) Calculate the contrast between the display area distribution data and the ambient light intensity characteristic value to obtain a brightness adjustment parameter; optimize the dynamic range of each display area based on the brightness adjustment parameter to obtain a regional brightness mapping table;

[0062] (5) performing regional optical compensation on the display content through a regional brightness mapping table to obtain a display compensation coefficient; adaptively matching the display compensation coefficient with the ambient light intensity characteristic value to generate a display correction matrix;

[0063] (6) Parameters of each display area are configured according to the display correction matrix, and the display effect is optimized in combination with the display area distribution data to establish regional display parameters.

[0064] Specifically, the eye tracking data is processed. The eye tracking data is collected by the on-board eye sensor and contains information such as the position, speed, and acceleration of the eye movement. Since the eye movement has the characteristics of rapid jitter, the original data contains a lot of noise, which needs to be eliminated through data filtering. The filtering process adopts the Kalman filter algorithm, which effectively removes random noise and mutation signals in the eye movement data through two steps of prediction and update to obtain smooth eye movement trajectory data. The obtained eye movement trajectory data is subjected to time series analysis. The fixed time window method is used to segment the continuous eye movement trajectory according to the time interval of 50 milliseconds, and the coordinates of the gaze points in each time window are extracted to form a sequence of gaze point coordinates. Based on the obtained gaze point coordinate sequence, the line of sight thermal distribution is calculated by the kernel density estimation algorithm. The algorithm takes each gaze point as the center, establishes a two-dimensional Gaussian kernel function, superimposes the distribution of multiple gaze points, and generates a continuous thermal distribution map. The different colors on the thermal map represent the driver's concentration level, red represents a highly concerned area, and blue represents a less concerned area. The obtained attention distribution map is spatially correlated with the display content priority data, and the weighted sum of the attention value and content priority of each area is calculated to generate a regional weight matrix. Each element in the regional weight matrix represents the comprehensive importance of a specific position on the display screen.

[0065] Using the generated regional weight matrix, the display screen is divided by the regional growing algorithm. Starting from the point with the highest weight, it grows outward until it encounters a boundary point with a large weight difference, forming an initial region. For this initial region scheme, the minimum spanning tree algorithm of graph theory is used to optimize the boundary. Each region is regarded as a node in the graph, and the boundary between regions is used as an edge. By minimizing the boundary length and the difference in regional weight, the regional division boundary is optimized, and finally the display region distribution data is formed. The display region distribution data is then combined with the ambient light intensity characteristic value to calculate the required contrast of each display region. According to the contrast perception characteristics of the human eye, the brightness level of the display region is adjusted under different ambient light intensities to obtain the brightness adjustment parameters. Based on these parameters, the dynamic range of each display region is optimized to ensure good visibility under different lighting conditions, and finally a regional brightness mapping table is generated.

[0066] The display content is optically compensated by region through the regional brightness mapping table. The compensation process takes into account the luminous characteristics of the LED display and the ambient light contrast requirements, and the display compensation coefficient is calculated. These compensation coefficients are adaptively matched with the real-time ambient light intensity characteristic values ​​to establish a display correction matrix, which contains the optimal display parameters of each display area under the current ambient light conditions. Finally, according to the display correction matrix, the parameters of each display area are configured, including display parameters such as brightness, contrast, and color saturation. Combined with the display area distribution data obtained previously, the overall display effect is optimized to ensure that the display effects between each area are coordinated and consistent, and the final regional display parameters are established.

[0067] For example, when the driver looks at the on-board LED display screen while driving, the eye movement sensor collects 50 eye movement data points per second. After Kalman filtering, the noise level is reduced from the original ±5 pixels to ±1 pixel. Timing analysis shows that the driver's gaze points are mainly concentrated in the middle and upper areas of the display screen, accounting for 65% of the total gaze time. The heat map generated by kernel density estimation shows that the attention weight of the middle and upper areas is 0.8, while that of the lower areas is only 0.3. Combined with the display content priority data (navigation information priority 0.9, music playback information priority 0.4), the generated regional weight matrix shows that the comprehensive weight of the middle and upper areas reaches 0.85. After regional division, three main display areas are obtained: high priority area (upper middle, accounting for 40% of the screen), medium priority area (lower middle, accounting for 35% of the screen), and low priority area (bottom, accounting for 25% of the screen). When the ambient light intensity is 40,000 lux, the display brightness of the high priority area is set to 2000 cd / m 2 , contrast ratio is 3:1, brightness in medium priority area is 1500cd / m 2 , contrast ratio is 2.5:1, low priority area brightness is 1000cd / m2 , with a contrast ratio of 2:1. This dynamic adjustment and optimization ensures that key information is clearly visible in strong light environments, while appropriately reducing the display intensity of secondary information, which not only ensures the effective transmission of information but also reduces energy consumption.

[0068] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0069] (1) Separate and extract the voice, touch and gesture data from the multimodal interaction data, and obtain the interaction feature sequence through signal processing; perform multi-dimensional correlation analysis on the interaction feature sequence to generate interaction mode features;

[0070] (2) Calculate the matching degree based on the interaction mode characteristics and the driving scene type to obtain the interaction scene adaptation value; perform threshold screening on the interaction scene adaptation value to obtain the interaction mode candidate set;

[0071] (3) Construct a decision tree structure based on the candidate set of interaction patterns to obtain an interaction decision graph; extract the optimal interaction path from the interaction decision graph to form an interaction control strategy;

[0072] (4) Perform collision detection on the interactive control strategy and the regional display parameters to obtain the interactive safety zone; divide the trigger boundary through the interactive safety zone and establish the interactive response rules;

[0073] (5) Performing timing constraint analysis on the interactive response rules to obtain a response timing table; using the response timing table to verify the interactive process and obtain the interactive verification results;

[0074] (6) Compare the interaction verification results with the historical interaction data, screen out the optimal interaction solution, and generate interaction response data.

[0075] Specifically, in the interactive control of the vehicle-mounted LED display, it is necessary to process multimodal interactive data first. Multimodal interactive data includes three types: voice commands, touch operations, and gesture actions. For voice data, the voice framing technology is used to segment the continuous voice signal according to the time window of 20ms, and the audio features including Mel frequency cepstral coefficients, fundamental frequency, sound intensity and other parameters are extracted; for touch data, the touch point coordinates, pressure value, touch duration and other information are recorded; for gesture data, the trajectory, speed, acceleration and other features of the hand movement are extracted. These features are denoised and standardized by signal processing technology to obtain a standardized interactive feature sequence. Subsequently, these feature sequences are subjected to multi-dimensional correlation analysis, the correlation between different features is calculated, the feature combination pattern is identified, and the interactive mode feature is generated. The generated interactive mode feature is matched with the current driving scene type. The matching degree calculation takes into account the relationship between the complexity of the interactive operation and the safety requirements of the driving scene. For example, in a high-speed driving scene, a complex touch operation will get a lower matching score, while a simple voice command will get a higher score. The scene adaptation value of each interaction mode is obtained by calculation, and then these adaptation values ​​are threshold screened and a safety threshold is set. Only interaction modes above the threshold are retained to form a candidate set of interaction modes.

[0076] Based on the obtained candidate set of interaction modes, a decision tree structure is constructed. Each node of the decision tree represents an interaction operation, and the connection between nodes represents the order and dependency of the interaction operations. By traversing all the paths of the decision tree, a complete interaction decision map is formed. In the interaction decision map, each path represents a possible sequence of interaction operations. By evaluating the safety, efficiency and user-friendliness of each path, the optimal interaction path is selected to form an interaction control strategy. The formed interaction control strategy is subjected to collision detection with the regional display parameters to avoid conflicts between the interaction operation and the display content. During the collision detection process, the overlap between the spatial range of the interaction operation and the display area is calculated to determine the safe interaction range and obtain the interaction safety interval. By dividing the boundary of the interaction safety interval, the correspondence between the trigger condition and the response action is established to form the interaction response rule.

[0077] Perform timing constraint analysis on the established interactive response rules to ensure that the timing logic of the interactive operation is reasonable. Timing constraint analysis includes operation response time limit, continuous operation interval requirements, operation priority processing, etc., and generates a response timing table containing detailed timing information. Use the response timing table to verify the entire interactive process, check whether there are timing conflicts or logical errors, and obtain the interactive verification results.

[0078] Finally, the interaction verification results are compared with the historical interaction data. The historical interaction data contains a large number of successful and failed interaction cases. Through comparative analysis, the historical solution that is closest to the current scenario and has a good interaction effect is found. Based on this, the optimal interaction solution is selected to generate the final interaction response data.

[0079] For example: Assume that in a highway scenario (vehicle speed 120km / h), the driver needs to adjust the navigation display. First, three interactive signals are collected: voice command "zoom in the map" (duration 1.2 seconds, audio sampling rate 16kHz), outward stretching gesture (duration 0.8 seconds, hand movement distance 15cm), and two-finger touch zoom operation (duration 0.5 seconds, touch pressure value 200g). After feature extraction, the confidence of the voice command is 0.95, the gesture recognition accuracy is 0.85, and the touch operation effectiveness is 0.9. These features are matched with the highway scene. The voice command obtains an adaptation value of 0.9 (higher than the safety threshold of 0.8), while the gesture and touch operations obtain adaptation values ​​of 0.6 and 0.5 respectively (lower than the threshold). Through decision tree analysis, the interaction path of "voice command-short prompt sound feedback-screen zoom" was selected. Collision detection shows that there is no conflict between the map zoom animation and the navigation information display area (60% area in the center of the screen). Timing analysis determined the timing of the entire interaction process: 0.2 seconds for voice recognition, 0.1 seconds for prompt sound playback, 0.3 seconds for screen zoom animation, and the total response time was controlled within 0.6 seconds. Comparing historical data, it was found that the success rate of this interaction solution in similar scenarios reached 98%, so it was selected as the final solution. The entire interaction process not only ensures driving safety, but also achieves efficient information display adjustment.

[0080] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0081] (1) Dynamically quantifying the display content using the ambient light intensity characteristic value to obtain a display content brightness matrix; performing nonlinear correction on the display content brightness matrix using an optical compensation algorithm to generate brightness compensation parameters;

[0082] (2) Mapping the brightness compensation parameters with the color space data to obtain a color gamut conversion matrix; dynamically expanding the color gamut conversion matrix through a color enhancement algorithm to form display rendering parameters;

[0083] (3) constructing a response timing relationship according to display rendering parameters and interactive response data to obtain a display timing diagram; performing frame synchronization optimization on the display timing diagram to obtain a frame synchronization sequence;

[0084] (4) spatially mapping the frame synchronization sequence to the pixel coordinates to obtain a display mapping table; smoothing the display mapping table through an interpolation algorithm to establish a pixel rendering strategy;

[0085] (5) performing jitter compensation calculation on the pixel rendering strategy to obtain a jitter elimination coefficient; performing display quality evaluation based on the jitter elimination coefficient to generate display quality parameters;

[0086] (6) The display content is finally optimized through display quality parameters, and the display effect is comprehensively evaluated by combining display rendering parameters and interactive response data, and the display effect data is output.

[0087] Specifically, the display content is dynamically quantified and analyzed. Based on the characteristic value of the ambient light intensity, the display content is converted into 8-bit grayscale data to generate an initial brightness matrix. Each element of the matrix represents the brightness value of the corresponding pixel on the display screen, and the value range is 0-255. Considering the nonlinearity of the luminous characteristics of the LED display screen, the brightness matrix is ​​corrected by the optical compensation algorithm. The optical compensation process adopts the gamma correction method, and different gamma values ​​are set for different brightness intervals: the dark area (0-63) uses a gamma value of 2.2, the midtone (64-191) uses a gamma value of 2.0, and the highlight area (192-255) uses a gamma value of 1.8, thereby generating brightness compensation parameters. Then the brightness compensation parameters are mapped to the color space data. The color space data adopts the RGB color model, and each pixel contains color information of three channels: red, green, and blue. The RGB color space is mapped to the device-specific color space through a 3x3 color gamut conversion matrix. On this basis, the color gamut conversion matrix is ​​optimized using an adaptive color enhancement algorithm. The enhancement algorithm includes three steps: first, local adjustment of color saturation is made to improve the color performance of low-saturation areas; then, the hue is fine-tuned to ensure the accurate restoration of key colors; finally, contrast enhancement is performed to expand the dynamic range and generate the final display rendering parameters.

[0088] According to the display rendering parameters and interactive response data, the display timing relationship is established. The display timing needs to consider the timing coordination of the three links of content update, rendering processing and display refresh. Through timing analysis, the display timing diagram is constructed, which includes the processing time, transmission delay and display duration of each display frame. The display timing diagram is optimized for frame synchronization, and adaptive frame rate control is adopted to increase it to 60fps in high-speed scenes and reduce it to 30fps in low-speed scenes. At the same time, the vertical synchronization signal is accurately aligned to avoid screen tearing, and the optimized frame synchronization sequence is obtained. The frame synchronization sequence is spatially mapped with the pixel coordinates on the display screen to establish a pixel-level display mapping relationship. The display mapping table records the display parameters of each pixel at different frame times, including brightness, color and update timing. The display mapping table is smoothed by the bilinear interpolation algorithm to eliminate the sudden changes between pixels and achieve a gradual transition effect. The pixel rendering strategy obtained after interpolation ensures the smooth transition and detail retention of the displayed content.

[0089] The established pixel rendering strategy is used for jitter compensation calculation. First, the jitter area in the picture is detected, and the pixel value changes between adjacent frames are calculated. When the change amplitude exceeds the threshold, the jitter compensation mechanism is activated. Through time domain filtering and spatial domain averaging, the jitter elimination coefficient is calculated, which is used to suppress rapidly changing pixel values. Based on the jitter elimination coefficient, the display quality is evaluated in multiple dimensions, including indicators such as brightness uniformity, color reproduction, and dynamic responsiveness, and display quality parameters are generated. Finally, the display quality parameters, display rendering parameters, and interactive response data are comprehensively integrated to finally optimize the display content. The optimization process includes: detail enhancement, edge sharpening, dynamic range compression, etc., to ensure the best display effect under various ambient light conditions. The display effect evaluation adopts a combination of objective indicators and subjective scoring, and finally outputs display effect data by calculating indicators such as structural similarity and peak signal-to-noise ratio.

[0090] For example, when a vehicle enters a tunnel, the ambient light intensity drops rapidly from 50,000 lux to 500 lux. The display content is a navigation map, which includes elements such as roads, text, and icons. First, the display content is quantitatively analyzed, and the map background brightness is set to 180 (0-255 range), the road line brightness is 220, and the text and icon brightness is 240. Through gamma correction, the corresponding compensation parameters are adjusted to 165, 208, and 235 respectively. After color space mapping, the road is converted from RGB (200, 200, 200) to RGB (190, 195, 205), which improves the color layering. The system increases the display frame rate from 30fps to 60fps, and the inter-frame interval is reduced from 33.3ms to 16.7ms. In the spatial mapping stage, bilinear interpolation of 4x4 pixel blocks is performed on the edge area of ​​the map to make the regional transition more natural. Slight jitter (pixel offset ±2 units) of the image due to vehicle vibration was detected, and the jitter elimination coefficient was calculated to be 0.85, which significantly reduced the jitter phenomenon after application. In the final optimization process, the contrast of road lines increased by 15%, the clarity of text increased by 20%, the structural similarity index of the overall display effect reached 0.92, and the peak signal-to-noise ratio increased to 38dB, ensuring that the navigation information is still clearly discernible in the rapidly changing lighting environment of the tunnel.

[0091] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0092] (1) Converting display effect data into display power consumption data through energy efficiency estimation algorithm; performing correlation analysis on display power consumption data and ambient light change prediction data to obtain power consumption influencing factors;

[0093] (2) Calculate regional energy consumption using the power consumption influencing factors and display area distribution data to obtain a regional power consumption matrix; perform energy consumption balancing on the regional power consumption matrix using an optimization algorithm to generate a power consumption allocation strategy;

[0094] (3) Based on the power consumption allocation strategy, the LED driving circuit is partitioned and controlled to obtain a driving control sequence; the driving control sequence is decomposed in time sequence to obtain pulse modulation parameters;

[0095] (4) constraining the dynamic range of the pulse modulation parameters to obtain a driving voltage threshold; performing graded control on the LED brightness according to the driving voltage threshold to form an energy-saving control parameter;

[0096] (5) performing current modulation calculation on the LED driving circuit through energy-saving control parameters to obtain a driving current waveform; optimizing the driving timing based on the driving current waveform to generate a driving timing table;

[0097] (6) Convert the drive timing table into a digital control signal, adjust the drive circuit in combination with the energy-saving control parameters, and generate a display control signal.

[0098] Specifically, the display effect data is processed through the energy efficiency estimation algorithm. The power consumption per unit area is calculated for the brightness, contrast and refresh rate of different display contents. For example, the power density of a full brightness white display is 80mW / cm 2 , while text content with a black background only requires 20mW / cm 2 . The display power consumption data is correlated with the ambient light change prediction data, and the corresponding relationship between illumination and power consumption is obtained through regression analysis to generate power consumption influencing factors. These factors reflect the degree of influence of ambient light intensity on display power consumption. For example, the display brightness needs to be increased under strong light, resulting in increased power consumption. Based on the power consumption influencing factors, the partition energy consumption is calculated in combination with the display area distribution data. The display screen is divided into multiple independently controlled areas, and each area is assigned different power consumption limits according to the importance of the display content and visual needs. The regional power consumption matrix is ​​obtained by calculation, and each element in the matrix represents the target power consumption value of the corresponding area. The regional power consumption matrix is ​​optimized using a dynamic programming algorithm to achieve energy consumption balance between areas while ensuring the display effect, and generate the optimal power consumption allocation strategy.

[0099] Based on the power consumption allocation strategy, the LED drive circuit is zoned and controlled. The display screen is divided into several independently driven areas, and each area is configured with independent drive circuit parameters. According to the power consumption requirements of the area, a drive control sequence containing drive voltage, current and timing information is generated. The sequence is decomposed in time to determine the duty cycle, frequency and other parameters of PWM modulation, and obtain detailed pulse modulation parameters. Dynamic range constraints are set for pulse modulation parameters to ensure that the drive voltage varies within a safe range. For example, the range of LED drive voltage variation is limited to between 2.8V and 3.3V to avoid display abnormalities caused by overvoltage or undervoltage. According to these voltage thresholds, the LED brightness is divided into multiple levels, such as 100%, 75%, 50%, and 25%, corresponding to different drive parameters to form an energy-saving control parameter set.

[0100] The LED drive circuit is precisely current modulated through energy-saving control parameters. According to the requirements of different brightness levels, the corresponding drive current waveform is calculated. For example, the current is set to 20mA when working at full brightness, and it is reduced to 15mA at 75% brightness. The smooth transition of different brightness levels is achieved through PWM modulation. Based on these drive current waveforms, the drive timing is optimized to generate a drive timing table containing detailed timing information. Finally, the drive timing table is converted into a digital control signal, including PWM waveform, enable signal and synchronization clock. Combined with the energy-saving control parameters, the various parameters of the LED drive circuit are adjusted in real time, and finally a digital control signal for controlling the display screen is generated.

[0101] For example, when a vehicle enters a dark tunnel (ambient light 500 lux) from bright outdoor light, the display needs to quickly adjust the display parameters. The power consumption of the main display area (accounting for 60% of the total area) of the display originally working under strong light is 48mW / cm 2 The secondary area (accounting for 40%) has a power consumption of 36mW / cm 2 After entering the tunnel, the energy efficiency estimation algorithm recalculates the power consumption requirements based on the new ambient light data, and the power consumption impact factor is reduced from 1.0 to 0.4. After regional power consumption optimization, the power consumption of the main display area is reduced to 20mW / cm 2 , the secondary area drops to 15mW / cm 2 The drive control adopts 400Hz PWM modulation, and the duty cycle of the main area is reduced from 80% to 40%, and the secondary area is reduced from 60% to 30%. The driving voltage is reduced from 3.2V to 2.9V, and the LED driving current is reduced from 18mA to 12mA. Through the dynamic adjustment of these parameters, the total power consumption of the display is reduced by 55%, while still maintaining a good display effect, the brightness uniformity is maintained at more than 90%, and the display response time is controlled within 16.7ms, ensuring the clear visibility and visual comfort of the displayed content.

[0102] The display method for the vehicle-mounted LED display screen in the embodiment of the present application is described above. The display system for the vehicle-mounted LED display screen in the embodiment of the present application is described below. Figure 2 In the embodiment of the present application, a display system for a vehicle-mounted LED display screen includes:

[0103] A computing module is used to perform multi-dimensional data fusion computing based on the collected ambient light intensity data, obtain ambient light intensity characteristic values, complete trend prediction analysis based on the ambient light intensity characteristic values, and form ambient light change prediction data;

[0104] An analysis module, configured to implement scene recognition operations based on real-time traffic data, generate driving scene types, perform priority analysis on the driving scene types and the ambient light change prediction data, and output display content priority data;

[0105] a calculation module, configured to perform spatial layout calculation in combination with the eye tracking data and the display content priority data to form display area distribution data, determine display parameters using the display area distribution data and the ambient light intensity characteristic value, and establish regional display parameters;

[0106] An interaction module, configured to perform interaction mode analysis on the multimodal interaction data and the driving scenario type, determine an interaction control strategy, perform safety verification on the interaction control strategy and the regional display parameters, and generate interaction response data;

[0107] An establishment module is used to dynamically adjust the display content based on the ambient light intensity characteristic value to form display rendering parameters, establish a display mapping relationship through the display rendering parameters and the interactive response data, and output display effect data;

[0108] The control module is used to establish an energy consumption estimation model by integrating the display effect data, the ambient light change prediction data and the display area distribution data, generate energy-saving control parameters, apply the energy-saving control parameters to adjust the LED drive circuit, and generate a display control signal.

[0109] Through the collaborative cooperation of the above components, by collecting ambient light intensity data and performing multi-dimensional data fusion operations, combined with trend prediction analysis to form ambient light change prediction data, the forward-looking adjustment of display parameters is realized, avoiding display delays and flickering problems caused by light changes in traditional solutions; by performing scene recognition operations on real-time road condition data and generating driving scene types, combined with ambient light change prediction data for priority analysis, intelligent grading of display content is realized, ensuring the timely presentation of key information; by combining eye tracking data and display content priority data for spatial layout calculations, display area distribution data is formed and regional display parameters are established, dynamic optimization layout of display content is realized, and the efficiency and accuracy of information transmission are improved; by analyzing multimodal interaction data and driving scene The scene type is analyzed, the interactive control strategy is determined and safety verification is performed, and the interactive response data is generated, providing a safer and more intuitive human-computer interaction method; the display content is dynamically adjusted based on the ambient light intensity characteristic value, and a display mapping relationship is established in combination with the display rendering parameters and the interactive response data, thereby achieving real-time optimization of the display effect; finally, an energy consumption estimation model is established by integrating the display effect data, the ambient light change prediction data and the display area distribution data, and the energy-saving control parameters are generated and the LED drive circuit is adjusted, thereby achieving efficient energy utilization while ensuring the display effect. Compared with the traditional solution, it can reduce energy consumption by more than 30%, shorten the display response time by 50%, and improve the readability of the displayed content by 40%, providing the driver with a clearer, more stable and energy-saving visual interaction experience.

[0110] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A display method for a vehicle-mounted LED display screen, characterized in that: The display method for the vehicle-mounted LED display screen includes: Perform multi-dimensional data fusion operation based on the collected ambient light intensity data to obtain ambient light intensity characteristic values, complete trend prediction analysis based on the ambient light intensity characteristic values, and form ambient light change prediction data; Implementing scene recognition calculation based on real-time road condition data, generating driving scene types, performing priority analysis on the driving scene types and the ambient light change prediction data, and outputting display content priority data; Perform spatial layout calculation in combination with the eye tracking data and the display content priority data to form display area distribution data, determine display parameters using the display area distribution data and the ambient light intensity characteristic value, and establish regional display parameters; Performing interaction mode analysis on the multimodal interaction data and the driving scenario type, determining an interaction control strategy, performing safety verification on the interaction control strategy and the regional display parameters, and generating interaction response data; Dynamically adjusting the display content based on the ambient light intensity characteristic value to form display rendering parameters, establishing a display mapping relationship through the display rendering parameters and the interactive response data, and outputting display effect data; An energy consumption estimation model is established by integrating the display effect data, the ambient light change prediction data and the display area distribution data, and energy-saving control parameters are generated. The energy-saving control parameters are applied to adjust the LED driving circuit to generate a display control signal.

2. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The method of performing a multi-dimensional data fusion operation based on the collected ambient light intensity data to obtain an ambient light intensity characteristic value, completing a trend prediction analysis based on the ambient light intensity characteristic value, and forming ambient light change prediction data includes: Synchronously collecting raw data of light intensity through multiple ambient light sensors, performing a weighted sum operation on the raw data of light intensity to obtain weighted light data; A polynomial regression equation is established based on the weighted illumination data and parameters such as vehicle speed and time, an illumination data coefficient matrix is ​​calculated by the least square method, and an ambient light intensity characteristic value is generated from the illumination data coefficient matrix; A time series convolutional network is used to extract periodic features from historical ambient light intensity feature values, and a light change trend factor is obtained by combining time series analysis. A light prediction vector is constructed based on the light change trend factor. Performing probability distribution calculation on the illumination prediction vector and the real-time ambient light intensity characteristic value through Bayesian reasoning to obtain illumination prediction probability distribution; Performing data calibration according to the illumination prediction probability distribution and the driver's personalized settings to generate an illumination correction coefficient, and adjusting the illumination prediction probability distribution by the illumination correction coefficient; The adjusted illumination prediction probability distribution is reconstructed into a time series, and key change points are extracted through sliding window analysis to form ambient light change prediction data.

3. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The method of implementing scene recognition operation based on real-time road condition data, generating driving scene types, performing priority analysis on the driving scene types and the ambient light change prediction data, and outputting display content priority data includes: Performing data dimension reduction processing on real-time traffic data through a feature extractor to obtain a traffic feature vector; performing principal component analysis based on the traffic feature vector to obtain key traffic features; Classify the key road condition features through a neural network to obtain a scene category probability value; perform similarity matching between the scene category probability value and historical scene data to generate a driving scene type; Using the ambient light change prediction data to perform weight calculation on the driving scene types to obtain a scene priority matrix; and constructing an information display level table based on the scene priority matrix; Extracting display content categories in different scenarios from the information display level table, performing dynamic threshold adjustment in combination with vehicle speed data, and obtaining display threshold parameters; Screening and grading the display content by the display threshold parameter to obtain a graded display sequence; optimizing the content combination according to the graded display sequence to generate priority sorting data; The priority sorting data is matched with the ambient light change prediction data for display strategy, the display content is prioritized, and display content priority data is output.

4. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The combining of the eye tracking data and the display content priority data to perform spatial layout calculation to form display area distribution data, determining display parameters using the display area distribution data and the ambient light intensity characteristic value, and establishing regional display parameters includes: Eliminate noise from the eye tracking data by data filtering to obtain eye trajectory data; perform time series analysis on the eye trajectory data to extract a gaze point coordinate sequence; Calculate the sight heat distribution based on the gaze point coordinate sequence to obtain an attention distribution map; perform spatial correlation analysis on the attention distribution map and the display content priority data to generate a regional weight matrix; Using the regional weight matrix to perform regional division operation on the display screen to obtain an initial regional plan; optimizing the boundaries of the initial regional plan through a graph theory algorithm to form display area distribution data; Performing contrast calculation on the display area distribution data and the ambient light intensity characteristic value to obtain a brightness adjustment parameter; performing dynamic range optimization on each display area based on the brightness adjustment parameter to obtain a regional brightness mapping table; Performing regional optical compensation on the display content by using the regional brightness mapping table to obtain a display compensation coefficient; adaptively matching the display compensation coefficient with the ambient light intensity characteristic value to generate a display correction matrix; Parameters of each display area are configured according to the display correction matrix, and the display effect is optimized in combination with the display area distribution data to establish regional display parameters.

5. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The performing of interaction mode analysis on the multimodal interaction data and the driving scenario type, determining an interaction control strategy, performing safety verification on the interaction control strategy and the regional display parameters, and generating interaction response data includes: Separate and extract the voice, touch and gesture data in the multimodal interaction data, and obtain an interaction feature sequence through signal processing; perform multi-dimensional correlation analysis on the interaction feature sequence to generate interaction mode features; Calculating the matching degree between the interaction mode feature and the driving scene type to obtain an interaction scene adaptation value; performing threshold screening on the interaction scene adaptation value to obtain an interaction mode candidate set; Constructing a decision tree structure according to the interaction mode candidate set to obtain an interaction decision map; extracting an optimal interaction path from the interaction decision map to form an interaction control strategy; Perform collision detection on the interaction control strategy and the area display parameter to obtain an interaction safety interval; divide the trigger boundary by the interaction safety interval to establish an interaction response rule; Performing timing constraint analysis on the interactive response rules to obtain a response timing table; performing interactive process verification using the response timing table to obtain an interactive verification result; The interaction verification result is compared with the historical interaction data to screen out the optimal interaction solution and generate interaction response data.

6. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The dynamically adjusting the display content based on the ambient light intensity characteristic value to form display rendering parameters, establishing a display mapping relationship through the display rendering parameters and the interactive response data, and outputting display effect data, includes: Performing dynamic quantitative analysis on the display content with the ambient light intensity characteristic value to obtain a display content brightness matrix; performing nonlinear correction on the display content brightness matrix with an optical compensation algorithm to generate brightness compensation parameters; The brightness compensation parameter is mapped to the color space data to obtain a color gamut conversion matrix; the color gamut conversion matrix is ​​dynamically expanded by a color enhancement algorithm to form a display rendering parameter; Constructing a response timing relationship according to the display rendering parameters and the interactive response data to obtain a display timing diagram; performing frame synchronization optimization on the display timing diagram to obtain a frame synchronization sequence; The frame synchronization sequence is spatially mapped with the pixel point coordinates to obtain a display mapping table; the display mapping table is smoothed by an interpolation algorithm to establish a pixel rendering strategy; Performing jitter compensation calculation on the pixel rendering strategy to obtain a jitter elimination coefficient; performing display quality evaluation based on the jitter elimination coefficient to generate a display quality parameter; The display content is finally optimized by using the display quality parameters, the display effect is comprehensively evaluated by combining the display rendering parameters with the interactive response data, and the display effect data is output.

7. The display method for a vehicle-mounted LED display screen according to claim 1, characterized in that: The step of establishing an energy consumption estimation model by integrating the display effect data, the ambient light change prediction data and the display area distribution data, generating energy-saving control parameters, and applying the energy-saving control parameters to adjust the LED driving circuit to generate a display control signal includes: The display effect data is converted into display power consumption data by using an energy efficiency estimation algorithm; a correlation analysis is performed on the display power consumption data and the ambient light change prediction data to obtain a power consumption influencing factor; Perform regional energy consumption calculation on the power consumption influencing factors and the display area distribution data to obtain a regional power consumption matrix; perform energy consumption balancing processing on the regional power consumption matrix through an optimization algorithm to generate a power consumption allocation strategy; Based on the power consumption allocation strategy, a partition control plan is performed on the LED driving circuit to obtain a driving control sequence; the driving control sequence is decomposed in time sequence to obtain pulse modulation parameters; Dynamic range constraints are imposed on the pulse modulation parameters to obtain a driving voltage threshold; LED brightness is graded and controlled according to the driving voltage threshold to form energy-saving control parameters; Performing current modulation calculation on the LED driving circuit according to the energy-saving control parameter to obtain a driving current waveform; optimizing the driving timing based on the driving current waveform to generate a driving timing table; The driving timing table is converted into a digital control signal, and the driving circuit is adjusted in combination with the energy-saving control parameter to generate a display control signal.

8. A display system for a vehicle-mounted LED display screen, used to implement the display method for a vehicle-mounted LED display screen as described in any one of claims 1 to 7, characterized in that: The display system for the vehicle-mounted LED display screen comprises: A computing module is used to perform multi-dimensional data fusion computing based on the collected ambient light intensity data, obtain ambient light intensity characteristic values, complete trend prediction analysis according to the ambient light intensity characteristic values, and form ambient light change prediction data; An analysis module, configured to implement scene recognition operations based on real-time traffic data, generate driving scene types, perform priority analysis on the driving scene types and the ambient light change prediction data, and output display content priority data; a calculation module, configured to perform spatial layout calculation in combination with the eye tracking data and the display content priority data to form display area distribution data, determine display parameters using the display area distribution data and the ambient light intensity characteristic value, and establish regional display parameters; An interaction module, configured to perform interaction mode analysis on the multimodal interaction data and the driving scenario type, determine an interaction control strategy, perform safety verification on the interaction control strategy and the regional display parameters, and generate interaction response data; An establishment module is used to dynamically adjust the display content based on the ambient light intensity characteristic value to form display rendering parameters, establish a display mapping relationship through the display rendering parameters and the interactive response data, and output display effect data; The control module is used to establish an energy consumption estimation model by integrating the display effect data, the ambient light change prediction data and the display area distribution data, generate energy-saving control parameters, apply the energy-saving control parameters to adjust the LED drive circuit, and generate a display control signal.

Images