Vehicle-mounted head-up display method and system based on Boltzman entropy

Abstract

The invention provides a Boltzman entropy-based vehicle-mounted head-up display method and system, and the method comprises the steps: mapping a driving scene into a plurality of types of microscopic states, and defining a quantifiable data index of each type of microscopic state as a microscopic feature of each type of microscopic state; obtaining each item of microscopic feature data of each type of microscopic state of the current driving scene, and calculating the Boltzmann entropy of the current driving scene; vehicle-mounted head-up display levels are determined according to the Boltzmann entropy of the current driving scene, and different vehicle-mounted head-up display levels comprise different numbers of display elements. According to the method, the absolute complexity of the current driving scene is quantized by introducing the Boltzmann entropy, and a complete microscopic state mapping system is constructed to realize the extraction of the microscopic state number in the driving scene, so that the Boltzmann entropy of the current driving scene is accurately calculated.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted human-computer interaction systems, and in particular to a vehicle-mounted head-up display method and system based on Boltzmann entropy. Background Art

[0002] With the development of intelligent driving technology, the use of electronic heads-up navigation maps is becoming increasingly common. While these maps provide convenience for drivers, excessive visual guidance can also distract drivers. Therefore, the proper configuration of guidance information is crucial, and this information needs to be determined based on the information entropy of the real-world head-up navigation map.

[0003] Traditionally, the Shannon entropy method is often used to calculate information uncertainty based on the modified distribution, thereby quantifying the complexity of the current driving scenario and configuring the guidance information of the vehicle head-up display system based on the complexity of the driving scenario.

[0004] However, the calculation method based on Shannon entropy has difficulty quantifying the absolute complexity of driving scenarios, resulting in a lack of accuracy in judging the complexity of actual driving scenarios and making it difficult to accurately adjust the guidance information of the vehicle's head-up display system. Summary of the Invention

[0005] The present invention provides a vehicle-mounted head-up display method and system based on Boltzmann entropy, which are used to overcome the defect of poor accuracy of the vehicle-mounted head-up display system based on Shannon entropy in the prior art, and realize a vehicle-mounted head-up display method with higher accuracy to quantify the absolute complexity of the driving scene.

[0006] The present invention provides a vehicle-mounted head-up display method based on Boltzmann entropy, comprising: Map the driving scene into several types of micro-states, and define quantifiable data indicators for each type of micro-state as the micro-features of each type of micro-state; Obtaining each microscopic feature data of each type of microscopic state of the current driving scene, and calculating the Boltzmann entropy of the current driving scene; The vehicle head-up display level is determined according to the Boltzmann entropy of the current driving scene, wherein different vehicle head-up display levels include different numbers of display elements.

[0007] According to a vehicle-mounted head-up display method based on Boltzmann entropy provided by the present invention, the step of calculating the Boltzmann entropy of the current driving scene specifically includes: Selecting a weight for each type of microstate according to a road type to which the current driving scene belongs, wherein a mapping relationship between the road type and the weight for each type of microstate is predefined; Determining the Boltzmann entropy state number of each microscopic feature data according to each microscopic feature data of each type of microscopic state, wherein a mapping relationship between each microscopic feature data and the Boltzmann entropy state number is predefined; Calculate the first product of all Boltzmann entropy state numbers corresponding to each type of microstate; The first product of each type of microstate is weighted using the weight of each type of microstate, and the weighted product of all microstates is calculated to obtain the Boltzmann entropy of the current driving scene.

[0008] According to a Boltzmann entropy-based vehicle head-up display method provided by the present invention, before the step of selecting the weight of each micro-feature according to the road type to which the current driving scene belongs, the method further includes: Based on the results of the driver cognition experiment, the weight of each micro-feature under each road type is defined, wherein the road types include expressways, urban main roads, urban secondary roads, rural roads and tunnels.

[0009] According to a Boltzmann entropy-based in-vehicle head-up display method provided by the present invention, the in-vehicle head-up display levels include a first level representing low-complexity scenes, a second level representing medium-complexity scenes, and a third level representing high-complexity scenes, and from the first level to the third level, the number of display elements included in the in-vehicle head-up display level decreases step by step.

[0010] According to a vehicle-mounted head-up display method based on Boltzmann entropy provided by the present invention, the microscopic states are defined as four categories, namely, people, vehicles, roads and environments.

[0011] According to a Boltzmann entropy-based vehicle head-up display method provided by the present invention, the step of obtaining each microscopic feature data of each type of microscopic state of the current driving scene specifically includes: Extracting an in-vehicle video of a current driving process, wherein the in-vehicle video represents the driver's visual field environment; Extracting image frames from the vehicle-mounted video and using an object detection model to detect object bounding boxes corresponding to people, vehicles, and road signs in each frame; Each item of microscopic feature data of each type of microscopic state of the current driving scene is determined based on the detection result of the target detection model.

[0012] The present invention also provides a vehicle-mounted head-up display system based on Boltzmann entropy, comprising: an acquisition module, configured to acquire each microscopic feature data of each type of microscopic state of the current driving scene and calculate the Boltzmann entropy of the current driving scene; a determination module, configured to determine a vehicle head-up display level according to the Boltzmann entropy of the current driving scene; The display module is used to adjust and display display elements according to the determined vehicle head-up display level.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the in-vehicle head-up display method based on Boltzmann entropy as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for implementing any of the above-described Boltzmann entropy-based vehicle head-up display methods is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-described methods for in-vehicle head-up display based on Boltzmann entropy.

[0016] The Boltzmann entropy-based in-vehicle head-up display method and system provided by the present invention introduces Boltzmann entropy to quantify the absolute complexity of the current driving scene, constructs a complete micro-state mapping system to realize the extraction of the number of micro-states in the driving scene, accurately calculates the Boltzmann entropy of the current driving scene, and determines the corresponding in-vehicle head-up display level according to the Boltzmann entropy of the driving scene for display, thereby controlling the number of display elements and reducing the driving load of the driver in complex scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 1 is a flow chart of a vehicle-mounted head-up display method based on Boltzmann entropy provided by the present invention; Figure 2 Schematic diagram of the structure of the vehicle head-up display system based on Boltzmann entropy provided by the present invention; Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0020] The following combination Figure 1 The present invention introduces a vehicle head-up display method based on Boltzmann entropy. Figure 1 As shown, including: Step 101: Map the driving scene into several types of micro-states, and define quantifiable data indicators for each type of micro-state as micro-features of each type of micro-state; The vehicle-mounted head-up display system, also known as the HUD (Head-up-Display), projects relevant information for assisted driving onto the windshield in front of the driver through the head-up display, allowing the driver to obtain relevant driving information without having to lower his head, thereby assisting the driver in driving.

[0021] But at the same time, the information projected by the HUD also increases the content of the driver's field of view. In complex driving scenarios, too much HUD display information will affect the driver's judgment of the current driving environment. Therefore, the number of display elements of the HUD projection needs to be increased or decreased according to the driver's actual driving environment.

[0022] Traditional quantification methods based on Shannon entropy usually choose to judge the uncertainty of driving scenarios based on probability distribution. That is, they believe that driving scenarios with higher uncertainty require fewer display elements. Probability distribution cannot accurately quantify the absolute complexity of the scene, which may lead to errors in its judgment of scene complexity.

[0023] Therefore, the present invention introduces Boltzmann entropy to map the driving scenario into microscopic state numbers and quantify the absolute complexity through the principles of statistical mechanics.

[0024] In order to realize the quantification of driving scenarios based on Boltzmann entropy, it is necessary to map the driving scenarios into micro-state numbers in advance.

[0025] Optionally, factors that have a greater impact on the driver are selected as micro-states, and quantifiable data indicators for each type of factor are determined as micro-features, and a quantification method for each micro-feature is defined.

[0026] As a preferred implementation, the microscopic states are defined as four categories: people, vehicles, roads, and environments. Then, the microscopic features corresponding to each category of microscopic states and their quantification methods are defined respectively.

[0027] Optionally, when the micro-state is a person, the micro-features may include pedestrian density, pedestrian movement direction, pedestrian movement speed, etc.; when the micro-state is a car, the micro-features may include its own vehicle speed, road vehicle density, road vehicle type, road vehicle taillight status, road vehicle speed, distance, etc.; when the micro-state is a road, the micro-features may include road type, number of lanes, traffic sign density, road curvature radius, road slope, road material, etc.; when the micro-state is the environment, the micro-features may include weather, light conditions, tunnels, elevated roads, backlighting, etc.

[0028] However, in order to reduce the amount of parallel micro-state computations in the vehicle host and enhance real-time performance, the following preferred solution is implemented based on the results of the driver questionnaire survey among the many optional micro-features mentioned above: For the microscopic state of a person, pedestrian density is defined as its microscopic feature. Optionally, the number of pedestrians in a frame of image is further defined as a calculation method for pedestrian density.

[0029] For the vehicle microstate, we define the vehicle's speed, road vehicle density, and road vehicle type as its microscopic features. Optionally, we further define the vehicle's speed as kilometers per hour, the number of vehicles in a frame as road vehicle density, and the composition of large vehicles, small vehicles, and non-motorized vehicles in a frame as road vehicle type.

[0030] For the microscopic state of a road, we define road type, number of lanes, and traffic sign density as its microscopic features. Optionally, road type refers to the type of road, such as expressway, urban main road, and urban secondary road; number of lanes refers to the number of drivable lanes; and traffic sign density refers to the number of traffic signs contained in a frame of image.

[0031] For the microscopic state of the environment, weather and light conditions are defined as its microscopic characteristics. Optionally, weather includes sunny, cloudy, overcast, rainy, foggy, and snowy days, and light conditions refer to the brightness of the light.

[0032] Among them, the questionnaire for drivers included the driver's basic information, micro-state, micro-characteristics and their weights (1-10 points). Through offline distribution, 56 valid questionnaires were collected. The analysis results showed that when the micro-state was human, the pedestrian density (9.8), pedestrian movement direction (8.2), and pedestrian movement speed (8.5) were included; when the micro-state was car, the micro-characteristics included vehicle speed (9.6), road vehicle density (9.2), road vehicle type (9.3), road vehicle taillight state (8.6), road vehicle speed (8.8), and distance (8.2); when the micro-state was road, the micro-characteristics included road type (9.2), number of lanes (9.6), traffic sign density (9.3), road curvature radius (8.7), road slope (7.8), and road material (6.5); when the micro-state was environment, the micro-characteristics included weather (9.8), light conditions (9.6), tunnel (9.1), elevated road (8.3), and backlight (8.6). Based on the above results, micro features with weights above 9 are selected.

[0033] On this basis, the microscopic feature data of the actual driving scene is obtained to characterize the microscopic state of the corresponding category, so as to calculate the Boltzmann entropy of the current driving scene based on the number of microscopic states and realize the quantification of scene complexity based on Boltzmann entropy.

[0034] Step 102: Obtain each microscopic feature data of each type of microscopic state of the current driving scene, and calculate the Boltzmann entropy of the current driving scene; Real-time vehicle video is acquired, and each microscopic feature data of each type of microscopic state of the current driving scene is determined based on each frame image of the vehicle video.

[0035] The in-vehicle video is a video acquired by an image acquisition device installed in front of the vehicle, so that the video content simulates the driver's field of view.

[0036] In a feasible implementation, the image acquisition device may be a camera shared with an intelligent driving / assisted driving system, which is typically installed at the center of the outer side of the vehicle's front windshield.

[0037] In other feasible embodiments, the image acquisition device may also be independently provided, for example, installed inside the vehicle and located in front of the main driver's seat, so that the shooting range of the image acquisition device can be closer to the driver's field of view, so as to better quantify the current driving scene based on the driver's posture field of view rather than the machine's field of view.

[0038] According to the definition of each micro-feature data, each micro-feature data of each type of micro-state in the current driving scene is extracted and calculated based on each frame image of the on-board video. On this basis, the micro-feature data is further discretized into state classification to obtain the Boltzmann state number corresponding to each micro-feature data, and then the Boltzmann entropy calculation formula is used to calculate the Boltzmann entropy of the current driving scene.

[0039] Step 103 : determining a vehicle head-up display level according to the Boltzmann entropy of the current driving scene, wherein different vehicle head-up display levels include different numbers of display elements.

[0040] The calculated Boltzmann entropy of the current driving scenario represents the absolute complexity of the current driving scenario. Specifically, when the number of microstates in the current driving scenario is fixed, the Boltzmann entropy value remains unchanged even if the distribution of the system's internal states changes. As the number of microstates in the current driving scenario increases, the larger the Boltzmann entropy value, the more complex the current driving environment.

[0041] On this basis, the vehicle head-up display levels are pre-divided, and the number and content of display elements corresponding to each vehicle head-up display level, as well as the Boltzmann entropy range corresponding to each vehicle head-up display level are determined.

[0042] It can be understood that the larger the starting value of the Boltzmann entropy range corresponding to the normal display level of the vehicle, the more complex the driver's current driving environment is, the fewer the corresponding display elements are, and the more concise the content is.

[0043] The present invention introduces the Boltzmann entropy to quantify the absolute complexity of the current driving scene, constructs a complete micro-state mapping system to realize the extraction of the micro-state number in the driving scene, accurately calculates the Boltzmann entropy of the current driving scene, and determines the corresponding on-board head-up display level according to the Boltzmann entropy of the driving scene for display, so as to control the number of display elements and reduce the driving load of the driver in complex scenes.

[0044] In the in-vehicle head-up display method based on Boltzmann entropy of the present invention, the step of calculating the Boltzmann entropy of the current driving scene specifically includes: Selecting a weight for each type of microstate according to a road type to which the current driving scene belongs, wherein a mapping relationship between the road type and the weight for each type of microstate is predefined; It is understandable that in different driving scenarios, different categories of micro-states have different effects on the complexity of the driving scenario.

[0045] For example, when the micro-state includes people and cars, if the current driving scene is an urban road, the appearance of pedestrians on the road has a greater impact on the complexity of the driving scene and the driver's load. If the current driving scene is a highway, the probability of pedestrians appearing is greatly reduced, and the impact of cars on the complexity of the driving scene and the driver's load increases.

[0046] Therefore, it is necessary to dynamically adjust the weight of each microstate to make it more suitable for the current driving scenario. However, it should be noted that if the weight is completely based on the real-time driving scene, such as the acquisition and adjustment of real-time in-vehicle video content, the computing resources required are high and it is difficult to apply to the in-vehicle environment. At the same time, the large amount of calculation will also affect the real-time performance of the Boltzmann entropy calculation.

[0047] To this end, in this embodiment, the current driving scene is represented based on the road type to which the current driving scene belongs, and the weight of each type of microstate under each road type is predetermined, thereby obtaining a mapping relationship table between the road type and the weight of each type of microstate.

[0048] On this basis, the road type to which the current driving scene belongs is determined, and the weight of each type of microstate can be queried and selected according to the mapping relationship table to realize the calculation of the weighted microstate number.

[0049] Optionally, under each road type, the weight of each type of micro-state can be an empirical value or determined by a machine learning model.

[0050] As a preference, based on the results of driver cognition experiments, the weight of each micro-feature under each road type is defined, wherein the road types include expressways, urban main roads, urban secondary roads, rural roads and tunnels.

[0051] Specifically, this embodiment aims to obtain the driver's real behavior and physiological data by conducting cognitive experiments on the driver, so as to obtain accurate micro-state weights with interpretability.

[0052] In a specific embodiment, an eye movement experiment is conducted on the driver to implement a cognitive experiment on the driver, so as to obtain the weight of each type of micro-state.

[0053] The eye movement experiment was conducted under different conditions, including highways, urban main roads, urban secondary roads, rural roads, and tunnels. The human, vehicle, road, and environmental elements were set up separately. By recording and analyzing the eye movement data of 31 testers under these conditions, the attention distribution of drivers under different microscopic conditions was obtained, and the weight matrix was obtained as follows: .

[0054] Therefore, the mapping relationship table constructed based on the results of the driver cognition test is shown in Table 1 below: Table 1

[0055] Specifically, when the road type is highway, the weight of people is considered to be low because highways mainly serve vehicle traffic and have fewer pedestrians; the weight of cars is considered to be high because the main purpose of highways is efficient vehicle transportation; the weight of roads is considered to be high because road conditions have an important impact on the safety and efficiency of vehicle driving; the weight of the environment is considered to be moderate because environmental factors such as weather and visibility have an impact on highway traffic, but are not the main factors.

[0056] When the road type is urban trunk roads, the weight of people is considered moderate because the flow of pedestrians and non-motor vehicles on urban trunk roads is large; the weight of cars is considered high because urban trunk roads are the main channels of urban traffic; the weight of roads is considered moderate because road conditions have an impact on traffic flow; the weight of the environment is considered low because environmental factors have relatively little impact on urban trunk roads.

[0057] When the road type is urban secondary roads, the weight of people is considered to be higher because the flow of pedestrians and non-motor vehicles on urban secondary roads is large; the weight of cars is considered to be moderate because urban secondary roads mainly serve the traffic within the region; the weight of roads is considered to be moderate because the road conditions have an impact on traffic flow; the weight of the environment is considered to be moderate because environmental factors have a certain impact on urban secondary road traffic.

[0058] When the road type is rural road, the weight of people is considered to be higher because there is a large flow of pedestrians and non-motor vehicles on rural roads; the weight of cars is considered to be moderate because rural roads mainly serve the traffic within the region; the weight of roads is considered to be moderate because road conditions have an impact on traffic flow; the weight of the environment is considered to be lower because environmental factors have relatively little impact on rural roads.

[0059] In the case of a tunnel, the weight of people is considered to be low because tunnels mainly serve vehicle traffic and have fewer pedestrians; the weight of vehicles is considered to be the highest because the smoothness and safety of tunnel traffic mainly depends on the driving conditions of vehicles; the weight of roads is considered to be moderate because road conditions have an important impact on the safety and efficiency of vehicle driving; the weight of the environment is considered to be moderate because environmental factors such as ventilation and lighting have an impact on tunnel traffic.

[0060] Through the above method, a mapping relationship table between road types and weights is pre-built. The current driving road type is determined based on the real-time vehicle video, and the corresponding micro-state weight can be directly selected by looking up the mapping relationship table.

[0061] Determining the Boltzmann entropy state number of each microscopic feature data according to each microscopic feature data of each type of microscopic state, wherein a mapping relationship between each microscopic feature data and the Boltzmann entropy state number is predefined; In order to normalize the microscopic feature data of different scales to the same range and calculate the Boltzmann entropy, in this embodiment, a mapping relationship table between each microscopic feature data and the Boltzmann entropy state number is predefined, as shown in Table 2 below: Table 2

[0062] Specifically, the Boltzmann entropy state number hierarchy is predefined, and then a mapping relationship is constructed between each micro-feature data item and the hierarchy. For example, when the pedestrian density is 0, the corresponding Boltzmann entropy state number is 1. When the pedestrian density is 1, the corresponding Boltzmann entropy state number is 2. Similarly, when the pedestrian density is 7, the corresponding Boltzmann entropy state number is 6. For another example, when the lighting conditions are good, the corresponding Boltzmann entropy state number is 1. That is, in the third column of Table 2, the Boltzmann entropy state number levels increase from 1 to right.

[0063] On this basis, after determining the various micro-feature data of the current driving scene, the Boltzmann entropy state number corresponding to each micro-feature data can be determined.

[0064] In a specific embodiment, the road type of scenario 1 is a highway, and the remaining microscopic feature data and their corresponding Boltzmann entropy state numbers are shown in Table 3 below: Table 3

[0065] Scenario 1 represents a driving scenario where a vehicle is traveling on a highway on a sunny day. There are no pedestrians on the road. The vehicle speed is between 106 km / h and 4 vehicles on the road. The vehicle types on the road are a mixture of small and large vehicles (as shown in Table 2). There are 3 lanes and the lighting conditions are good.

[0066] The road type in scenario 2 is a secondary urban road. The remaining microscopic feature data and their corresponding Boltzmann entropy state numbers are shown in Table 4 below: Table 4

[0067] Scenario 2 represents a driving scenario in which a vehicle is traveling on a secondary urban road on a sunny day. There are three pedestrians on the road, the vehicle is traveling at a speed of 45 km / h, the vehicle density is seven, the vehicle type is a small car, there are two lanes, and the lighting conditions are good.

[0068] Calculate the first product of all Boltzmann entropy state numbers corresponding to each type of microstate; The first product of each type of microstate is weighted using the weight of each type of microstate, and the weighted product of all microstates is calculated to obtain the Boltzmann entropy of the current driving scene.

[0069] On this basis, the Boltzmann entropy of the current driving scene can be calculated based on the Boltzmann entropy state number corresponding to the microscopic state of the current driving scene.

[0070] Specifically, the calculation formula of Boltzmann entropy is: ; Where, is a constant, is the number of possible microscopic states of the system. Therefore, the entropy value increases monotonically with the increase of the number of states. The bigger, S The higher the value, the lower the entropy value. When the number of states is fixed, the entropy value remains unchanged, that is, Even if the internal state distribution of the system changes, the entropy value remains constant.

[0071] because is a constant. In order to simplify the calculation, the simplified Boltzmann entropy calculation formula is: ; Furthermore, in order to highlight the impact of different types of microstates on driver load in different driving scenarios, each type of microstate is weighted according to its weight, and the resulting weighted Boltzmann entropy calculation formula is: ; Where, represents the weighted microstate number, represents the product symbol, Indicates the i Micro-state data, Indicates the i The weight corresponding to each micro-feature data, n Represents the total number of micro-feature data.

[0072] Among them, the i-th micro-state data It is the first product of the number of Boltzmann entropy states corresponding to each type of microstate.

[0073] Taking scenario 1 in the above text as an example, the calculated Boltzmann entropy of scenario 1 is: ; ; That is, the first products corresponding to the micro-state categories of human, vehicle, road, and environment are 1, 24, 9, and 1, respectively. After weighting the weight of each type of micro-state as an exponent, the product of the weighted micro-state numbers of all categories is calculated to obtain the final weighted micro-state number.

[0074] Taking the above scenario 2 as an example, the Boltzmann entropy of scenario 2 is calculated as: ; ; That is, the first products corresponding to the micro-state categories of human, vehicle, road, and environment are 4, 9, 18, and 1, respectively.

[0075] Through the above method, the calculation of the Boltzmann entropy of the current driving scene is achieved, so that the HUD display level is determined according to the Boltzmann entropy of the current driving scene.

[0076] It's important to emphasize that the absolute complexity of a driving scenario, as quantified by the Boltzmann entropy, remains constant regardless of how the microstates interact with each other, as long as the number of microstates in the scenario remains constant. In other words, the entropy remains constant if the driver's speed remains constant, and the number of vehicles, pedestrians, and / or available lanes ahead of them remains constant.

[0077] This is because cognitive experiments on drivers have shown that the number of vehicles, pedestrians and / or drivable lanes ahead remains unchanged. Even if the status of the vehicle ahead changes, such as deceleration or lane change in the lane ahead, the driver's driving load remains unchanged. This is because the driver naturally maintains attention on the status of the vehicles and pedestrians ahead during normal driving. When the number of vehicles and pedestrians ahead that require attention remains unchanged, the driver's load does not change.

[0078] On this basis, this embodiment chooses to introduce Boltzmann entropy into the calculation of driving scene complexity to characterize the absolute complexity of the driving scene. By measuring the absolute complexity of the driving scene, the driving burden of the driver in the current scene is characterized, thereby configuring the appropriate number of HUD display elements for the driver.

[0079] In addition, compared with using a neural network model to determine dynamic weights, this embodiment selects weights based on road types and calculates entropy values through a table lookup method, reducing dependence on chip computing power and meeting the real-time requirements of the vehicle system.

[0080] In the vehicle-mounted head-up display method based on Boltzmann entropy of the present invention, the vehicle-mounted head-up display level includes a first level representing low-complexity scenes, a second level representing medium-complexity scenes, and a third level representing high-complexity scenes, and from the first level to the third level, the number of display elements included in the vehicle-mounted head-up display level decreases step by step.

[0081] In this embodiment, the vehicle's normal display level, i.e., the HUD display level, is divided into three levels, namely the first level, the second level, and the third level in order of increasing scene complexity. Therefore, from the first level to the third level, the number of display elements contained in each level of display decreases step by step.

[0082] Specifically, in the first level display level, the display elements include complete navigation information; in the second level display level, the display elements include simplified secondary information; and in the third level display level, the display elements only include key instructions.

[0083] Among them, when the calculated Boltzmann entropy is less than or equal to 1.0, the current driving scene is considered to belong to the first level; when the calculated Boltzmann entropy is greater than 1.0 and less than 1.5, the current driving scene is considered to belong to the second level; when the calculated Boltzmann entropy is greater than or equal to 1.5, the current driving scene is considered to belong to the third level.

[0084] In the in-vehicle head-up display method based on Boltzmann entropy of the present invention, the step of obtaining each microscopic feature data of each type of microscopic state of the current driving scene specifically includes: Extracting an in-vehicle video of a current driving process, wherein the in-vehicle video represents the driver's visual field environment; Extracting image frames from the vehicle-mounted video and using an object detection model to detect object bounding boxes corresponding to people, vehicles, and road signs in each frame; Each item of microscopic feature data of each type of microscopic state of the current driving scene is determined based on the detection result of the target detection model.

[0085] The vehicle video stream of the current driving process is extracted. In this embodiment, the video stream resolution is 1280×70 and the frame rate is 30fps. Continuous frame images and vehicle sensor data, including vehicle speed and GPS position, are extracted.

[0086] It is understandable that the content of the in-vehicle video is considered to be able to represent the driver's visual field environment, that is, the absolute complexity of the image frames of the in-vehicle video reflects the driver's driving burden.

[0087] A lightweight target detection model (such as YOLOv8-tiny) is used to apply the trained model to real-time extracted image frames to detect vehicles, pedestrians, and traffic signs. The model outputs information such as the bounding box coordinates, category, and confidence of the identified target, and then retains the target bounding box with a confidence level ≥ 0.5. The vehicle, pedestrian, traffic sign, and other data for each frame are calculated, and based on this, the pedestrian density Pt, pedestrian distance Dt, road vehicle density Nt, type composition Tt, number of lanes Lt, and traffic sign density Mt are calculated; the road type Rt, vehicle speed St, and weather information Et are obtained based on Beidou positioning information, and the light conditions Bt are obtained based on the light sensor.

[0088] It is understandable that when the defined microscopic feature data is different, different sensing devices or detection models can be selected and corresponding settings and calculations can be performed.

[0089] The following describes the vehicle-mounted head-up display system based on Boltzmann entropy provided by the present invention. The vehicle-mounted head-up display system based on Boltzmann entropy described below and the vehicle-mounted head-up display method based on Boltzmann entropy described above can refer to each other.

[0090] like Figure 2 As shown, the vehicle-mounted head-up display system based on Boltzmann entropy of the present invention includes an acquisition module 201, a determination module 202 and a display module 203; An acquisition module 201 is configured to acquire each microscopic feature data of each type of microscopic state of a current driving scene and calculate the Boltzmann entropy of the current driving scene; Real-time vehicle video is acquired, and each microscopic feature data of each type of microscopic state of the current driving scene is determined based on each frame image of the vehicle video.

[0091] The in-vehicle video is a video acquired by an image acquisition device installed in front of the vehicle, so that the video content simulates the driver's field of view.

[0092] In a feasible implementation, the image acquisition device may be a camera shared with an intelligent driving / assisted driving system, which is typically installed at the center of the outer side of the vehicle's front windshield.

[0093] In other feasible embodiments, the image acquisition device may also be independently provided, for example, installed inside the vehicle and located in front of the main driver's seat, so that the shooting range of the image acquisition device can be closer to the driver's field of view, so as to better quantify the current driving scene based on the driver's posture field of view rather than the machine's field of view.

[0094] According to the definition of each micro-feature data, each micro-feature data of each type of micro-state in the current driving scene is extracted and calculated based on each frame image of the on-board video. On this basis, the micro-feature data is further discretized into state classification to obtain the Boltzmann state number corresponding to each micro-feature data, and then the Boltzmann entropy calculation formula is used to calculate the Boltzmann entropy of the current driving scene.

[0095] A determination module 202 is configured to determine a vehicle head-up display level according to the Boltzmann entropy of the current driving scene; The calculated Boltzmann entropy of the current driving scenario represents the absolute complexity of the current driving scenario. Specifically, when the number of microstates in the current driving scenario is fixed, the Boltzmann entropy value remains unchanged even if the distribution of the system's internal states changes. As the number of microstates in the current driving scenario increases, the larger the Boltzmann entropy value, the more complex the current driving environment.

[0096] On this basis, the vehicle head-up display levels are pre-divided, and the number and content of display elements corresponding to each vehicle head-up display level, as well as the Boltzmann entropy range corresponding to each vehicle head-up display level are determined.

[0097] It can be understood that the larger the starting value of the Boltzmann entropy range corresponding to the normal display level of the vehicle, the more complex the driver's current driving environment is, the fewer the corresponding display elements are, and the more concise the content is.

[0098] The present invention introduces the Boltzmann entropy to quantify the absolute complexity of the current driving scene, constructs a complete micro-state mapping system to realize the extraction of the micro-state number in the driving scene, accurately calculates the Boltzmann entropy of the current driving scene, and determines the corresponding on-board head-up display level according to the Boltzmann entropy of the driving scene for display, so as to control the number of display elements and reduce the driving load of the driver in complex scenes.

[0099] The display module 203 is configured to adjust and display display elements according to the determined vehicle head-up display level.

[0100] Optionally, the display module includes a HUD display, and the HUD display adjusts the number of display elements for projection display based on the determined normal vehicle display level.

[0101] The vehicle's normal display level, that is, the HUD display level, is divided into three levels, namely level one, level two and level three in order of increasing scene complexity. Therefore, from level one to level three, the number of display elements contained in each level of display decreases step by step.

[0102] Specifically, in the first level display level, the display elements include complete navigation information; in the second level display level, the display elements include simplified secondary information; and in the third level display level, the display elements only include key instructions.

[0103] Among them, when the calculated Boltzmann entropy is less than or equal to 1.0, the current driving scene is considered to belong to the first level; when the calculated Boltzmann entropy is greater than 1.0 and less than 1.5, the current driving scene is considered to belong to the second level; when the calculated Boltzmann entropy is greater than or equal to 1.5, the current driving scene is considered to belong to the third level.

[0104] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3 As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 830 communicate with each other via the communications bus 340. The processor 310 may call logic instructions in the memory 330 to execute a vehicle head-up display method based on Boltzmann entropy. The method includes: mapping the driving scene into several types of micro-states and defining quantifiable data indicators for each type of micro-state as micro-features of each type of micro-state; obtaining each micro-feature data item of each type of micro-state in the current driving scene and calculating the Boltzmann entropy of the current driving scene; and determining a vehicle head-up display level based on the Boltzmann entropy of the current driving scene, wherein different vehicle head-up display levels include different numbers of display elements.

[0105] Furthermore, the logic instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0106] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the in-vehicle head-up display method based on Boltzmann entropy provided by the above methods. The method includes: mapping the driving scene into several types of micro-states, and defining quantifiable data indicators for each type of micro-state as micro-features of each type of micro-state; obtaining each micro-feature data of each type of micro-state in the current driving scene, and calculating the Boltzmann entropy of the current driving scene; determining the in-vehicle head-up display level according to the Boltzmann entropy of the current driving scene, wherein different in-vehicle head-up display levels include different numbers of display elements.

[0107] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the in-vehicle head-up display method based on Boltzmann entropy provided by the above-mentioned methods, the method comprising: mapping the driving scene into several types of micro-states, and defining quantifiable data indicators for each type of micro-state as micro-features of each type of micro-state; obtaining each micro-feature data of each type of micro-state in the current driving scene, and calculating the Boltzmann entropy of the current driving scene; determining the in-vehicle head-up display level according to the Boltzmann entropy of the current driving scene, wherein different in-vehicle head-up display levels include different numbers of display elements.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0109] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 various embodiments of the present invention.

Claims

1. A vehicle head-up display method based on Boltzmann entropy, characterized in that: include: Map the driving scene into several types of micro-states, and define quantifiable data indicators for each type of micro-state as the micro-features of each type of micro-state; Obtaining each microscopic feature data of each type of microscopic state of the current driving scene, and calculating the Boltzmann entropy of the current driving scene; The vehicle head-up display level is determined according to the Boltzmann entropy of the current driving scene, wherein different vehicle head-up display levels include different numbers of display elements.

2. The vehicle-mounted head-up display method based on Boltzmann entropy according to claim 1, characterized in that: The step of calculating the Boltzmann entropy of the current driving scene specifically includes: Selecting a weight for each type of microstate according to a road type to which the current driving scene belongs, wherein a mapping relationship between the road type and the weight for each type of microstate is predefined; Determining the Boltzmann entropy state number of each microscopic feature data according to each microscopic feature data of each type of microscopic state, wherein a mapping relationship between each microscopic feature data and the Boltzmann entropy state number is predefined; Calculate the first product of all Boltzmann entropy state numbers corresponding to each type of microstate; The first product of each type of microstate is weighted using the weight of each type of microstate, and the weighted product of all microstates is calculated to obtain the Boltzmann entropy of the current driving scene.

3. The vehicle-mounted head-up display method based on Boltzmann entropy according to claim 1, characterized in that: Before the step of selecting the weight of each micro-feature according to the road type to which the current driving scene belongs, the method further includes: Based on the results of the driver cognition experiment, the weight of each micro-feature under each road type is defined, wherein the road types include expressways, urban main roads, urban secondary roads, rural roads and tunnels.

4. The vehicle-mounted head-up display method based on Boltzmann entropy according to claim 1, characterized in that: The vehicle-mounted head-up display levels include a first level representing low-complexity scenes, a second level representing medium-complexity scenes, and a third level representing high-complexity scenes, and from the first level to the third level, the number of display elements included in the vehicle-mounted head-up display level decreases step by step.

5. The vehicle-mounted head-up display method based on Boltzmann entropy according to claims 1-4, characterized in that: The microscopic states are defined as four categories, namely, people, vehicles, roads and environments.

6. The method for in-vehicle head-up display based on Boltzmann entropy according to claim 5, characterized in that: The step of obtaining each item of microscopic feature data of each type of microscopic state of the current driving scene specifically includes: Extracting an in-vehicle video of a current driving process, wherein the in-vehicle video represents the driver's visual field environment; Extracting image frames from the vehicle-mounted video and using an object detection model to detect object bounding boxes corresponding to people, vehicles, and road signs in each frame; Each item of microscopic feature data of each type of microscopic state of the current driving scene is determined based on the detection result of the target detection model.

7. A vehicle head-up display system based on Boltzmann entropy, characterized in that: include: an acquisition module, configured to acquire each microscopic feature data of each type of microscopic state of the current driving scene and calculate the Boltzmann entropy of the current driving scene; a determination module, configured to determine a vehicle head-up display level according to the Boltzmann entropy of the current driving scene; The display module is used to adjust and display display elements according to the determined vehicle head-up display level.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the vehicle-mounted head-up display method based on Boltzmann entropy as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the in-vehicle head-up display method based on Boltzmann entropy as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the in-vehicle head-up display method based on Boltzmann entropy as claimed in any one of claims 1 to 6 is implemented.

Images