Vehicle-mounted display loading method and system

Abstract

The invention relates to a vehicle-mounted display loading method and system, and belongs to the technical field of vehicle-mounted display.The vehicle-mounted display loading method comprises the steps that a to-be-loaded texture resource set is obtained, and texture resources and preset interface positions of the texture resources are stored in the texture resource set; for each texture resource, determining a visual focus weight of the texture resource based on a pre-constructed visual focus prediction model according to a preset interface position configured to be rendered; according to the visual focus weight of each texture resource, determining a corresponding target texture quality grade for the texture resource; obtaining the current texture quality grade of each texture resource, and calculating the loading priority of the texture resources based on the visual focus weight of each texture resource, the target texture quality grade and the current texture quality grade; and determining a loading sequence of the texture resources according to the loading priorities, and loading the corresponding texture resources according to the loading sequence and the texture quality corresponding to the target texture quality grade.

Description

Technical Field

[0001] This application relates to the technical field of vehicle-mounted displays, and more particularly to a method and system for loading vehicle-mounted displays. Background Technology

[0002] In automotive display systems, especially AVM (Around View Monitor) systems, to provide a high-quality, highly realistic graphical interface and interactive experience, the system typically needs to call and render a large number of texture resources, such as 3D vehicle models, surround view background images, function icons, dynamic indicators, and text labels. Currently, the industry practice is to read, decode, and load all required texture resources from storage devices into system memory and video memory all at once during system initialization, so that they can be directly called during subsequent graphics rendering. While this full pre-loading mode ensures the immediate availability of interface elements, it significantly increases the processing burden during system startup, and its limitations are becoming increasingly apparent, especially in automotive embedded environments where real-time performance and resource efficiency are critical.

[0003] The aforementioned full-load method has the following limitations. First, the concentrated execution of reading, decoding, and uploading operations for a large number of texture resources during system startup instantly consumes extremely high CPU and GPU computing resources, leading to system response delays, extended startup time, and impacting the immediacy of user operations. Second, the hardware resources of automotive embedded platforms are relatively limited. Simultaneously handling multiple high-load tasks during the initialization phase can easily cause resource contention and system lag, and in severe cases, may even affect the normal rendering and display of safety-related functions. Furthermore, because the loading process is completely independent of the user's actual visual focus and driving scenario, a large number of texture resources in areas not currently of interest are loaded with equal priority, resulting in unnecessary resource waste and failing to achieve an effective match between resource allocation and visual perception value. Overall system efficiency and user experience need improvement. Summary of the Invention

[0004] This application addresses, to at least some extent, one of the technical problems in the related art.

[0005] Therefore, this application aims to provide a method and system for loading vehicle-mounted displays.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for loading an in-vehicle display, comprising: Obtain a set of texture resources to be loaded, wherein the set of texture resources stores texture resources and their preset interface positions; For each texture resource, based on the preset interface position where it is configured to be rendered, the visual focus weight of the texture resource is determined according to a pre-built visual focus prediction model. Based on the visual focus weight of each texture resource, a corresponding target texture quality level is determined for the texture resource; Obtain the current texture quality level of each texture resource, and calculate the loading priority of the texture resource based on the visual focus weight, the target texture quality level, and the current texture quality level of each texture resource; The loading order of each texture resource is determined according to the loading priority, and the corresponding texture resource is loaded according to the loading order and the texture quality corresponding to the target texture quality level.

[0007] In this technical solution, a visual focus prediction model is introduced to dynamically determine the visual focus weight of texture resources based on their preset interface positions. This weight is then used to further determine the texture quality level and loading priority, ultimately achieving differentiated loading of texture resources. The loading priority is calculated by comprehensively considering the visual focus weight, the target texture quality level, and the current texture quality level reflecting the resource's current state. This loading priority reflects both the visual importance of the texture resource and the urgency of quality improvement, thus determining the order in which texture resources are loaded. Finally, loading is performed according to this priority order and its corresponding target quality level, ensuring that limited system memory, bandwidth, and computing resources are preferentially allocated to texture resources with high visual importance and urgent need for quality improvement. This method effectively overcomes the resource contention problem caused by existing full-load uniform loading methods, enabling dynamic and orderly resource scheduling during system initialization and operation. This reduces display latency in critical areas, improves system response efficiency, and optimizes the overall visual experience under resource-constrained conditions.

[0008] In some embodiments of this application, determining the visual focus weight of the texture resource based on a pre-built visual focus prediction model includes: Based on the preset interface position, determine the basic display weight of the preset interface position; Acquire real-time vehicle status data, and calculate a status adjustment factor based on the real-time vehicle status data; Obtain the system runtime and calculate the time decay factor based on the system runtime; The visual focus weight of the texture resource is calculated based on the state adjustment factor, the time decay factor, and the basic display weight.

[0009] In this technical solution, a more accurate and dynamic visual focus weight is calculated by comprehensively considering the basic display weight of the preset interface position, the state adjustment factor determined by the vehicle's real-time status data, and the time decay factor related to the system's running time. This method not only considers the static influence of interface layout and human visual characteristics but also introduces dynamic adjustments based on the driving scenario and time dimension. This allows the visual focus prediction to adapt to changes in vehicle status and system operation stages in real time, significantly improving the scenario fit and prediction accuracy of the weight calculation. This provides a more reliable and adaptive decision-making basis for subsequent differentiated loading.

[0010] In some embodiments of this application, a gaze point optimization step is also included, comprising: Acquire real-time gaze information of the driver, collected by the vehicle's vision sensors; Calculate the attention enhancement factor based on the real-time gaze point information; The visual focus weight of the texture resource is adjusted according to the attention enhancement factor.

[0011] In this technical solution, an onboard vision sensor is introduced to capture the driver's gaze point in real time. Based on this, an attention enhancement factor is generated to dynamically adjust the pre-calculated visual focus weights, achieving real-time collaboration and closed-loop optimization between the prediction model and the driver's true intentions. This effectively compensates for the potential biases that may exist when relying solely on preset rules and vehicle status for static or semi-static predictions, enabling system resources to be more accurately and agilely directed towards the screen area that the driver is currently focusing on.

[0012] In some embodiments of this application, the basic display weight for determining the preset interface position includes: Based on the layout features, visual characteristics, and information distribution features of the preset interface position in the display interface, the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor of the preset interface position are calculated. The basic display weight is obtained by weighting and summing the interface layout weight factor, the visual center weight factor, the information density weight factor, and the reading habit weight factor.

[0013] In this technical solution, the basic display weights of preset interface locations are decomposed and quantified into weight factors across four specific dimensions: interface layout, visual center, information density, and reading habits. These weighted factors are then fused to achieve a refined and structured model of the visual importance of static interfaces. This method comprehensively considers multiple factors, including interface layout design principles, human eye physiological characteristics, information functional value, and user cognitive habits. This makes the calculation of basic weights not only more comprehensive and objective but also more interpretable and adjustable. The resulting basic display weights can more accurately and scientifically reflect the inherent visual importance of different interface areas under normal conditions. This provides a reliable and high-precision initial benchmark for subsequent weight adjustments based on dynamic states and time factors, thereby fundamentally improving the accuracy and reliability of the entire visual focus prediction model.

[0014] In some embodiments of this application, the step of acquiring real-time vehicle status data and calculating a status adjustment factor based on the real-time vehicle status data includes: The current vehicle status is determined based on the real-time vehicle status data. Based on the preset adjustment mapping relationship between vehicle status and interface position, the status adjustment factor corresponding to the preset interface position is determined.

[0015] In this technical solution, the current driving scenario is determined based on the vehicle's real-time status, and the state adjustment factor for the corresponding area is quickly determined based on a preset adjustment mapping relationship between vehicle status and interface position, achieving intelligent linkage between visual focus weight and driving scenario. This method enables resource loading strategies to dynamically match the driver's actual focus in different states, thereby ensuring that critical driving information always receives higher loading priority and rendering quality. This not only significantly improves the system's adaptability and practicality under different operating conditions, but also ensures the real-time and deterministic nature of state responses through a preset rule mapping mechanism, enhancing driving safety. Furthermore, it eliminates the need for complex calculations, facilitating efficient and stable operation in in-vehicle embedded environments.

[0016] In some embodiments of this application, determining the corresponding target texture quality level for the texture resource includes: The visual focus weight of the texture resource is compared with a preset quality threshold and a second quality threshold; If the visual focus weight is greater than the first quality threshold, then the target texture quality level is determined to be a high quality level; If the visual focus weight is less than or equal to the second quality threshold, then the target texture quality level is determined to be low quality. If the visual focus weight is between the first quality threshold and the second quality threshold, then the target texture quality level is determined to be a medium quality level.

[0017] In this technical solution, the visual focus weight of texture resources is quantified and compared with preset first and second quality thresholds, thereby classifying them into three distinct texture quality levels: high, medium, and low. This threshold-based grading mechanism allows the system to allocate appropriate texture quality to areas of different importance based on their weight values, achieving refined resource management. On the one hand, it ensures that high-weight areas receive high-quality texture loading, guaranteeing the display effect of critical content and user experience; on the other hand, it proactively reduces the texture quality of low-weight areas, effectively saving memory usage and loading time, and avoiding resource waste.

[0018] In some embodiments of this application, a dynamic quality adjustment step is also included, comprising: Real-time utilization rate of hardware resources within the monitoring system; Based on the real-time utilization rate of the aforementioned hardware resources, the overall system load index is calculated. The load status determination result is obtained by comparing the system's comprehensive load index with a preset load threshold. Based on the load status determination result, the current texture quality level of the texture resource is adjusted.

[0019] The technical solution incorporates a dynamic quality adjustment step, enabling the system to flexibly respond to the current operating load based on real-time monitoring of hardware resource usage, further optimizing the quality level configuration of texture resources. When the system is under high load, the quality level of some non-critical texture resources can be promptly reduced to free up processor or graphics unit resources, ensuring the rendering stability and real-time performance of critical visual information. Conversely, when system resources are sufficient, texture quality can be appropriately increased to enhance visual performance. This mechanism gives the texture loading strategy adaptive and resource-aware capabilities, improving not only the overall system efficiency and stability but also the responsiveness and visual consistency of the in-vehicle display interface under different operating conditions.

[0020] In some embodiments of this application, the calculation of the visual center weight factor for the preset interface position includes: Based on the relative positional relationship between the preset interface position and the visual center area of ​​the display interface, the visual center weight factor is calculated using a preset distance attenuation function.

[0021] In this technical solution, by utilizing the relative positional relationship between a preset interface location and the visual center area, and combining this with a preset distance attenuation function to calculate the visual center weight factor, the system can quantify the attention given to each area of ​​the interface relative to the user's visual focus center. This approach introduces a spatial attenuation mechanism into the visual focus weight evaluation, helping to highlight the importance of the central area of ​​the display interface and ensuring that high-quality texture resources are prioritized for the areas most likely to attract user attention, thereby improving the perceptual efficiency of interface content presentation and the user's interactive experience. This method not only enhances the human-engineering adaptability of the loading strategy but also provides a calculable and adjustable basis for prioritizing texture resource allocation.

[0022] In some embodiments of this application, the calculation of the information density weighting factor for the preset interface position includes: The display interface is divided into multiple sub-display areas; Determine the target sub-display area to which the preset interface position belongs; Based on the number and type weight of interface elements in the target sub-display area, the interface element density of the target sub-display area is calculated. Based on the density of the interface elements, determine the information density weighting factor of the preset interface position.

[0023] In this technical solution, the display interface is divided into multiple sub-display areas. By combining the number and type weights of interface elements within each sub-area, the density of interface elements is calculated, thereby determining the information density weighting factor. This allows the system to identify areas where information content is relatively concentrated within the interface. This method not only improves the accuracy of recognizing the display value of texture resources but also enables the loading strategy to prioritize areas with high information carrying capacity, thus enhancing the information delivery effect and visual focus of the interface. Through this quantitative evaluation method based on information density, the system can rationally allocate loading priorities and optimize the organization and presentation logic of image information even with limited resources.

[0024] Secondly, this application provides an in-vehicle display loading system for implementing the in-vehicle display loading method as described in the first aspect.

[0025] As can be seen from the above technical solutions, additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0026] Figure 1 This is a schematic flowchart of an in-vehicle display loading method according to an embodiment of this application; Figure 2 This is a schematic diagram of the visual focus weight calculation process according to the embodiments of this application; Figure 3 This is a schematic diagram of the basic display weight calculation process according to the embodiments of this application; Figure 4 This is a schematic diagram of the state adjustment factor calculation process according to the embodiments of this application; Figure 5 This is a schematic flowchart of the fixation point optimization steps according to the embodiments of this application; Figure 6 This is a schematic diagram of the dynamic quality adjustment steps according to the embodiments of this application; Figure 7 This is a schematic diagram of a computer device according to an embodiment of this application. Detailed Implementation

[0027] The present application will now be described in detail through exemplary embodiments. However, it should be understood that, without further description, elements, structures, and features in one embodiment may be advantageously incorporated into other embodiments. It's worth noting that in the automotive field, with the rapid development of intelligent driving and human-machine interaction technologies, in-vehicle display systems are gradually evolving into highly integrated information display and interaction platforms. These systems are required not only to possess clear images and rapid response interface presentation capabilities, but also to ensure operational stability and real-time performance in resource-constrained embedded environments. Especially in typical applications such as AVM (Around View Monitor), central control displays, and driver assistance interfaces, the system often needs to dynamically render a large amount of complex graphical content to support the driver's real-time perception and operation of the vehicle's surrounding environment, functional status, and navigation information.

[0028] Several key terms are involved in the above application process, including: AVM (Around View Monitor) acquires images from multiple cameras around the vehicle, generates a bird's-eye view of the vehicle's surroundings, and displays it on the in-vehicle screen. This helps the driver to make precise operations at low speeds or while parking. It has extremely high requirements for the real-time performance and clarity of texture resource loading and graphics rendering.

[0029] OpenGL (Open Graphics Library) is a widely used graphics rendering programming interface that allows developers to control the image drawing process in automotive systems. It is used for loading, managing, and visualizing texture resources, supporting the underlying implementation of the entire graphics display module.

[0030] In an in-vehicle system, the CPU (Central Processing Unit) is responsible for performing logical control tasks such as system initialization, data parsing, and scheduling. During the startup phase, the reading and decoding of a large number of texture resources mainly rely on the CPU, so its computational load has a significant impact on the system response speed.

[0031] GPU (Graphics Processing Unit) is dedicated to highly parallel computing tasks. It undertakes the main computing tasks in graphics rendering, such as texture mapping, lighting processing, and graphics transformation. It is a key component for improving interface rendering performance and real-time interactive response capabilities.

[0032] Texture resources refer to image data used to give detail to graphic objects such as 3D models, icons, and backgrounds, and are the core materials for the visual presentation of in-vehicle interfaces. The quantity and quality of texture resources directly affect the display effect and loading efficiency.

[0033] In existing technologies, to ensure a fast response of the interface after system startup, a full preloading strategy is usually adopted, that is, loading all possible texture resources at once during the system startup phase.

[0034] While the full preloading strategy can avoid runtime resource shortages, it also brings many problems. Loading a large amount of resources at once will put the central processing unit and graphics processing unit under tremendous load in a short period of time, resulting in slow system startup, UI lag, and even affecting the timely rendering of critical graphics content.

[0035] Meanwhile, because the full preloading strategy does not differentiate between the user's current visual focus or driving scenario, many temporarily unnecessary texture resources are also loaded equally, resulting in a waste of memory and video memory resources. In automotive systems with limited embedded hardware resources, this resource scheduling mode is prone to causing system instability, reducing interface rendering efficiency and user experience, and hindering further improvements in graphics processing performance.

[0036] In the following, embodiments of this application will be described in detail with reference to the accompanying drawings.

[0037] As attached Figure 1 As shown, the in-vehicle display loading method includes the following steps S1-S5.

[0038] S1: Get the set of texture resources to be loaded. The texture resource set stores the texture resources and their preset interface positions.

[0039] Preferably, after the vehicle system starts up, the AVM system first completes the environment initialization, including powering on the system, starting the underlying real-time operating system, and initializing the graphics rendering environment to support subsequent texture loading and graphics drawing tasks.

[0040] Furthermore, the system will automatically read preset configuration files, such as texture resource configuration lists, UI layout configuration files, and vehicle model parameter files. These preset configuration files constitute the basic information structure for texture resource scheduling.

[0041] Texture resource configuration lists are typically described in structured formats such as JSON or XML. They record detailed information about all available texture resources, including the storage path of the texture image files, default resolution, compression encoding format, and the functional module to which they belong, providing fundamental support for resource classification, management, and parsing.

[0042] The UI layout configuration file defines the position, size, hierarchy, and corresponding texture resource identifiers of UI elements in various display interfaces, enabling the system to establish a correspondence between texture resources and interface structure at the logical level.

[0043] The vehicle model parameter file provides personalized parameter information related to specific vehicle models, such as camera installation angle, radar distribution scheme, and interface theme color scheme, to ensure that texture resources can be reasonably adapted and selected according to vehicle model differences.

[0044] Furthermore, after the system completes the parsing of the configuration file, it can construct a structured collection of texture resources based on this information.

[0045] Each texture resource in the texture resource set is clearly marked with its preset interface position. This position is generally defined by a rectangular display area, bounding box, or key point coordinates in the screen coordinate system to ensure that the subsequent visual focus weight evaluation can obtain an accurate spatial reference.

[0046] Once the texture resource set is constructed, it will be loaded into memory as the input basis for visual focus prediction and differential loading processes, thereby achieving more accurate and efficient graphics resource management and display optimization.

[0047] S2: For each texture resource, determine the visual focus weight of the texture resource based on the preset interface position configured for rendering and a pre-built visual focus prediction model.

[0048] Preferably, the visual focus prediction model is used to quantitatively evaluate the degree to which different areas of the in-vehicle display interface attract the driver's visual attention. The principle behind the construction of the visual focus prediction model is that the driver's attention to different areas on the in-vehicle display screen is not uniformly distributed during driving, but is influenced by a combination of factors.

[0049] The visual focus prediction model dynamically assigns a numerical visual focus weight to each position by analyzing the preset interface positions where texture resources are configured for rendering.

[0050] The visual focus weight represents the relative visual importance of the content corresponding to the preset interface position in the current context. The higher the weight, the more likely the area is to become the driver's visual focus, and therefore it should have a higher priority in the allocation of system resources.

[0051] It should be noted that the visual focus prediction model protected in this application is not limited to any specific implementation method, but covers any technical means that can evaluate the visual importance of texture resources based on a preset interface position. For example, the visual focus prediction model can comprehensively consider one or more factors such as the static attributes of the position, dynamic scene information, or the driver's real-time physiological signals, and perform calculations through predefined rules or algorithms.

[0052] By introducing a visual focus prediction model, the loading strategy for texture resources can be transformed from a fully indiscriminate loading strategy to a differentiated loading strategy based on visual importance.

[0053] Specifically, firstly, a target texture quality level is assigned to each texture resource based on the visual focus weight, while simultaneously obtaining its current texture quality level. Then, by combining the visual focus weight, the target quality level, and the current quality level, the loading priority of each texture resource is calculated.

[0054] Loading priority reflects both the visual importance of a region and the urgency of quality improvement. Ultimately, texture resources are loaded according to the loading priority, at the quality level corresponding to the target level. This resolves resource contention and latency issues caused by full loading under resource constraints, achieving an optimal balance between system startup speed, smooth operation, and visual quality in critical areas.

[0055] The following specific examples further illustrate the construction of the visual focus prediction model and the calculation process of the visual focus weight.

[0056] In some embodiments, such as Figure 2 As shown, determining the visual focus weight of texture resources based on a pre-built visual focus prediction model includes the following steps S21-S24.

[0057] S21: Determine the basic display weight of the preset interface position based on the preset interface position.

[0058] In some embodiments, such as Figure 3 As shown, determining the basic display weight of the preset interface position includes the following steps S211-S212.

[0059] S211: Based on the layout features, visual characteristics, and information distribution features of the preset interface position in the display interface, calculate the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor of the preset interface position.

[0060] The calculation methods for the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor are explained in detail below.

[0061] Preferably, the interface layout weight factor is calculated as follows.

[0062] Furthermore, the interface layout weight factor is calculated based on the golden ratio aesthetic principle and a two-dimensional Gaussian distribution model, specifically manifested as the golden ratio factor. The calculation of the golden ratio factor aims to quantify the inherent visual appeal of the screen area due to its conformity to the golden ratio.

[0063] Specifically, the calculation function of the interface layout weight factor G(x,y) is defined as:

[0064] In the above formula, A is the peak weight, which is usually set to 1; The coordinates of the visual center of the display interface are (e.g., (960, 540) on a display interface with a resolution of 1920*1080). and These are the standard deviations in the horizontal and vertical directions, respectively, and their values ​​are determined based on the golden ratio and screen size; the golden ratio is approximately 0.618.

[0065] Taking a display interface with a resolution of 1920*1080 as an example, the standard deviations in the horizontal and vertical directions are calculated as follows:

[0066] The aforementioned two-dimensional Gaussian function simulates a weight distribution that smoothly decays outward from the center of the screen, thereby continuously and differentiably quantifying the visual importance of the aligned golden section region.

[0067] The following calculations are performed on the interface layout weight factors for different preset interface positions corresponding to different texture resources in the display interface. The specific calculations are as follows: Selecting the center point of the screen (960, 540), the result of calculating the interface layout weight factor is: G(960, 540) = 1.0; indicating that the center point of the screen conforms to the ideal focus of the golden ratio layout and has the highest visual appeal. Selecting a point (700, 400) within the golden section region, the calculated result of the interface layout weight factor is: G(700, 400) = 0.78; indicating that this position is still within the sensitive area defined by the golden section and has strong visual importance, but is slightly lower than the center point. Selecting the screen edge point (0, 0), the result of calculating the interface layout weight factor is: G (0, 0) = 0.03; indicating that this position is far from the golden section area, and the visual appeal is significantly reduced.

[0068] By introducing interface layout weight factors, the golden ratio layout principle can be transformed into calculable weight parameters, allowing interface areas that conform to human visual preferences to receive higher priority in resource allocation. This improves visual comfort while enhancing the rationality and scientific nature of resource scheduling.

[0069] In some embodiments, the calculation of the visual center weight factor at the preset interface position includes: Based on the relative positional relationship between the preset interface position and the visual center area of ​​the display interface, the visual center weight factor is calculated using a preset distance attenuation function.

[0070] Preferably, to quantify the characteristic that human eye attention to the central area of ​​the screen decreases with increasing distance, the distance decay function is configured as a monotonically decreasing function. The value of the visual center weighting factor is inversely proportional to the geometric distance from the preset interface position to the visual center point of the display interface.

[0071] Furthermore, the visual center weight factor D(x,y) can be calculated through the following steps.

[0072] First, determine the visual center, which is usually defined as the geometric center of the display interface. .

[0073] For a given preset interface position, calculate its Euclidean distance to the visual center point. The specific expression is as follows: ; Euclidean distance, also known as straight-line distance or Euclidean distance, is a commonly used metric for measuring the shortest straight-line length between two points in two-dimensional or multi-dimensional space. It describes the spatial distance of any interface location relative to the visual center. Locations closer to the visual center are generally more likely to attract the driver's attention and should therefore be given a higher visual center weight.

[0074] Euclidean distance Substitute the preset decay function.

[0075] Alternatively, a linearly normalized decay function can be used, with the following specific expression:

[0076] in, The maximum Euclidean distance from the visual center point to the farthest corner of the screen is used for normalization. d is the visual center weight factor for this interface location; d is the Euclidean distance from this location to the visual center.

[0077] Taking a display interface with a resolution of 1920×1080 as an example, .

[0078] The linearly normalized decay function ensures that the weight is 1 at the visual center point and decays to 0 at the farthest point, thus achieving weight normalization and an intuitive linear mapping.

[0079] The visual center weight factor constructs a spatial decay model of visual attention by evaluating the Euclidean distance between each preset position on the interface and the visual center point.

[0080] The visual center weight factor adopts a linear normalized decay function to convert the distance into a weight value in the interval [0,1], realizing a quantitative mapping from the physical location of the interface to visual importance.

[0081] The closer a texture is to the visual center, the higher its weight value, thus receiving higher priority and higher quality texture resource allocation in the loading strategy; conversely, resources are allocated less when the texture is further away from the center. This mechanism not only conforms to the natural distribution of human eye gaze behavior but also improves the interpretability of the loading logic and the targeted nature of resource allocation.

[0082] In some embodiments, calculating the information density weighting factor for a preset interface location includes: The display interface will be divided into multiple sub-display areas; Determine the target sub-display area to which the preset interface position belongs; Based on the number and type weight of interface elements in the target sub-display area, the interface element density of the target sub-display area is calculated. Based on the density of interface elements, determine the information density weight factor for the preset interface position.

[0083] Preferably, in order to achieve quantification and smooth mapping of information distribution, this embodiment adopts a calculation method based on gridded statistics and bilinear interpolation, and the specific steps are as follows.

[0084] First, the display interface is evenly divided into a rectangular grid of M rows × N columns, with each grid serving as a sub-display area.

[0085] For example, a 1920×1080 resolution screen can be divided into a 12-row × 8-column grid, with each grid cell measuring approximately 160 pixels × 135 pixels. This method of partitioning achieves a good balance between computational complexity and accuracy.

[0086] Furthermore, all possible interface elements in the display interface are categorized by type, and each type is assigned a preset type weight to reflect the importance of the element to the driver's information or the degree of interaction criticality.

[0087] For example, a weight table can be defined as follows:

[0088] Table 1 Element Weight Table Furthermore, the aforementioned weight table is designed to transform the abstract importance of information into quantifiable values. Its assignment logic follows the principles of prioritizing safety criticality, followed by information dynamism, and then functional interactivity.

[0089] Warning icons carry the highest weight, directly indicating immediate safety risks in scenarios such as reversing or navigating narrow lanes. They represent the highest priority visual information that must be communicated clearly and immediately. Any rendering delays or quality degradation of this type of information can directly impact driving safety decisions.

[0090] Dynamic information displays, such as dynamic reversing guide lines, are given high weight because they change in real time with the driver's operation, providing key direction and path predictions for vehicle movement. They are the core visual feedback that assists the driver in precise control.

[0091] The control buttons involve interactive control of the AVM view itself, such as switching perspectives to adapt to different scenarios. Although they are not continuously focused when performing critical operations such as reversing, their availability is crucial for flexible system use, and their weight is moderate.

[0092] Static information displays provide a stable environmental reference system, such as vehicle outlines and distance scales, which help drivers judge relative position and distance. They are the foundation for building spatial cognition and have a moderate weight.

[0093] Status indicators, such as camera obstruction warnings or system mode indicators, are used to convey the health status and operating mode of the AVM system itself. Their importance lies in ensuring the reliability of the information source, but they are not always the visual focus.

[0094] Decorative and background elements are used only to enhance visual appeal and do not carry any functional or security information, therefore they have the lowest weight.

[0095] Furthermore, it should be noted that the element types and corresponding weights shown in Table 1 are merely an exemplary configuration for implementing this solution. In practical applications, the specific element classifications, number of types, and weight values ​​for each type can be adjusted and expanded according to different vehicle configurations, user interface design specifications, and specific functional safety level requirements. For example, new types such as "autonomous driving status indicator" or "emergency call button" can be added, and appropriate weights can be assigned to them.

[0096] Furthermore, for each sub-display area, the following calculations are performed.

[0097] Identify all interface elements that fall completely or partially within the grid area and accumulate their weight values ​​according to their type. To simplify the calculation, the element's affiliation can be determined solely by whether its center point falls within the grid.

[0098] The raw interface element density value RawDensity(i, j) of the mesh is calculated using the following formula:

[0099] in, Let k be the number of elements of type k in the grid. For the corresponding type weight, Let be the area of ​​the grid.

[0100] Normalize the original density values ​​of all grids to obtain the normalized grid density GridDensity(i, j), ensuring its value range falls within the [0,1] interval. For example:

[0101] For any given preset interface position (x, y), determine the grid (i, j) it belongs to based on its coordinates.

[0102] Furthermore, to obtain a continuous and smooth density distribution and avoid abrupt weight changes due to grid boundaries, bilinear interpolation is preferably used. The information density weight factor I(x, y) of the precise coordinate point (x, y) is calculated using the normalized density values ​​GridDensity of the four nearest grids ((i,j), (i+1,j), (i,j+1), (i+1,j+1)). This method effectively simulates the continuous distribution characteristics of information on the interface.

[0103] Information density weighting factors can continuously reflect the differences in visual attention caused by the varying information content and importance in different areas of the screen. By transforming the abstract concept of information functional density into a calculable weight value, which, together with factors such as layout and visual center, constitutes the basic display weight, the accuracy of predictions and the rationality of resource allocation decisions are significantly improved.

[0104] Preferably, the calculation of the reading habit weighting factor aims to quantify the general visual scanning habits of the user group into a weight distribution on the screen space. The specific calculation method is as follows: a product model based on a power function is used.

[0105] For any preset interface position (x, y) on the display interface, the formula for calculating its reading habit weighting factor R(x, y) is:

[0106] Where W and H are the total width and total height of the display interface, respectively. For example, for a screen with a resolution of 1920×1080, W = 1920 and H = 1080. m and n are empirical exponential parameters greater than 0, used to adjust the weight attenuation intensity in the horizontal and vertical directions, respectively, and their values ​​are determined based on the analysis of human eye reading habits.

[0107] Furthermore, the calculation of the reading habit weighting factor is based on a core assumption that, under typical left-to-right and top-to-bottom reading or visual scanning habits, the top left corner of the screen will receive the highest initial attention.

[0108] Horizontal dimension The smaller the x value, the closer the preset interface position is to the left, and the larger this value is. The parameter m controls the rate at which the weight decays from left to right.

[0109] Vertical dimension The smaller the y-value, the closer the preset interface position is to the top, and the larger this value is. The parameter n controls the rate at which the weight decays from top to bottom.

[0110] And usually n>m, to reflect the habit that vertical attention shifts, i.e., from top to bottom, are usually more significant than horizontal attention shifts, i.e. from left to right.

[0111] Alternatively, the parameters can be set to m = 0.8 and n = 1.2. Based on the above formula and parameter settings, the following exemplary calculations can be performed.

[0112] For the preset interface position (0, 0) in the upper left corner of the interface, the reading habit weighting factor is R(0, 0) = 1.0; For the preset interface position (1920,0) in the upper right corner of the interface, the reading habit weight factor is R(1920,0) = 0.0; For the preset interface position (480, 270) in the upper left area of ​​the center of the interface, the reading habit weighting factor is R(480, 270) = 0.57; For the preset interface position (1920, 1080) in the lower right corner of the interface, the reading habit weight factor is R(1920,1080) = 0.0; For the preset interface position (0, 1080) in the lower left corner of the interface, the reading habit weight factor is R(0, 1080) = 0.0.

[0113] Based on the above calculations, the reading habit weighting factor R(x, y) assigns a value between 0.0 and 1.0 to each position on the display interface. This value is proportional to the degree to which the position conforms to the reading habit of the upper left starting point.

[0114] Incorporating reading habit weighting factors into the calculation of basic display weights means that when allocating resources, the system will tend to assign higher initial importance to areas that align with the starting point of the human natural visual flow. This supplements weighting models based solely on geometric centers or fixed layouts, making visual focus prediction more consistent with cognitive patterns. Consequently, even when the interface design is not strictly symmetrical or the information distribution is unique, it can still make more reasonable resource loading decisions.

[0115] S212: The basic display weight is obtained by weighting and summing the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor.

[0116] Preferably, the weighted summation adopts a linear weighted model, and its general calculation formula is as follows: .

[0117] in: Indicates the screen coordinates The basic display weight at the preset interface position; , , , These represent the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor for that location, respectively. , , , Let be the weighting coefficients of each factor, and satisfy . All coefficient values ​​are greater than or equal to 0.

[0118] Furthermore, the weighting coefficients of each factor are determined based on a multi-factor fusion optimization strategy. A preferred implementation method is to determine them through a combination of user eye-tracking experiments and data-driven tuning.

[0119] First, data collection was conducted. In representative driving simulation scenarios, a large amount of eye-tracking data was collected when users viewed the AVM interface, recording the screen coordinates of their gaze points and dwell time.

[0120] The actual fixation frequency at each location was statistically analyzed, and its correlation with the four weighting factors was calculated. Numerical correlation of ).

[0121] To maximize prediction weights With the goal of minimizing the correlation between the weighting coefficients and the actual gaze frequency distribution or minimizing the prediction error, an optimization algorithm is used to solve for the optimal combination of weighting coefficients. .

[0122] Optionally, after optimization based on the above method, the weighting coefficients of each factor are: , , , .

[0123] Furthermore, the weighted summation step, in its specific implementation, also includes boundary handling and normalization safeguards.

[0124] Specifically, regarding input normalization, it ensures that the four weighting factors of the input computation are... The range of values ​​is all in Within the range to ensure the weighted summation result The range of values ​​is also within a controllable and interpretable range.

[0125] In terms of output smoothing, a slight Gaussian smoothing filter is applied to the calculated basic display weight matrix covering all preset interface positions. This aims to eliminate local weight abrupt changes or jagged edges that may be introduced by the initial gridding, UI element boundary discreteness, or discontinuous sampling. Through smoothing, the distribution of weight values ​​in screen space becomes more continuous and natural, and the smooth gradient transition is more in line with the perceptual characteristics of the human visual system, thus providing a higher-quality and more reliable input data foundation for subsequent dynamic adjustments and resource scheduling.

[0126] In addition, weighting coefficients These parameters are not fixed and can be designed as configurable. The system supports dynamically loading or switching different coefficient profiles based on different vehicle configurations, user interface themes, or personalized calibration data for specific users.

[0127] For example, for a car model whose UI design emphasizes central instrument panel information, the UI could be appropriately improved. The weight of the visual center weight factor; while for another UI design with dense shortcut control buttons set at the edge of the screen, the weight can be increased accordingly. Information density weighting factor and The weighting of reading habits is a key factor. This flexibility ensures that the solution can be widely adapted to diverse product designs and user needs, enhancing the versatility and practicality of the technology.

[0128] The calculation of basic display weights not only achieves the scientific integration of multi-dimensional visual influence factors, but also possesses data-based optimizability, implementation robustness, and configurability for different application scenarios, laying a solid foundation for generating accurate and reliable visual focus weights.

[0129] S22: Obtain real-time vehicle status data and calculate the status adjustment factor based on the real-time vehicle status data.

[0130] In some embodiments, such as Figure 4 As shown, the process of acquiring real-time vehicle status data and calculating a status adjustment factor based on the real-time vehicle status data includes the following steps S221-S222.

[0131] S221: Determine the current vehicle status based on real-time vehicle status data.

[0132] S222: Based on the preset adjustment mapping relationship between vehicle status and interface position, determine the status adjustment factor corresponding to the preset interface position. The status adjustment factor dynamically changes the importance of specific areas according to different driving states.

[0133] Preferably, the vehicle's real-time status data is acquired in real time via the vehicle bus, including at least the gear position signal, steering angle signal, and parking assist system activation signal.

[0134] The current driving scenario is determined by analyzing real-time vehicle status data, for example: The system determines the driving status (D gear forward or R gear reverse) based on the gear position signal, the left or right turn status based on the steering angle signal, and the parking assist status or P gear parking status based on the system activation signal.

[0135] Furthermore, the adjustment mapping relationship between vehicle status and interface position is pre-set in the system in the form of a lookup table or rule function. This mapping relationship explicitly defines the status adjustment factors applied to specific areas of the interface under different vehicle statuses, and these specific areas are usually specified by coordinate ranges or predefined area identifiers.

[0136] The following will elaborate on the forward, reverse, steering, and parking assist states as examples.

[0137] When the vehicle is in reverse, the system assigns differentiated state adjustment factors to different areas of the screen based on a preset mapping relationship. The specific rules are as follows: If coordinates If it belongs to the rear environment area, then ; If coordinates If it falls within the detection range of rear radar or ultrasonic sensors, then ; If coordinates If it falls within the area of ​​the environment in front of the vehicle, then ; For other screen areas not covered above, then .

[0138] In reverse mode, the above mapping relationship increases the importance weight of the rear and rear side areas that the driver needs to focus on, while appropriately reducing the weight of the front area that is not currently being focused on.

[0139] When the vehicle is moving forward, the state adjustment factor The allocation rules are as follows: If coordinates If it belongs to the area of ​​the environment in front, then ; If coordinates If it falls within the left or right blind zone monitoring area, then ; If coordinates If it belongs to the rear area, then ; For other areas, ; When moving forward, follow the principle of maintaining a high level of attention to the road conditions ahead and blind spots on the sides, while giving basic attention to the area behind.

[0140] When the vehicle is turning, the state adjustment factor Based on real-time steering angle Perform dynamic calculations: If coordinates If it belongs to the outer area of ​​the steering, then ; If coordinates If it belongs to the inner area of ​​the steering wheel, then ; For other areas, ; The above mapping relationship realizes the dynamic shift of visual focus with steering operation. The weight of the outer area increases with the increase of steering angle, while the weight of the inner area decreases accordingly, which is consistent with the typical visual behavior of drivers looking at the outer side in a curve.

[0141] When the parking assist system is activated, the state adjustment factor The allocation focuses on parking guidance information: If coordinates If it falls within the marked parking space area, then ; If coordinates If it belongs to the path guide line area, then ; For other areas, .

[0142] The above mapping relationship ensures that, in parking scenarios, the key visual elements most relevant to parking space recognition and path planning receive the highest resource allocation priority.

[0143] Furthermore, the mapping relationship is adjusted according to the principle that driving tasks are strongly correlated with visual attention. Under specific driving conditions, the key information that the driver needs to complete the current main task is usually concentrated in a specific area of ​​the screen.

[0144] By presetting higher state adjustment factors for these interface areas that are strongly related to the current task in the above mapping relationship, the system can automatically assign them higher importance when calculating the visual focus weight of texture resources. This ensures that, under limited system resources, the relevant texture resources can obtain higher loading priority and rendering quality, thereby prioritizing the clear and timely presentation of visual information that is crucial to driving safety and operational convenience.

[0145] It should be noted that the region divisions and specific multiplier values ​​in the above-mentioned mapping relationship are merely examples. In practical applications, their specific definitions can be calibrated and optimized based on the camera angles, display layouts, human-machine interface (HMI) designs, and in-depth driver behavior studies for different vehicle models to ensure the accuracy and effectiveness of the state adjustments.

[0146] S23: Obtain the system runtime and calculate the time decay factor based on the system runtime.

[0147] Preferably, time decay factor Using the exponential decay function, its calculation formula is as follows:

[0148] in: This indicates the running time since the system started up, in seconds; It is a decay coefficient greater than zero, used to control the decay rate.

[0149] Time decay factor It is used to characterize the natural process by which a user's visual attention evolves from global scanning to focused dwell after the system is started.

[0150] Specifically, at the moment of startup ( ), This means that the time decay factor does not affect the initial weight of any area, and the system prioritizes the rapid presentation of the basic content of the entire interface; as time goes on ( (increase) The value starts from 1 and decreases monotonically, which allows the weight of non-critical areas loaded in the early stages to decay, thereby guiding system resources to gradually concentrate on the visually important areas that are continuously important.

[0151] Furthermore, the attenuation coefficient The value of λ is optimized through experiments and system performance targets. The typical range of values ​​for the attenuation coefficient λ is as follows: .

[0152] For example, take At that time, the specific value of the time decay factor is expressed as follows: seconds, ; seconds, ; seconds, ; seconds, .

[0153] The time decay factor ensures that the system prioritizes loading the basic textures of all areas during the initial startup phase, and then focuses on optimizing key areas later.

[0154] S24: Based on the state adjustment factor, time decay factor, and basic display weight, the visual focus weight of the texture resource is calculated.

[0155] Preferably, the visual focus weight is calculated using a product fusion model, and the calculation formula is as follows:

[0156] in: Indicates the time at which the system is running Vehicle status Below, located at interface coordinates Visual focus weight at the location; This represents the basic display weight located at interface coordinates (x, y), reflecting the inherent importance of this location based on the characteristics of human visual perception; This represents the time decay factor, reflecting the changing trend of attention over time; This represents the state adjustment factor, reflecting the impact of the current driving state on the importance of the area.

[0157] Furthermore, the visual focus weight is determined by the basic display weight. Time decay factor With state adjustment factor The result is obtained by multiplying the three factors. Multiplication means that a significant change in any factor will have a decisive impact on the final weight.

[0158] For example, even if a region has a high base display weight, if its state adjustment factor is low under the current driving condition, its final weight will become extremely low, reflecting the actual shift in the driver's attention in that scenario.

[0159] It should be understood that the differentiated loading in this application is not a one-time operation performed only during the system initialization phase, but a resource scheduling process that continues throughout the entire system operation cycle and can be dynamically adjusted based on vehicle status, system running time, and visual focus prediction results.

[0160] In some embodiments, such as Figure 5 As shown, it also includes a fixation optimization step, which includes the following steps.

[0161] S61: Acquire real-time gaze information of the driver through onboard vision sensors.

[0162] S62: Calculate the attention enhancement factor based on real-time gaze point information.

[0163] S63: Adjust the visual focus weight of texture resources based on the attention enhancement factor.

[0164] Preferably, attention-enhancing factors The calculation is based on real-time gaze coordinates. With any preset interface position on the screen Euclidean distance between It is modeled using a Gaussian decay function, and its calculation formula is as follows: .

[0165] in, This is the attenuation coefficient, used to control the rate at which attention diminishes with increasing distance.

[0166] A(x,y) ensures that the closer the screen area is to the driver's real-time gaze point, the closer the attention enhancement factor is to the maximum value of 1; the farther away, the factor value decays exponentially to close to 0.

[0167] For example, it can be set Pixels, when a certain position is 50 pixels away from the gaze point, At a distance of 360 pixels, .

[0168] Furthermore, the visual focus weights are adjusted based on the attention enhancement factor, and a weight fusion mechanism is adopted.

[0169] Specifically, the calculated attention enhancement factor Weight of original visual focus The images are then fused to generate the optimized final visual focus weights. An effective fusion method is weighted enhancement:

[0170] in, To enhance the intensity coefficient, the influence of real-time gaze information on the final weights is adjusted. This fusion method retains the core weights based on model predictions while significantly increasing the weight of the driver's real-time gaze area.

[0171] For example, if the original weight of a certain region is If it is located near the driver's gaze point, its weight will be increased after optimization. This allows it to receive higher priority in subsequent loading decisions.

[0172] By introducing real-time physiological signals, it is possible to more accurately capture the driver's instantaneous intentions and attention jumps, which is suitable for rapid switching of driving scenarios or when sudden warning information appears on the interface, ensuring that critical information can always be guaranteed by resources and displayed in a timely manner.

[0173] S3: Determine the corresponding target texture quality level for each texture resource based on the visual focus weight of each texture resource.

[0174] In some embodiments, determining the corresponding texture quality level for a texture resource includes: The visual focus weight of the texture resource is compared with a first quality threshold and a second quality threshold; If the visual focus weight is greater than the first quality threshold, the target texture quality level is determined to be high quality. If the visual focus weight is less than or equal to the second quality threshold, the target texture quality level is determined to be low quality. If the visual focus weight is between the first quality threshold and the second quality threshold, the target texture quality level is determined to be medium quality level.

[0175] Preferably, the setting of the first quality threshold T1 and the second quality threshold T2 is based on a balanced analysis of the statistical distribution of visual focus weights and user experience. Specifically, the threshold division aims to accurately allocate limited rendering resources to the most important areas while ensuring the overall usability of the interface.

[0176] For example, the first quality threshold T1 and the second quality threshold T2 can be configured as follows: T1 = 0.65, T2 = 0.25.

[0177] Based on the above configuration, the visual focus weight of texture resources will be divided into three intervals, corresponding to different target texture quality levels: Areas with a visual focus weight higher than 0.65 are identified as critical focus areas, corresponding to a high quality level; Areas with a visual focus weight of 0.25 or less are classified as edge or minor areas, corresponding to a low quality level; The area with a visual focus weight between 0.25 and 0.65 is considered a transitional area, corresponding to a medium quality level.

[0178] Furthermore, as shown in Table 2, different target texture quality levels correspond to differentiated resource processing strategies in order to achieve an optimized balance between performance and image quality.

[0179]

[0180] Table 2. Correspondence between Texture Quality Levels and Resource Strategies The high-quality level corresponds to the core area with the highest visual focus weight, such as the rear view when reversing and the simulated road conditions in front when driving.

[0181] The high-quality level fully decodes and loads the texture resources at their original resolution to ensure the clarity, detail, and color accuracy of key information, providing users with the best visual experience and a basis for security judgment.

[0182] Medium quality level corresponds to transitional areas that are of some importance but not core, such as side blind spot auxiliary views and status information panels that are not currently in focus.

[0183] Medium quality levels can be achieved by downsampling, for example, loading at 50% of the original resolution. This significantly reduces texture memory usage and GPU bandwidth consumption while allowing for only a slight loss of visual detail.

[0184] Low quality levels correspond to the least important edge or background areas, such as non-interactive decorative backgrounds or distant areas far from the current driving focus.

[0185] Lower quality levels can further reduce resource requirements significantly, for example, by using simplified textures at 25% of the original resolution, or even temporarily replacing them with solid colors or minimalist geometric shapes, thereby achieving maximum resource savings and ensuring system startup speed and smooth operation.

[0186] S4: Obtain the current texture quality level of each texture resource, and calculate the loading priority of the texture resource based on the visual focus weight, the target texture quality level, and the current texture quality level of each texture resource.

[0187] Preferably, the current texture quality level refers to a state attribute stored in the system resource management unit, which records the quality level of the texture resource currently configured and used in memory or video memory. The current texture quality level reflects the result of the system's previous loading decision and execution, including the initial loading and any subsequent dynamic adjustments.

[0188] During system initialization and startup, since no loading operations have been performed, there is no valid texture data in memory or video memory. Therefore, the current texture quality level of all texture resources to be loaded is initialized to a preset minimum quality level.

[0189] The initialization strategy ensures that the system can begin the first round of loading priority calculation and loading scheduling based on a clear and consistent baseline state after startup. This ensures that even in critical areas with high visual focus weight, the initial quality gap—that is, the difference between the target texture quality level and the current texture quality level—is maximized, thereby achieving extremely high loading priority in the first round of calculation and driving the system to prioritize loading high-quality textures in these critical areas.

[0190] Preferably, the loading priority is calculated using a product model, and the calculation formula is as follows:

[0191] in: Indicates the location on the interface. The loading priority of texture resources; a higher value indicates that the texture is loaded earlier in the order. The visual focus weight of the texture resource located at interface position (x,y) is calculated by the aforementioned steps and reflects the visual importance of this area in its current position and state. The target texture quality level for the texture resource located at interface position (x,y) is the desired texture quality level determined based on its visual focus weight. This refers to the current texture quality level of the texture resource located at interface position (x, y), i.e. The actual texture quality level that is currently loaded and in effect in the system.

[0192] Loading priority is positively correlated with and sensitive to both visual focus weight and texture quality gap. This ensures that the system can immediately identify and prioritize the scheduling of texture resources that are most critical to visual experience and safety, and have the largest quality gap (i.e., currently insufficient quality), such as the rear view when reversing. This significantly reduces the instantaneous load and resource contention during system initialization, shortens the presentation delay of critical safety information, and achieves an optimal balance between response speed and visual experience under limited resource conditions.

[0193] In some embodiments, such as Figure 6 As shown, it also includes a dynamic quality adjustment step, which specifically includes the following steps.

[0194] S71: Real-time utilization rate of hardware resources within the monitoring system.

[0195] S72: The overall system load index is calculated based on the real-time utilization of hardware resources; S73: Compare the overall system load index with the preset load threshold to obtain the load status determination result; S74: Adjust the current texture quality level of the texture resource based on the load status determination result.

[0196] Preferably, the real-time utilization of hardware resources includes at least: central processing unit (CPU) utilization, graphics processing unit (GPU) utilization, and system memory utilization.

[0197] Real-time hardware resource utilization data is obtained in real time through performance interfaces provided by the operating system or dedicated drivers.

[0198] System overall load index The calculation formula, expressed using the weighted summation model, is as follows:

[0199] in: The real-time utilization rate of the central processing unit at time t is normalized and its value ranges from [0,1]. The real-time utilization of the graphics processor at time t is normalized and its value ranges from [0,1]. The real-time utilization rate of the system memory at time t is normalized and its value range is [0,1]. Weight the importance of each resource with a coefficient, and satisfy the following conditions: .

[0200] Optionally, the importance weighting coefficient for each resource can be configured as follows: , , This is to demonstrate the core roles of the CPU and GPU in graphics loading and rendering.

[0201] Furthermore, the preset load thresholds include a low load threshold. and high load threshold .

[0202] Based on the calculated system comprehensive load index The system executes the following dynamic quality adjustment rules to adaptively balance image quality and performance: when This indicates that the system load is low and system resources are sufficient, allowing for an improved visual experience. At this point, the system will appropriately increase the texture quality by one level at the current location. But not exceeding the highest quality level supported by the system. .

[0203] when This indicates that the system load is normal and the system resources are in a balanced state. At this time, the current quality level of all textures remains unchanged. .

[0204] when This indicates a high system load and limited system resources, necessitating a priority on ensuring smooth operation. In this situation, the system will appropriately reduce texture quality to free up resources, by lowering the texture quality level of the current location by one level. But not lower than the minimum quality level supported by the system. .

[0205] Optionally, the preset load threshold can be set to: This indicates that texture quality can be improved when the system load is below 50%. When the system load exceeds 80%, the texture quality needs to be reduced.

[0206] Through the aforementioned closed-loop adjustment mechanism, the system can flexibly manage texture quality under different system load conditions, thereby achieving overall optimization of system performance and resource utilization efficiency while ensuring the experience of the visual focus area.

[0207] In addition, the dynamic quality adjustment steps work in conjunction with the incremental quality improvement mechanism.

[0208] After the system completes its initial loading and enters the idle phase, the dynamic quality adjustment steps will dominate the gradual optimization process: during low-load periods, the quality level of textures in non-critical areas will be gradually and in batches; once an increase in load is detected, the quality improvement operation will be immediately paused or rolled back to ensure system responsiveness.

[0209] The progressive quality improvement mechanism enables the system to intelligently balance image quality and performance throughout its entire lifecycle, always prioritizing resources to ensure driving safety and core interactive experience, while maximizing the use of idle resources to improve the overall visual experience.

[0210] S5: Determine the loading order of each texture resource according to the loading priority, and load the corresponding texture resource according to the loading order and the texture quality corresponding to the target texture quality level.

[0211] Preferably, the loading order is determined based on the calculated loading priority, and system resources are dynamically scheduled to load texture resources asynchronously based on the loading order and the corresponding target texture quality level.

[0212] Furthermore, a priority loading queue can be constructed, in which the system arranges all texture resource tasks to be loaded in descending order according to their loading priority values, forming an ordered task list.

[0213] The loading engine continuously retrieves the highest priority tasks from the head of the queue, loads, decodes, and uploads them to the GPU memory, thus ensuring that the most important visual content is presented first.

[0214] Furthermore, the differentiated loading process includes several key mechanisms to ensure efficiency and smoothness.

[0215] For texture resources marked as highest priority, the system uses preemptive preloading to ensure they are completed as quickly as possible. For medium- and low-priority resources, streaming loading is used, loading them gradually when system I / O is idle or the load is low to avoid blocking high-priority tasks.

[0216] For texture resources with extremely low priority or located in completely invisible areas, the system implements lazy loading, meaning that no actual loading task is assigned for them. Before these areas need to be displayed, minimalist placeholder textures, such as monochrome blocks or low-resolution mosaics, can be used as temporary substitutes. The actual loading will be triggered when their priority is increased or system resources become sufficient.

[0217] Furthermore, the differentiated loading process is closely linked to the target texture quality level determined in the aforementioned steps, realizing resource scheduling that combines priority-based sequential differentiation with quality differentiation allocated according to visual needs.

[0218] Specifically, different loading quality processing strategies are implemented based on the target texture quality level of the texture resource.

[0219] For texture resources with a target texture quality level of high quality, the system loads its original or high-resolution texture data and performs complete and high-precision decoding to ensure that key visual focus areas achieve optimal clarity and detail.

[0220] For texture resources with a target texture quality level of medium, the system can employ optimized resource processing strategies, such as loading texture data versions that have undergone moderate downsampling or use a higher compression ratio. This significantly reduces data volume, memory usage, and decoding overhead while maintaining basic recognizability.

[0221] For resources with a target texture quality level of low quality, the system can load a minimalist placeholder texture with low data volume, such as a solid color texture or an extremely low resolution mosaic, or directly use the system's preset general simplified texture, thereby meeting its display requirements with minimal resource cost.

[0222] If the vehicle status suddenly changes during the loading process, causing the visual focus weight and loading priority to be recalculated, the system can immediately interrupt the currently running low-priority loading task and seize resources to immediately start loading a new high-priority task, thereby quickly responding to sudden changes in the driving scenario.

[0223] Based on adaptive throughput control according to system load, the loading engine monitors the utilization of CPU, GPU, and memory in real time. When the overall system load is too high, it automatically reduces the scheduling frequency or batch size of loading threads to reduce concurrent loading tasks and prevent loading operations from exacerbating system lag. Conversely, when the system is idle, it increases loading throughput to accelerate the gradual improvement of texture quality.

[0224] The present invention provides an in-vehicle display loading system for implementing the in-vehicle display loading method as described in the first aspect.

[0225] Combination Figure 7 As shown, this embodiment discloses a specific implementation of a computer device. The computer device may include a processor 41 and a memory 42 storing computer program instructions.

[0226] Specifically, the processor 41 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0227] The memory 42 may include a large-capacity storage device for data or instructions. For example, and not limitingly, the memory 42 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 42 may include removable or non-removable (or fixed) media. Where appropriate, the memory 42 may be internal or external to the data processing device. In a particular embodiment, the memory 42 is non-volatile. Volatile memory. In a particular embodiment, memory 42 includes read-only memory. ROM (ROM-only memory) and RAM (Random Access Memory). Where appropriate, the ROM can be a mask-programmed ROM or a programmable ROM. Only Memory (PROM) and Erasable Programmable Read-Only Memory (EPRROM) The RAM can be a type of RAM, such as EPROM (Electrically Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), EAROM (Electrically Alterable Read-Only Memory), or FLASH (Flash Memory), or a combination of two or more of these. Where appropriate, the RAM can be a Static Random Access Memory (SRAM). Access Memory (SRAM) or Dynamic Random Access Memory (DRAM) can be Fast Page Mode Dynamic Random Access Memory (FPMDRAM), Extended Data Out Dynamic Random Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0228] The memory 42 can be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 41.

[0229] The processor 41 implements the vehicle display loading method in the above embodiment by reading and executing computer program instructions stored in the memory 42.

[0230] In some embodiments, the computer device may further include a communication interface 43 and a bus 40. For example, Figure 7 As shown, the processor 41, memory 42, and communication interface 43 are connected through bus 40 and complete communication with each other.

[0231] Communication interface 43 is used to enable communication between modules, devices, units and / or equipment in the embodiments of this application.

[0232] Communication port 43 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0233] Bus 40 includes hardware, software, or both, that couples components of a computer device together. Bus 40 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. For example, and not as a limitation, bus 40 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, and a PCI bus. Express (PCI The bus may be an X-bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 40 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0234] Furthermore, in conjunction with the vehicle display loading method in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the vehicle display loading methods in the above embodiments.

[0235] The present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described vehicle display loading method.

[0236] The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described vehicle display loading method.

Claims

1. A method for loading an in-vehicle display, characterized in that, It includes: Obtain a set of texture resources to be loaded, wherein the set of texture resources stores texture resources and their preset interface positions; For each texture resource, based on the preset interface position where it is configured to be rendered, the visual focus weight of the texture resource is determined according to a pre-built visual focus prediction model. Based on the visual focus weight of each texture resource, a corresponding target texture quality level is determined for the texture resource; Obtain the current texture quality level of each texture resource, and calculate the loading priority of the texture resource based on the visual focus weight, the target texture quality level, and the current texture quality level of each texture resource; The loading order of each texture resource is determined according to the loading priority, and the corresponding texture resource is loaded according to the loading order and the texture quality corresponding to the target texture quality level.

2. The vehicle display loading method according to claim 1, characterized in that, Based on a pre-built visual focus prediction model, the visual focus weights of the texture resource are determined as follows: Based on the preset interface position, determine the basic display weight of the preset interface position; Acquire real-time vehicle status data, and calculate a status adjustment factor based on the real-time vehicle status data; Obtain the system runtime and calculate the time decay factor based on the system runtime; The visual focus weight of the texture resource is calculated based on the state adjustment factor, the time decay factor, and the basic display weight.

3. The vehicle display loading method according to claim 1, characterized in that, It also includes a foveation optimization step, including: Obtain real-time driver gaze point information; Calculate the attention enhancement factor based on the real-time gaze point information; The visual focus weight of the texture resource is adjusted according to the attention enhancement factor.

4. The vehicle display loading method according to claim 2, characterized in that, The basic display weight for determining the preset interface position includes: Based on the layout features, visual characteristics, and information distribution features of the preset interface position in the display interface, the interface layout weight factor, visual center weight factor, information density weight factor, and reading habit weight factor of the preset interface position are calculated. The basic display weight is obtained by weighting and summing the interface layout weight factor, the visual center weight factor, the information density weight factor, and the reading habit weight factor.

5. The vehicle display loading method according to claim 2, characterized in that, The process of acquiring real-time vehicle status data and calculating a status adjustment factor based on the real-time vehicle status data includes: The current vehicle status is determined based on the real-time vehicle status data. Based on the preset adjustment mapping relationship between vehicle status and interface position, the status adjustment factor corresponding to the preset interface position is determined.

6. The vehicle display loading method according to claim 1, characterized in that, Determining the corresponding target texture quality level for the texture resource includes: The visual focus weight of the texture resource is compared with a first quality threshold and a second quality threshold; If the visual focus weight is greater than the first quality threshold, then the target texture quality level is determined to be a high quality level; If the visual focus weight is less than or equal to the second quality threshold, then the target texture quality level is determined to be low quality. If the visual focus weight is between the first quality threshold and the second quality threshold, then the target texture quality level is determined to be a medium quality level.

7. The vehicle display loading method according to claim 1, characterized in that, It also includes dynamic quality adjustment steps, including: Real-time utilization rate of hardware resources within the monitoring system; Based on the real-time utilization rate of the aforementioned hardware resources, the overall system load index is calculated. The overall system load index is compared with a preset load threshold to obtain the load status determination result; Based on the load status determination result, the current texture quality level of the texture resource is adjusted.

8. The vehicle display loading method according to claim 4, characterized in that, The visual center weight factor for calculating the preset interface position includes: Based on the relative positional relationship between the preset interface position and the visual center area of ​​the display interface, the visual center weight factor is calculated using a preset distance attenuation function.

9. The vehicle display loading method according to claim 4, characterized in that, The information density weighting factor for the preset interface position is calculated as follows: The display interface is divided into multiple sub-display areas; Determine the target sub-display area to which the preset interface position belongs; Based on the number and type weight of interface elements in the target sub-display area, the interface element density of the target sub-display area is calculated. Based on the density of the interface elements, determine the information density weighting factor of the preset interface position.

10. A vehicle-mounted display loading system, characterized in that, Used to implement the vehicle display loading method as described in any one of claims 1 to 8.

Images