Visual field adjusting method of electronic rearview mirror and electronic equipment

Abstract

The embodiment of the invention discloses a visual field adjusting method of an electronic rearview mirror and electronic equipment. According to the method, a target scene template is determined from a scene template library through at least one of current driving state information, road information and environment information of a vehicle so as to recommend a user to perform view adjustment on an electronic rearview mirror of the vehicle by adopting configuration parameters corresponding to the target scene template. Therefore, the technical problem that the view configuration of the electronic rearview mirror is rigid can be solved, it is ensured that the electronic rearview mirror can provide the optimal view range for a user under different working conditions of the vehicle, the view blind area of the user is effectively reduced, the diversified requirements of the user under different driving scenes are met, and the driving safety is improved.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and in particular to a method and electronic device for adjusting the field of view of an electronic rearview mirror. Background Technology

[0002] With the continuous evolution of intelligent connected vehicle technology, electronic rearview mirrors, as an innovative replacement for traditional physical rearview mirrors, have become a core component of the vehicle's active safety system. Electronic rearview mirrors use high-precision cameras to collect real-time images of the vehicle's surroundings. After optimization by an image processing unit, clear rear and side view information is presented on the in-vehicle display screen, providing crucial visual support for the driver.

[0003] However, electronic rearview mirrors generally suffer from rigid field configurations, causing drivers to adjust parameters too frequently when switching between scenarios. This not only makes the operation process cumbersome and time-consuming, but also easily distracts the driver in emergency situations, resulting in a poor user experience. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a method and electronic device for adjusting the field of view of an electronic rearview mirror, solving the technical problem of rigid field of view configuration of electronic rearview mirrors, ensuring that the electronic rearview mirror can provide users with the optimal field of view under different operating conditions, and meeting the diverse needs of users in different driving scenarios.

[0005] In a first aspect, this application provides a method for adjusting the field of view of an electronic rearview mirror, including:

[0006] Acquire current vehicle driving status information, road information, and environmental information; Based on at least one of the following information: driving status information, road information, and environmental information, a target scene template is determined from the scene template library; the scene template library includes at least one preset scene template. Generate and display recommended information for the target scene template, and after receiving the user's application instructions for the recommended information, adjust the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0007] Secondly, this application also provides a field-of-view adjustment device for an electronic rearview mirror, comprising: The acquisition unit is used to acquire the vehicle's current driving status information, road information, and environmental information. The determining unit is used to determine a target scene template from a scene template library based on at least one of driving status information, road information, and environmental information; the scene template library includes at least one preset scene template. The adjustment unit is used to generate and display recommended information for the target scene template, and after receiving the user's application instructions for the recommended information, adjusts the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the field of view adjustment method of the electronic rearview mirror as provided in the first aspect above.

[0009] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the field-of-view adjustment method for the electronic rearview mirror provided in the first aspect.

[0010] Fifthly, embodiments of this application also provide a computer program product, including a computer program or instructions, wherein the computer program or instructions are executed by a processor to provide the field of view adjustment method for the electronic rearview mirror provided in the first aspect.

[0011] The electronic rearview mirror field of view adjustment method provided in this application determines a target scene template from a scene template library by using at least one of the vehicle's current driving status information, road information, and environmental information. This recommends that the user adjust the electronic rearview mirror field of view using the configuration parameters corresponding to the target scene template. This solves the technical problem of rigid field of view configuration of electronic rearview mirrors, ensuring that the electronic rearview mirror can provide the user with the optimal field of view under different operating conditions. It effectively reduces the user's blind spots, meets the diverse needs of users in different driving scenarios, and improves driving safety. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 An application scenario diagram of the electronic rearview mirror field of view adjustment method provided in the embodiments of this application; Figure 2 A flowchart illustrating the field of view adjustment method for an electronic rearview mirror provided in an embodiment of this application; Figure 3 An architecture diagram of the scene template library provided in the embodiments of this application; Figure 4This is a schematic diagram of the data acquisition and processing flow of various sensors provided in the embodiments of this application; Figure 5 A schematic block diagram illustrating dual-mode switching provided in an embodiment of this application; Figure 6 A schematic diagram illustrating the process of creating a user-defined scene template provided in this application embodiment; Figure 7 An architecture diagram of the multi-user personalized configuration management architecture provided in the embodiments of this application; Figure 8 An architecture diagram of the AI ​​learning and recommendation engine provided in the embodiments of this application; Figure 9 A schematic diagram illustrating the smooth transition of scene switching provided in the embodiments of this application; Figure 10 A schematic diagram of the field of view adjustment method for an electronic rearview mirror provided in an embodiment of this application; Figure 11 A schematic block diagram of the field of view adjustment device for an electronic rearview mirror provided in an embodiment of this application; Figure 12 A schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0016] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0017] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0018] In related technologies, the requirements for field of vision configuration vary significantly depending on the road environment in actual driving scenarios: when driving on highways, it is necessary to expand the rear field of vision to monitor distant vehicles, while reducing the proportion of sky and ground; urban parking operations require enhancing the visibility of the lower edge of the vehicle body and ground markings, and increasing the brightness of the screen to adapt to low-light environments such as underground parking lots; driving on mountain curves requires expanding the horizontal field of vision to cover the side blind spots.

[0019] However, electronic rearview mirrors generally suffer from rigid field configurations, with most only offering fixed modes or basic manual adjustment functions. This causes drivers to adjust parameters too frequently when switching between scenarios, making the operation process cumbersome and time-consuming. In addition, it can easily distract the driver in emergency situations, resulting in a poor user experience.

[0020] For example, Chinese invention patent application number 202111146613.2 discloses an electronic rearview mirror mode switching method. Although this method attempts to trigger the display mode switching through driving operations such as turn signals and gear status, the electronic rearview mirror is unable to provide a precise and adaptable vision solution when facing complex and ever-changing actual driving environments, which seriously restricts the technical potential of electronic rearview mirrors in improving driving safety.

[0021] To address this, this application provides a method for adjusting the field of view of an electronic rearview mirror. By using at least one of the vehicle's current driving status information, road information, and environmental information, a target scene template is determined from a scene template library. This method recommends that the user adjust the field of view of the vehicle's electronic rearview mirror using the configuration parameters corresponding to the target scene template. This solves the technical problem of rigid field of view configuration of electronic rearview mirrors, ensuring that the electronic rearview mirror can provide the user with the optimal field of view under different operating conditions. It effectively reduces the user's blind spots, meets the diverse needs of users in different driving scenarios, and improves driving safety.

[0022] Please see Figure 1 , Figure 1 This diagram illustrates an application scenario of the electronic rearview mirror field-of-view adjustment method provided in this application. The electronic rearview mirror field-of-view adjustment method provided in this application can be applied to a vehicle's in-vehicle terminal, such as a first in-vehicle terminal 111 of a first vehicle 110 or / and a second in-vehicle terminal 121 of a second vehicle 120. The vehicle's in-vehicle terminal can communicate with a cloud 130. The electronic rearview mirror field-of-view adjustment method provided in this application can be synchronized with the cloud 130 and used across vehicles. Furthermore, the electronic rearview mirror field-of-view adjustment method can employ… Figure 10 The architecture diagram shown is used.

[0023] The vehicle terminal is equipped with a processor, memory, and display. The processor can be an automotive-grade SoC chip with a six-core ARM Cortex-A72@2.0GHz and a Mali-G76 GPU. The memory can consist of 4GB of LPDDR4 and 16GB of eMMC. The display can be a 12.3-inch LCD screen with a resolution of 1920×720 and support touch.

[0024] In addition, user scenario templates and usage habit data can be uploaded to cloud-based 130 server storage. When a user changes vehicles or uses multiple vehicles, they can synchronize their personal configurations to the new vehicle by logging into their cloud account, achieving one-time configuration for all applications. Here, "user" can be understood as the driver.

[0025] Meanwhile, cloud synchronization can employ an incremental synchronization strategy, transmitting only changed configuration items to reduce data transmission volume. End-to-end encryption can be used during the synchronization process to protect user privacy and data security.

[0026] It should be noted that the application scenarios described in the following embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0027] The following is a detailed description of the field of view adjustment method of the electronic rearview mirror provided in this application.

[0028] like Figure 2 As shown, the method includes the following steps S210~S230.

[0029] S210: Obtain the vehicle's current driving status information, road information, and environmental information; S220. Determine a target scene template from the scene template library based on at least one of the following: driving status information, road information, and environmental information; the scene template library includes at least one preset scene template. S230: Generate and display recommended information for the target scene template, and after receiving the user's application instruction for the recommended information, adjust the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0030] In this embodiment, driving status information can be used to determine whether the vehicle is stationary, driving at low speed, driving at high speed, reversing, or changing lanes. It can be understood as data on the current operating status of the vehicle. Driving status information can be obtained from real-time information collected by various sensors installed on the vehicle, such as the vehicle's instantaneous speed, gear status, steering angle, and turn signal.

[0031] Road information can be understood as characteristic data related to the road where the vehicle is currently located, such as road type (e.g., highway, urban road, rural road, mountain road), road curvature or number of lanes, etc. Road information can be obtained through in-vehicle navigation systems, electronic map data interfaces or visual recognition systems.

[0032] Environmental information can be understood as characteristic data of the external environment surrounding a vehicle, such as current weather conditions (e.g., rainfall intensity, fog level), ambient light intensity (e.g., daytime, dusk, nighttime), or the distance between the vehicle and obstacles. Environmental information can be collected by various sensors such as rain sensors, ambient light sensors, and reversing radar.

[0033] The scene template library can be understood as a collection that stores a variety of preset or custom view configuration schemes. Each configuration scheme can be called a scene template. The scene template contains view adjustment parameters and trigger conditions optimized for specific driving scenarios.

[0034] Preset scene templates can be understood as scene templates pre-configured at the factory. Preset scene templates are designed for common driving scenarios, such as highway driving scenarios, urban parking scenarios, and night driving scenarios.

[0035] The scene template library provides a complete CRUD (create, read, update, delete) operation interface. Users can create new templates, edit the parameters of existing templates, delete templates that are no longer needed, search for templates by scene name or usage frequency, and adjust the sorting position of templates in the list. Furthermore, the scene template library uses a structured data format for storage, with each template containing metadata such as a unique identifier, creation time, modification time, usage count, and usage duration, facilitating management and statistical analysis.

[0036] Preset scene templates are vision schemes optimized for common driving scenarios and pre-configured by the system at the factory. Preset scene templates can include highway driving scene templates, urban parking scene templates, mountain road curve scene templates, rainy / dewy weather scene templates, and night driving scene templates, etc.

[0037] In the high-speed driving scene template, the field of view focuses on the area behind the vehicle, the cropped area is shifted downward to reduce the proportion of sky and increase the field of view of the road behind, and the contrast of the image is moderately improved to facilitate the identification of distant vehicles.

[0038] The urban parking scene template focuses on covering the lower edge of the vehicle and the ground, with the cropped area extended downwards to increase the field of view of the ground and nearby obstacles. At the same time, the screen brightness is increased to adapt to the low-light environment of the parking lot, and parking auxiliary reference lines can be superimposed.

[0039] The field of view in the mountain road curve scene template expands to the side, increases the horizontal field of view, reduces the vertical proportion of sky and ground, strengthens the coverage of side blind spots, and can also enable the blind spot warning function.

[0040] The purpose of the rain and fog weather scene template is to improve image contrast and brightness, enable defogging algorithms, enhance image clarity, and help users obtain clearer rear view information in low visibility conditions.

[0041] Night driving scene templates can significantly improve image brightness, enable night vision enhancement algorithms and anti-glare functions to reduce glare interference from headlights of vehicles behind, while maintaining image clarity.

[0042] The preset scene templates contain complete parameter configurations, including field of view cropping parameters, image enhancement parameters, display mode parameters, and accessibility function on / off parameters.

[0043] Among them, the field of view cropping parameter defines the position of the top, bottom, left and right boundaries of the image, and the field of view cropping parameter can be expressed as a percentage relative to the original video image.

[0044] Image enhancement parameters include brightness adjustment values. Contrast adjustment value and saturation adjustment value The value range is usually 100%. .

[0045] The display mode parameter defines one of the following: normal mode, wide-angle mode, or split-screen mode.

[0046] The assist function switch parameters define the activation status of various assist functions (such as blind spot warning, lane line overlay, defogging algorithm, etc.).

[0047] For example, the field of view cropping parameters of the high-speed driving scene template are 20% for the upper boundary, 65% for the lower boundary, 10% for the left boundary, and 90% for the right boundary (percentage positions relative to the original image); the image enhancement parameters are brightness ΔB=+5%, contrast ΔC=+10%, and saturation ΔS=0%; the display mode is normal mode; the auxiliary functions are blind spot warning enabled and lane line overlay enabled; the triggering condition is: GPS speed greater than 80 km / h AND road type equal to highway.

[0048] The field-of-view cropping parameters for the urban parking scene template are: upper boundary 10%, lower boundary 75%, left boundary 5%, and right boundary 95%; image enhancement parameters are ΔB=+15%, ΔC=+5%, and ΔS=0%; the display mode is close-up mode; and the auxiliary function is reversing reference line enabled. The triggering conditions are: reverse gear engaged OR (GPS speed less than 10 km / h AND minimum obstacle distance less than 2 meters).

[0049] The field-of-view cropping parameters for the rain and fog weather scene template inherit the current configuration; the image enhancement parameters are ΔB=+20%, ΔC=+25%, ΔS=-5%; the display mode is normal mode; the auxiliary function is defogging algorithm enabled; the triggering condition is: rainfall intensity greater than 50% OR ambient illuminance less than 200 lux (during daytime).

[0050] The field of view cropping parameters for the night driving scene template inherit the current configuration; the image enhancement parameters are ΔB=+30%, ΔC=+15%, ΔS=0%; the display mode is normal mode; the auxiliary functions are night vision enhancement on and anti-glare on; the trigger conditions are: ambient illuminance less than 50 lux OR time period between 18:00 and 06:00 the next day.

[0051] The field-of-view cropping parameters for the mountain road curve scene template are: upper boundary 25%, lower boundary 60%, left boundary 5%, and right boundary 95% (expanding the horizontal field of view); image enhancement parameters are ΔB=+10%, ΔC=+10%, and ΔS=0%; display mode is wide-angle mode; auxiliary function is enabled for blind spot warning; trigger condition is: road type equals mountain road OR lane line curvature is greater than 0.3.

[0052] The target scenario template can be understood as a scenario template that is highly consistent with the current driving situation, identified from the scenario template library based on the currently acquired driving status information, road information, and environmental information.

[0053] Recommended information can be understood as suggestions presented to users in a non-intrusive manner by the system based on the identified target scenario template. Recommended information may include the name, icon, and operation options of the recommended template.

[0054] In this application, the recommendation information can be presented as a floating card, popping up in the edge area (usually the upper right corner or lower) of the central control display screen or electronic rearview mirror display screen. The card content may include: the name and icon of the recommended scene template, a brief description of the recommendation reason (such as high-speed scene detected), and two operation buttons (Apply and Ignore). At the same time, the recommendation card uses a semi-transparent background to not completely obscure the main display content, ensuring that the user's line of sight is not excessively disturbed.

[0055] An application instruction can be understood as a confirmation instruction that a user sends to the system through a human-computer interaction interface (such as clicking the "Apply" button on the screen) after receiving recommendation information, indicating that the user accepts and wants to apply the currently recommended scenario template.

[0056] Configuration parameters can be understood as the specific adjustment values ​​and function switches that constitute the scene template. Configuration parameters include, but are not limited to, the boundary position of the field of view cropping area, the brightness, contrast, and saturation adjustment values ​​of the image, the display mode (such as wide-angle, split screen), and the enabled status of auxiliary functions (such as blind spot warning, defogging algorithm).

[0057] Electronic rearview mirrors can be understood as a new type of device that replaces traditional physical rearview mirrors. They use cameras to collect images of the vehicle's surroundings and display them on an in-car screen, providing users with rear and side view information.

[0058] Specifically, in the process of adjusting the field of view of the electronic rearview mirror, this application can adjust the parameters of the electronic rearview mirror display to achieve the field of view adjustment, including but not limited to the field of view cropping area, image brightness, contrast, saturation, display mode, and auxiliary function switch.

[0059] In this application, during the process of acquiring the vehicle's current driving status information, road information, and environmental information, vehicle speed signals, gear signals, turn signal signals, etc., can be obtained through the vehicle's controller local area network (CAN) bus to obtain the vehicle's current driving status information; at the same time, the current road type identifier (such as highway, urban road, etc.) can be read from the in-vehicle navigation system or a pre-installed electronic map database to obtain the vehicle's current road information; in addition, this application can also detect rainfall intensity through a rain sensor installed on the outside of the vehicle and detect ambient light through an ambient light sensor to obtain the vehicle's current environmental information.

[0060] Furthermore, in the process of determining the target scenario template, this application can determine the target scenario template based on a single driving state information, or based on the vehicle's current driving state information and road information, or based on the vehicle's current driving state information and environmental information, or based on the vehicle's current road information and environmental information. It can also identify a scenario template that is suitable for the current driving environment based on the vehicle's current driving state information, road information, and environmental information, and determine it as the target scenario template.

[0061] Finally, in the process of generating and displaying recommendation information that matches the target scene template, the recommendation information can be generated by the field of view parameters corresponding to the target scene template, and the recommendation information can be presented in the form of a graphical user interface, such as a pop-up prompt box or floating card on the vehicle display screen.

[0062] The prompt box can include the name of the target scene template, a brief description, and an option for the user to accept or ignore it. Meanwhile, the generated recommendation information aims to inform the user of the system's suggested adaptive field-of-view configuration in a non-intrusive manner.

[0063] In the process of adjusting the field of view of the vehicle's electronic rearview mirror according to the recommended information, when the user responds to the recommended information, that is, after receiving the user's application instruction for the recommended information, such as confirming acceptance of the recommended scene template, the display of the electronic rearview mirror will be adjusted according to the preset field of view parameters in the target scene template.

[0064] The field of view adjustment of the electronic rearview mirror can include changing the cropping area of ​​the field of view, adjusting the brightness, contrast or saturation of the image, and enabling or disabling certain auxiliary functions, thereby switching the display effect of the electronic rearview mirror to a state that matches the target scene template, so as to provide an optimized driving field of view.

[0065] This application solves the technical problem of rigid field of view configuration of electronic rearview mirrors by comprehensively acquiring vehicle information from multiple dimensions to determine the target scenario template and introducing a user confirmation mechanism. As a result, users can obtain a field of view configuration that matches the current driving situation without frequent manual adjustment, thereby improving driving safety and user experience.

[0066] In some embodiments, the scene template library may further include custom scene templates, and the adjustment method of the electronic rearview mirror may further include: determining a first triggering condition and a first field of view adjustment parameter corresponding to the creation of the custom scene template based on the user's operation instructions; and generating the custom scene template according to the first triggering condition and the first field of view adjustment parameter.

[0067] In this embodiment, the scene template library can be a multi-layered architecture scene template library, such as... Figure 3 The three-tiered scene template library shown has a first tier for preset scene templates, a second tier for custom scene templates, and a third tier for intelligent scene templates.

[0068] Custom scene templates can be understood as view configuration schemes that users can create and save independently based on their personal needs and preferences. Custom scene templates allow users to set trigger conditions and adjust view parameters. Custom scene templates include scene templates for commuting mode, scene templates for picking up and dropping off children, and scene templates for long-distance travel mode.

[0069] In the custom scene templates, users can create scene templates according to their personal needs and preferences. This application provides a complete scene template creation workflow, allowing users to set scene trigger conditions, adjust field of view parameters, name the scene, select icons, and save the created templates to their personal template library.

[0070] Furthermore, user-defined templates can be infinitely expanded, allowing each user to create any number of personalized scenarios based on their driving habits. For example, a commuter mode scenario template can be created for daily commutes, a child pick-up and drop-off mode scenario template can be created for low-speed driving and frequent stopping around schools, and a long-distance travel mode scenario template can be created for long-distance driving on highways.

[0071] In addition, user-defined templates are bound to user identities, and templates for different users are isolated from each other and do not affect each other.

[0072] In the process of generating a custom scene template, this application can determine the first triggering condition corresponding to the creation of the custom scene template; based on the first triggering condition, obtain the first field of view parameters input by the user and the scene description information of the custom scene template; and generate the custom scene template according to the first triggering condition and the first field of view parameters, while adding scene description information to the custom scene template.

[0073] The first trigger condition can be based on a combination of sensor data and logical operators. For example... Figure 6 As shown, the first trigger condition allows users to define in what specific driving situations their custom field of view configuration should be recommended or activated by the system.

[0074] Specifically, users can select vehicle sensor data (such as GPS speed sensor, rain sensor, ambient light sensor, etc.) as input and set corresponding logical judgment conditions (such as speed greater than a certain threshold, rainfall intensity within a certain range, ambient light intensity below a certain level). They can further combine these conditions using logical operators (such as AND, OR, NOT) to construct complex and precise triggering rules to form the first triggering condition.

[0075] As an example, in the process of obtaining the user's input of the first field of view parameters and the scene description information of the custom scene template, the user's specific personalized settings for the electronic rearview mirror's field of view can be collected, and the user can name and describe the created template.

[0076] As another example, users can perform intuitive operations through a graphical user interface (GUI), such as adjusting the field of view in real time by dragging the cropping box, or adjusting image enhancement parameters through slider controls and previewing the adjustment effect in real time.

[0077] The scene description information allows users to enter an easily recognizable name (such as "My Commuting Mode" or "Rainy Night Driving Mode") and an optional detailed description for custom scene templates, enabling users to manage and understand the purpose of the template later. Additionally, users can select an icon or label to further visually identify the template.

[0078] The first field of view parameters may include, but are not limited to, field of view cropping parameters (such as adjusting the top, bottom, left and right boundaries of the field of view to change the display area), image enhancement parameters (such as brightness adjustment values, contrast adjustment values, saturation adjustment values ​​and other image display effect adjustment values), display mode parameters (such as selecting normal mode, wide-angle mode or split-screen mode), and auxiliary function switch parameters (such as whether to enable blind spot warning, lane line overlay or defogging algorithm).

[0079] During the process of generating a custom scene template, all user-defined elements can be integrated and stored as a custom scene template that the system can recognize and use.

[0080] Specifically, this application can encapsulate the user-defined first trigger condition (in the form of a logical expression or event identifier), the user-input first field of view parameters (various adjustment values ​​stored in a structured data format), and scene description information (name, description, icon, and other metadata) into a complete data structure, which is then stored in the scene template library as part of the custom scene template.

[0081] In addition, during the generation of custom scene templates, user input can be validated to ensure the validity and completeness of parameters. Furthermore, a summary page can be provided before saving, allowing users to confirm that all settings are correct before persistent storage and binding them to the user's identity information, ensuring the isolation and management of personalized configurations.

[0082] As an example, the process of creating a custom scene template can specifically include four steps: setting the first trigger condition, adjusting the first field of view parameters, inputting scene information, and saving and sharing.

[0083] Setting the first trigger condition: The user defines the first trigger condition for the scenario, which is to recommend the scenario under specific circumstances. The first trigger condition is formed based on a combination of sensor data and logical operators. Specifically, this application can provide a visual condition editing interface, where users can select sensor types (such as GPS speed, road type, rainfall, light intensity, etc.), set comparison operators (greater than, less than, equal to, greater than or equal to, less than or equal to) and threshold values, and use logical operators (AND, OR, NOT) to combine multiple conditions.

[0084] For example, a user can set the first trigger condition as: GPS speed greater than 80 AND road type equal to highway AND rainfall greater than 50, meaning that the scenario is triggered when driving at a speed of 80 km / h or higher on a highway and encountering moderate to heavy rain. This application represents the first trigger condition as a logical expression tree, stored in the template's metadata, facilitating subsequent scenario matching calculations.

[0085] First-view parameter adjustment: Users can adjust first-view parameters on the real-time preview interface, including cropping area, image enhancement parameters, display mode, and accessibility functions. This application also provides a touch-drag cropping frame, allowing users to directly drag the four boundaries of the cropping frame on the preview screen to observe changes in the field of view in real time. Image enhancement parameters are adjusted via a slider control; users can drag the slider to adjust brightness, contrast, and saturation in real time, with the preview screen updating synchronously. Display mode and accessibility functions are selected via checkboxes or toggle buttons. Furthermore, the delay in adjusting first-view parameters must be controlled within 50 milliseconds to ensure real-time synchronization between adjustment operations and visual feedback.

[0086] Scene information input: Users enter a name for the scene template, select an icon, and write a description. The scene template name supports Chinese, English, and numbers, with a maximum length of 20 characters. The icon can be selected from a preset icon library, including various scene-related emoticons and vector icons. Writing a description is optional; users can record information such as the applicable circumstances and precautions for the scene, facilitating later recall and management.

[0087] Saving and Sharing: After completing the previous steps, users can choose the save location (local storage, cloud synchronization, binding to seat memory, etc.) and sharing method (generating a share code, uploading to the community, etc.). If cloud synchronization is selected, the template data will be encrypted and uploaded to the cloud server, bound to the user's account. If a share code is generated, this application can generate a unique alphanumeric combination (e.g., AB3F-KL9M-XZ7Q), which other users can import the template by entering.

[0088] This application enables users to transform from passive configuration users to active scene creators by creating custom scene templates. This greatly enhances the personalization and adaptability of the electronic rearview mirror's field of vision adjustment. It not only solves the problem of users being unable to create personalized templates, but also enriches the content of the scene template library, allowing the electronic rearview mirror system to better adapt to the diverse needs of different users and complex and ever-changing driving environments, significantly improving the driving experience and safety.

[0089] For example, if a user creates a custom scene template for a child pick-up / drop-off mode, the operation process is as follows: The user sets the first trigger condition by specifying a time period between 15:00-16:00 or between 08:00-09:00 in the condition editing interface. This indicates that the scenario will be triggered during the daily drop-off and pick-up times for children. The user also adds auxiliary conditions: GPS speed less than 40 km / h AND road type equal to urban road, ensuring that the scenario template is only recommended when driving at low speeds on urban roads near schools.

[0090] Adjusting the second field-of-view parameters involves dragging the cropping box on the real-time preview interface, increasing the lower boundary from the default 60% to 75%, thus expanding the field of view along the bottom edge of the vehicle and the ground, making it easier to observe children and obstacles on the roadside. Simultaneously, the user adjusts the brightness to +12% and the contrast to +8% for a clearer image. Additionally, the user enables blind spot warning and low-speed pedestrian detection assistance. After each adjustment, the preview screen updates synchronously within 50 milliseconds, allowing the user to observe the effects of the adjustments in real time.

[0091] Input scenario information: The user enters the name of the scenario template as the "pick-up and drop-off mode", selects the corresponding icon, and fills in the description field with "for picking up and dropping off children around the school", "drive at low speed", "beware of children on the roadside", etc.

[0092] For saving and sharing, users can check "Save to local device" and "Sync to cloud," and also check "Bind to current seat memory." Users can also check "Generate sharing code," generating the sharing code KC8D-PM4N-RL2X. This sharing code can then be sent to other parents for importing and using the template.

[0093] Once saved, the generated scene template will immediately appear in the user-defined template category of the scene template list. At the same time, the data corresponding to the template can be uploaded to the cloud and linked to the user's account.

[0094] In some embodiments, the scene template library further includes intelligent scene templates, and the adjustment method of the electronic rearview mirror further includes: obtaining a preset number of recent adjustment records of the user, each adjustment record including scene information and field of view condition parameters; performing cluster analysis on the adjustment records to confirm the second triggering condition and the second field of view condition parameters; and generating an intelligent scene template based on the second triggering condition and the second field of view condition parameters.

[0095] In this application, by creating intelligent scene templates, the system can accurately match users' personalized needs, significantly reducing the tedious operation of manually creating or frequently adjusting field of vision parameters. It can also continuously adapt to changes in users' driving habits, making the electronic rearview mirror's field of vision adjustment more intelligent, personalized, and convenient, effectively improving the driving experience and safety. The preset quantity can be understood as the number of adjustments the user makes within the current period.

[0096] In this embodiment, the scene template library can be a multi-layered architecture scene template library, such as... Figure 3 The three-tiered scene template library shown has a first tier for preset scene templates, a second tier for custom scene templates, and a third tier for intelligent scene templates.

[0097] Intelligent scene templates can analyze users' usage habits and scene selection preferences through machine learning algorithms to automatically generate or suggest personalized view configuration schemes. Intelligent scene templates include a morning rush hour recommendation mode, a rainy highway safety mode, and AI-suggested scenarios.

[0098] Specifically, in the process of acquiring the user's most recent preset number of adjustment records and performing cluster analysis on the adjustment records to confirm the second trigger condition and the second field of view condition parameters, the second field of view adjustment parameters and scene information of the user adjusting the electronic rearview mirror each time within a preset first number of times can be acquired. Then, a clustering algorithm is used to process the second field of view adjustment parameters and scene information of the user adjusting the electronic rearview mirror each time, thereby determining the second number of times the user adjusts the electronic rearview mirror in the same scene in the same adjustment method. If the second number is greater than or equal to a preset number threshold, the scene information corresponding to the second number is used as the second trigger condition, and an intelligent scene template is generated based on the second trigger condition and the second field of view adjustment parameters corresponding to the second number.

[0099] In addition, during the process of generating an intelligent scene template based on the second triggering condition and the second field of view adjustment parameter corresponding to the second number, when the second number is greater than or equal to the number of times threshold, the present application can first generate suggestion information for creating an intelligent scene template; if the user responds to the return information of the suggestion information indicating that an intelligent scene template is to be created, then the intelligent scene template is generated based on the second triggering condition and the second field of view adjustment parameter corresponding to the second number.

[0100] This application utilizes AI learning and intelligent recommendation engines to acquire the user's most recent preset number of adjustment records. AI learning and intelligent recommendation engines are the core technologies for achieving personalization and intelligence in this solution. Figure 8 As shown, an AI learning module can be used to record users' scene switching history, recommendation feedback, manual adjustment records, and driving environment data in specific scenarios. Through algorithms such as collaborative filtering, reinforcement learning, or temporal pattern mining, the user's scene preference patterns and adjustment habits can be identified.

[0101] The scene switching history records the time, scene type, triggering method (intelligent recommendation or manual switching), and parameter changes before and after each scene switching.

[0102] The recommendation feedback information records the user's response to each recommendation (accept, reject, or no response), as well as the response time.

[0103] The manual adjustment record records the user's manual adjustment operations in various scenarios, including the type of adjustment parameter, adjustment range, and adjustment frequency.

[0104] Driving environment data records the GPS trajectory, time period, weather conditions, road type, and other environmental characteristics of each drive.

[0105] Based on the above data, spatiotemporal features, scene preference vectors, feedback pattern features, and adjustment habit features can be extracted for machine learning. Specifically, collaborative filtering algorithms, reinforcement learning algorithms, and time-series pattern mining can be implemented. The combination of collaborative filtering and reinforcement learning algorithms can achieve a two-stage recommendation decision and select the scene with the highest C value for recommendation output.

[0106] Temporal pattern mining can generate new template suggestions and provide recommendations. It can identify recurring manual adjustment patterns in specific scenarios. When it detects that a user performs the same or similar manual adjustment operations multiple times (e.g., 3 or more times) in a certain scenario, it can be determined that this adjustment pattern represents the user's preferred configuration, and the system can proactively suggest creating new intelligent scenario templates for the user.

[0107] The detection algorithm for temporal pattern mining can be as follows: Maintain a sliding window with a window size equal to the last 10 adjustment records. For each adjustment record within the window, extract its scene features (such as weather, road type, time period, etc.) and adjustment parameters (such as brightness adjustment amount, cropping area change, etc.).

[0108] Furthermore, this application can also identify cluster centers for adjusting parameters using clustering algorithms (such as K-means or DBSCAN). If the sample size of a certain cluster center exceeds a threshold (such as 30%), and the scene features corresponding to that cluster are consistent, it is determined to be a repeated adjustment mode.

[0109] When generating intelligent scene template suggestions, the second adjustment parameter of the cluster center can be extracted as the default configuration of the template, and the intersection of scene features can be extracted as the second trigger condition. At the same time, these are presented to the user as recommendation options, which the user can accept, modify, or reject. The system continuously optimizes the learning algorithm based on user feedback.

[0110] The recommendation card can explain to the user: Your adjustment preferences for rainy highway scenes have been detected (brightness +20%, contrast +15%, defogging algorithm enabled). Do you want to create a rainy highway safety mode? Users can click "Create Template" to save directly, or click "Remind Me Later" to postpone processing.

[0111] In addition, when performing scene matching, the intelligent scene template can use the same matching degree calculation method as the user-defined template: specifically, the identified scene features (such as weather, road type, time period, etc.) can be converted into logical expressions, and the matching degree score can be calculated using the same logical operation rules as the user-defined template.

[0112] In this application, the AI ​​learning module in the AI ​​learning and intelligent recommendation engine can learn not only the usage of preset scene templates, but also the usage of custom templates and intelligent scene templates. After a custom template is created, its triggering conditions and parameter configurations can be recorded. Simultaneously, the usage frequency, usage scenarios, and effects of the template can be tracked. Based on the usage of the template, the recommendation weights of relevant scenarios can be adjusted. Furthermore, the similarity between the template and other templates can be identified, generating similar template suggestions for other users. Thus, user-created custom templates can become input for AI learning to generate more optimized suggestions, forming a virtuous cycle from user creation to AI learning and then to intelligent optimization.

[0113] The AI ​​learning module can continuously optimize the recommendation algorithm based on user feedback. Each recommendation result (acceptance, rejection, or no response) is used as a training sample and adjusted to adjust the weight parameters of scene recognition and the formula for calculating recommendation confidence.

[0114] Furthermore, this application can also employ an online learning mechanism, thus eliminating the need for offline retraining and allowing for real-time updates of model parameters. The optimization objective is to maximize the recommendation acceptance rate and minimize the recommendation rejection rate; the objective function can be defined as:

[0115] in, Number of times accepted; Number of rejections; Total number of recommendations; The rejection penalty coefficient is usually set to 1.5, indicating that the negative impact of rejection outweighs the positive impact of acceptance, and the system is more inclined to make cautious recommendations to avoid disturbing the user.

[0116] It should be noted that there was no response ( The processing method can be to remove it from The objective function is calculated after deducting certain factors, meaning the system only optimizes recommendations with explicit responses (acceptance or rejection). Meanwhile, too many unresponsive recommendations will lower the user activity score. This can indirectly influence recommendation strategies.

[0117] In some embodiments, determining a target scene template from a scene template library based on at least one of driving state information, road information, and environmental information includes: determining the scene matching degree of each scene template in the scene template library based on at least one of driving state information, road information, and environmental information; and determining a target scene template from the scene template library based on the scene matching degree of each scene template.

[0118] In this embodiment, driving status information can be obtained from data collected by sensors such as GPS speed sensors and steering signal detectors; road information can be obtained from data collected by electronic maps, reversing radar sensors, and lane line detection cameras; and environmental information can be obtained from data collected by sensors such as rain sensors and ambient light sensors.

[0119] The GPS speed sensor can be provided by the vehicle's wheel speed sensors or a GNSS module. It can provide real-time vehicle speed, distinguishing between different speed levels such as stationary, low, medium, and high speeds. The GPS speed sensor can use BeiDou + GPS dual-mode positioning, achieving an accuracy within 5 meters.

[0120] Electronic map data can be obtained from pre-installed offline map databases or online map services. The electronic map data interface can provide the current road type (such as highway, urban road, rural road, mountain road, etc.) and road attributes (such as curve curvature, number of lanes, etc.). The electronic map can be integrated with the APIs of Amap or Baidu Maps.

[0121] Rain gauges can be optical or millimeter-wave radar sensors, providing rainfall intensity values ​​ranging from 0% to 100%. The sensitivity of a rain gauge is between 0% and 100%, with a response time of less than 1 second.

[0122] Ambient light sensors can be provided by photoresistors or CMOS image sensors. They can provide ambient illuminance to distinguish between different lighting conditions such as daytime, dusk, and nighttime. The measurement range of ambient light sensors is between 0 and 10000 lux.

[0123] Reversing radar sensors can be provided by ultrasonic radar or millimeter-wave radar. They can provide the minimum distance to obstacles around the vehicle and the reverse gear status. A reversing radar system can consist of four sensors, with a detection range between 0.3 and 2.5 meters.

[0124] The turn signal detector can obtain information from the vehicle's BCM (Body Control Controller). The turn signal detector can provide turn signal status and steering angle to determine whether the vehicle is turning or changing lanes.

[0125] Lane detection cameras can combine a forward-facing camera with image recognition algorithms to provide lane curvature information, used to determine the degree of road curvature. These cameras can be forward-facing and support lane departure warnings.

[0126] Specifically, in determining the scene matching degree of each scene template in the scene template library based on at least one of the following information: driving status information, road information, and environmental information, this application can determine the scene matching degree of each scene template based on the dimensional data corresponding to each scene template. This allows for the quantification of the similarity between the current driving environment and various scene templates. Consequently, the target scene template can be determined from the scene template library based on the scene matching degree of each scene template, avoiding the excessive influence of a single factor. This improves the accuracy and rationality of the matching, enabling the electronic rearview mirror's field of vision adjustment to adapt to complex and ever-changing driving scenarios more intelligently and accurately, thereby enhancing driving safety and comfort.

[0127] In some embodiments, determining the scene matching degree of each scene template in the scene template library based on at least one of driving status information, road information, and environmental information includes: for each scene template, if the scene template is a preset scene template, the scene template matching degree calculation formula is: ; Let be the matching degree of the i-th preset scene template. Let i be the weight of the i-th preset scene template in the j-th dimension. The dimensional data of the i-th preset scene template in the j-th dimension; if the scene template is a custom scene template or an intelligent scene template, determine the multiple dimensional data of the scene template, and calculate the corresponding scene matching degree based on the multiple dimensional data using a logical expression tree method; each dimensional data is one of driving status information, road information, and environmental information.

[0128] In this embodiment, the scene template library can be a multi-layered architecture scene template library, such as... Figure 3 The three-tiered scene template library shown has a first tier for preset scene templates, a second tier for custom scene templates, and a third tier for intelligent scene templates.

[0129] Specifically, in determining the scene matching degree corresponding to the preset scene template, the custom scene template and the intelligent scene template, the preset scene template can be calculated using a multi-dimensional weighted summation method, while the custom scene template and the intelligent scene template can be calculated using a logical expression tree method.

[0130] The formula for multi-dimensional weighted summation can be:

[0131] in, Let be the matching degree of the i-th preset scene template. Let i be the weight of the i-th preset scene template in the j-th dimension. This represents the dimensional data of the i-th preset scene template in the j-th dimension. Meanwhile, It can be adopted Furthermore, the formula for multi-dimensional weighted summation can also be expressed as:

[0132] in, This can be understood as the i-th preset scene template. The matching score between the i-th preset scene template and the current driving environment is called the scene matching score. The number of evaluation dimensions; For the i-th preset scene template, the first... The weight coefficients of each dimension satisfy... ; For the i-th preset scene template In the The scoring function is given by the requirements of the i-th preset scene template and sensor data. The output is the score, which ranges from 0 to 1.

[0133] It should be noted that the use ( (For scene template index) or , The matching degree is represented by "etc." (using text subscripts to indicate specific scenario types), and the two labeling methods are equivalent. Furthermore, the recommendation confidence formula consistently uses... ( (For scene type index).

[0134] like Figure 4 As shown, taking the high-speed driving scenario as an example, the matching degree calculation includes the following four dimensions, and the specific calculation method is as follows: High-speed driving scenarios require vehicle speeds greater than 80 km / h, and the speed matching score function is defined as follows:

[0135] in, Given the current vehicle speed, the speed matching score function scores 0 when the speed is below 60 km / h, increases linearly between 60 and 80 km / h, and scores 1 when the speed exceeds 80 km / h.

[0136] The road type matching function is defined as follows:

[0137] in, Given the current road type, the road type matching function returns the corresponding matching score based on the road type.

[0138] For high-speed driving scenarios, recommendations are prioritized based on clear weather conditions. The weather matching score function is as follows:

[0139] in, For rainfall intensity, the weather matching score function can ensure a score of 1 when there is no rain and a score of 0.5 when there is heavy rain, thus retaining a certain probability of recommendation.

[0140] In addition, the high-speed driving scenario is not applicable to the stationary state, so the stationary state exclusion degree also needs to be calculated:

[0141] Based on the above dimensions, the formula for calculating the matching score for high-speed driving scenarios can be:

[0142] Furthermore, after calculating the matching score for all preset scene templates, this application selects the preset scene template with the highest score as the recommended candidate.

[0143] For example, assume the vehicle's current GPS speed km / h, road type is highway, rainfall intensity Reverse gear Minimum obstacle distance rice.

[0144] The scene matching degree calculation method for high-speed driving scenarios can be as follows: Speed ​​matching: (Because 95 > 80) Road type matching degree: (The current road is a highway) Weather compatibility:

[0145] Parking status exclusion rate:

[0146] The scene matching degree for high-speed driving scenarios can be: At the same time, due to Therefore, high-speed driving scenario templates can be included in the sequence of scenario templates to be recommended.

[0147] In determining custom and intelligent scene templates, the scene matching degree can be calculated using a logical expression tree approach based on multi-dimensional data. Specifically, the scene matching degree can be obtained by calculating from the bottom up using the logical expression tree. During the calculation of scene matching degree using the logical tree, the degree of fulfillment for each basic condition can be calculated based on the current sensor data; for logical operators, scores for each basic condition can be set, and the final score can be determined according to the corresponding calculation formula.

[0148] For example, the basic condition is velocity v, if The score is ;like The score is (Linear transition); if The score is .

[0149] If the basic condition is road type, the score is 1 if the road type is highway, 0.5 if the road type is urban expressway, and 0 if the road type is urban road.

[0150] At the same time, the scores for each basic condition can be... The formula for the AND operation is: For the OR operation, the formula can be: For the NOT operation, the formula can be: .

[0151] For example, if the logical expression tree of the custom scene template and the smart scene template is: GPS speed AND road type, when the GPS speed is greater than 80 and the road type is highway, the score corresponding to the GPS speed greater than 80 is 1, and the score corresponding to the road type is urban expressway is 0.5. It can be calculated using the AND operation formula, and then the matching degree of the custom scene template and the smart scene template can be obtained as 0.5.

[0152] In some embodiments, after receiving the user's application instruction for the recommendation information, the method further includes: generating a reward signal for the target scene template; and dynamically adjusting the weight of the preset scene template in the j-th dimension based on the reward signal.

[0153] Specifically, after receiving the user's application instruction for the recommended information, this application can generate a reward signal for the target scene template and dynamically adjust the weight of the preset scene template in the j-th dimension through the reward signal. Thus, when the preset scene template is accepted for recommendation, the weight of the dimension with higher weight in the process of calculating the scene matching degree of the preset scene template can be further increased, so that the vehicle can more easily identify the preset scene template in similar scenarios in the future. Alternatively, when the preset scene template is rejected for recommendation, the weight of the dimension with higher weight in the process of calculating the scene matching degree of the preset scene template can be further reduced to suppress misidentification.

[0154] In this embodiment, the formula for dynamically adjusting the weight of the preset scene template in the j-th dimension can be:

[0155] in, Let be the weight of the j-th dimension after the t-th update, and its value can be between 0.05 and 0.5. R is the learning rate, which can range from 0.01 to 0.05; R is the reward signal, which can be +1, -1.5, or 0. This is the score value of the j-th dimension in this recommendation, and its value can be between 0 and 1.

[0156] In some embodiments, determining the target scene template from the scene template library based on the scene matching degree of each scene template includes: determining the target scene template from scene templates whose scene matching degree reaches a preset matching degree threshold.

[0157] In this embodiment, after determining the scene matching degree of each scene template in the scene template library based on at least one of the driving status information, road information, and environmental information, the scene matching degree of each scene template can be compared with a preset matching degree threshold. Multiple scene templates with scene matching degrees higher than the preset matching degree threshold are then selected, and one or more seat-specific scene templates from the selected multiple scene templates are selected as the target scene template. The matching degree threshold can be a fixed value, such as 0.7.

[0158] Furthermore, in some embodiments, the electronic rearview mirror field of view adjustment method further includes: determining a first rejection rate of the user rejecting the vehicle recommended scene template in the current period; and determining a preset matching degree threshold based on the preset target rejection rate and the first rejection rate.

[0159] In this embodiment, the preset matching threshold can be dynamically adjusted based on the user's historical behavior. Specifically, the first rejection rate of the user rejecting the vehicle recommendation scenario template in the current period can be obtained in real time, and the first rejection rate can be compared with the system's preset target rejection rate. The preset matching threshold can be dynamically adjusted based on the degree of deviation between the two to achieve more accurate personalized services.

[0160] The formula for dynamically adjusting the preset matching threshold can be:

[0161] in, For the first The updated threshold can range from 0.5 to 0.9, and its initial value can be 0.7. The threshold adjustment step size can be 0.05; This represents the number of rejections in the current period. This represents the total number of recommendations for the current period. The target rejection rate can be set to 0.15.

[0162] In this application, if the first rejection rate of a user rejecting a vehicle recommendation scenario template in the current period is higher than the target rejection rate, the preset matching threshold is automatically increased to make the recommendation strategy more conservative, recommending only scenario templates with extremely high matching degree, thereby reducing the rejection rate.

[0163] Additionally, it should be noted that the number of unresponsive requests is not counted in the number of rejections. This application can only optimize and evaluate recommendations that have a clear response (acceptance or rejection). An excessive number of unresponsive requests can reduce user activity, thereby indirectly affecting the recommendation strategy.

[0164] In some embodiments, determining a target scene template from scene templates whose scene matching degree reaches a preset matching degree threshold includes: determining scene templates whose scene matching degree reaches a preset matching degree threshold as scene templates to be recommended; determining the recommendation confidence of each scene template to be recommended, and determining the target scene template from multiple scene templates to be recommended based on the recommendation confidence.

[0165] Specifically, to avoid making incorrect recommendations when the scenario is unclear, this application, in the process of determining the target scenario template from scenario templates with a scenario matching degree reaching a preset matching degree threshold, can filter out scenario templates with a scenario matching degree greater than or equal to the preset matching degree threshold from the scenario template library and use them as scenario templates to be recommended. Then, the recommendation confidence of each scenario template to be recommended is determined, and finally, the templates are sorted according to the recommendation confidence of the highest recommendation confidence. This not only improves the accuracy of the recommendation, but also fully adapts to the user's preferences and the current environment.

[0166] Furthermore, in some embodiments, determining the recommendation confidence of each scenario template to be recommended includes: determining the historical acceptance rate of each scenario template to be recommended, the spatiotemporal similarity between the spatiotemporal environment in which the vehicle is located when the user uses each scenario template to be recommended and the spatiotemporal environment in which the vehicle is currently located, and the user activity in response to the vehicle recommending each scenario template to be recommended; and generating the recommendation confidence of each scenario template to be recommended based on the scenario matching degree, the first weight corresponding to the scenario matching degree, the historical acceptance rate, the second weight corresponding to the historical acceptance rate, the spatiotemporal similarity, the third weight corresponding to the spatiotemporal similarity, the user activity, and the fourth weight corresponding to the user activity.

[0167] In this embodiment, during the process of determining the recommendation confidence of each scenario template to be recommended, the confidence of each scenario template can be processed based on the scenario matching degree, historical acceptance rate, spatiotemporal similarity between the spatiotemporal environment in which the vehicle is located when the user uses each scenario template and the spatiotemporal environment in which the vehicle is currently located, and the user activity in response to the vehicle recommending each scenario template. This results in the recommendation confidence of each scenario template, thereby providing a more accurate vision adjustment solution that better meets the user's personalized needs and real-time context. This effectively solves the problems of inaccurate recommendations, poor adaptability, and low user acceptance that may result from relying solely on sensor data matching, thereby improving the intelligence level of electronic rearview mirror vision adjustment and user experience.

[0168] Specifically, the recommended confidence level The calculation formula can be:

[0169] in, For the scene Scene matching degree (obtained by multi-sensor fusion calculation); The historical acceptance rate for this scenario is calculated as follows: , To accept the number of times, Number of rejections; Spatiotemporal similarity measures the degree of similarity between the current spatiotemporal environment and the historical spatiotemporal environment in which the scene was used. User activity is measured by the frequency of user responses to the recommendation system; the confidence level of recommendations for users with high activity levels is appropriately increased.

[0170] In some embodiments, after receiving the user's application instruction for the recommendation information, the method further includes: determining the number of times the user has rejected the recommendation and the number of times the user has accepted the recommendation within the current period; and adjusting the first weight, the second weight, the third weight, and the fourth weight based on the number of rejections and the number of acceptances to obtain the adjusted first weight, the second weight, the third weight, and the fourth weight.

[0171] Specifically, after receiving the user's application instruction for the recommendation information, this application can also adjust the first weight, second weight, third weight, and fourth weight based on the number of rejections and second acceptances by the user in the current period, so that the adjusted first weight, second weight, third weight, and fourth weight can more accurately support the generation of subsequent recommendation information and better match the user's true preferences.

[0172] In this embodiment, during the process of adjusting the first weight based on the number of rejections and the number of acceptances by the user in the current period, the first weight can be updated using the average value of the scene matching degree within the number of rejections, the average value of the scene matching degree within the number of acceptances, and the first weight corresponding to the current scene matching degree.

[0173] In the process of adjusting the second weight based on the number of rejections and the number of acceptances by the user in the current period, the second weight can be updated by the average of the historical acceptance rate within the number of rejections, the average of the historical acceptance rate within the number of acceptances, and the second weight corresponding to the current historical acceptance rate.

[0174] In the process of adjusting the third weight based on the number of rejections and the number of acceptances by the user in the current period, the third weight can be updated by the average value of spatiotemporal similarity within the number of rejections, the average value of spatiotemporal similarity within the number of acceptances, and the third weight corresponding to the current spatiotemporal similarity.

[0175] In the process of adjusting the fourth weight based on the number of rejections and the number of acceptances by the user in the current period, the fourth weight can be updated by the average user activity within the number of rejections, the average user activity within the number of acceptances, and the fourth weight corresponding to the current user activity.

[0176] The formulas for adjusting the first, second, third, and fourth weights can be:

[0177] in, Let be the weight of the i-th factor after the t-th update. The factor can be any one of scene matching degree, historical acceptance rate, spatiotemporal similarity and user activity. The step size for adjusting the coefficients can be 0.02; This represents the average value of the i-th factor in the second number of acceptances; It represents the average value of the i-th factor among the number of rejections.

[0178] In some embodiments, the electronic rearview mirror field of view adjustment method further includes: determining the temporal similarity between the time when the vehicle is in each recommended scene template and the current time of the vehicle, and the spatial similarity between the position of the vehicle when the user uses each recommended scene template and the current position of the vehicle; and weighted summing the temporal similarity and spatial similarity to obtain the spatiotemporal similarity.

[0179] In this application, spatiotemporal similarity is determined by weighted summation of temporal and spatial similarity. This allows for flexible adjustment of the relative importance of time and space factors based on actual application scenarios and the driver's personalized preferences, thereby generating a more reasonable spatiotemporal similarity value that better matches the driver's current situation. This enables the intelligent recommendation system of the electronic rearview mirror to more accurately identify the driver's true intentions and needs, avoiding the bias that may be caused by single-dimensional evaluation. As a result, the accuracy and personalization level of scenario template recommendation are improved, ultimately optimizing the driver's field of vision experience.

[0180] Specifically, time similarity can quantify the degree of matching between the time when a user used a certain recommended scenario template in the past and the current time of the vehicle, in order to capture the user's driving preference patterns in different time periods.

[0181] For example, matching can be performed based on time periods of the day (such as morning rush hour, daytime, evening rush hour, and nighttime). If the historical usage time and the current time fall within the same or similar time period, the time similarity is high.

[0182] Meanwhile, time similarity can also be quantified by calculating the absolute time difference between historical usage time and current time, and using a decay function (such as exponential decay or Gaussian decay). The smaller the time difference, the higher the similarity.

[0183] In addition, time similarity can also take into account periodic factors such as day of the week, weekdays / weekends, and holidays, to correct for time differences or introduce additional weights, so as to more accurately reflect time preferences.

[0184] Spatial similarity can quantify the degree of matching between the vehicle's location when a user used a recommended scenario template in the past and the vehicle's current location, reflecting the impact of geographical location on user driving behavior and field of vision preferences.

[0185] For example, matching can be performed based on geographic region type (such as highway, urban road, rural road, residential area, commercial area, parking lot, etc.). If the historical location and the current location fall into the same or similar geographic region type, the spatial similarity is high.

[0186] Alternatively, spatial similarity can be quantified by calculating the geographical distance between historical usage locations and the current location, and using inverse distance weighting or a Gaussian decay function; the closer the distance, the higher the similarity. When calculating spatial similarity, factors such as road type and traffic conditions can also be incorporated for correction to improve the accuracy of the assessment.

[0187] In this embodiment, spatiotemporal similarity can be achieved by comparing the current spatiotemporal environment with the user's historical usage scenarios. The calculation is based on the spatiotemporal environment. This application can maintain historical usage records for each scenario, including temporal characteristics (hour, day of the week) and spatial characteristics (road type, region) of usage. Among these, spatiotemporal similarity... The calculation formula can be:

[0188] in, This refers to time similarity (the degree of matching between the current time and historical usage times). Spatial similarity (the degree of matching between the current location and historical usage locations). (usually taken) , When calculating spatial similarity, the system will refer to the scene preference vector. User preferences Higher similarity scores are given for scenes with higher similarity scores.

[0189] In some embodiments, determining a target scenario template from multiple scenario templates to be recommended based on recommendation confidence includes: if the confidence difference between the recommendation confidence of each scenario template to be recommended meets a preset condition, determining the user's preference score for each scenario template to be recommended; and determining the target scenario template from multiple scenario templates to be recommended based on the preference score.

[0190] Specifically, since the recommendation confidence levels of multiple scenario templates to be recommended are close to each other, it may be difficult to effectively distinguish them using a single confidence level indicator. Therefore, this application determines whether the confidence level difference between the recommendation confidence levels of each scenario template to be recommended meets a preset condition, thereby determining the user's preference score for each scenario template to be recommended. Based on the preference score, the target scenario template is determined from multiple scenario templates to be recommended. This effectively solves the problem of decision ambiguity and unsatisfactory recommendation results that may result from relying solely on confidence level, significantly improves the personalized accuracy of electronic rearview mirror field of view adjustment, avoids inconvenience to the driver caused by system misjudgment, and thus improves user experience and driving safety.

[0191] The preset condition can be a specific numerical threshold. For example, the condition is considered met when the absolute difference between the highest recommendation confidence and the second highest recommendation confidence is less than 0.05, or when the recommendation confidence of the top three recommended scenario templates all fall within a certain preset small range (such as 0.75 to 0.85).

[0192] In addition, the preset conditions can also be dynamically adjusted. For example, the threshold can be adaptively adjusted through a machine learning model based on historical recommendation data and user feedback to more accurately determine whether the confidence levels are close under different driving scenarios and user habits, thereby triggering subsequent preference score calculations.

[0193] Preference scores quantify a user's level of liking for a specific template to be recommended. Preference scores can be calculated based on various user behavior data.

[0194] For example, this application can count the number of times a user manually selects and applies the template in history, the number of times they accept system recommendations and apply the template, and the duration of time a user uses the template, and then perform a weighted average or a comprehensive evaluation through a machine learning model to obtain a value that reflects the degree of user preference.

[0195] Alternatively, this application can calculate a preference score based on explicit user feedback to the template (such as likes, favorites, ratings) or implicit feedback (such as frequency of use, duration of use, degree of fine-tuning of parameters). For example, the more times a user accepts a particular template, or the more frequently they manually select that template, the higher their preference score will be.

[0196] For example, suppose we detect that a user is driving at high speed and need to decide whether to recommend a high-speed driving scenario template. We can first filter by scenario matching score, where the scenario matching score of the high-speed driving scenario template is... ,because If the score is greater than 0.7, the high-speed driving scenario template passes the scene matching degree filter and enters the recommendation confidence degree ranking.

[0197] Meanwhile, assuming other scene templates also passed the scene matching filter, such as the rainy / foggy weather scene template with a low scene matching score... .

[0198] Furthermore, during the recommendation confidence ranking process, the recommendation confidence of the high-speed driving scenario template and the rain and fog weather scenario template can be calculated. .

[0199] If the scene matching degree of the high-speed driving scene template Historical acceptance rate (Recommended to accept 12 times, reject 2 times), Spatiotemporal similarity User activity The recommendation confidence of the high-speed driving scenario template .

[0200] Meanwhile, assuming the recommendation confidence of the rainy / foggy weather scenario template. .

[0201] because The recommended high-speed driving scenario template will then be selected as the target scenario template. Additionally, when displaying the recommendation card, confidence information, such as a recommendation accuracy of 91.6%, can be attached to enhance user trust.

[0202] Furthermore, in some embodiments, the preference score is calculated as follows: determining the number of times each scene template to be recommended is used, the first number of times each scene template is accepted, and the fifth weight corresponding to the first number of acceptances; and generating a user's preference score for each scene template to be recommended based on the number of uses, the first number of acceptances, and the fifth weight corresponding to the first number of acceptances.

[0203] Specifically, usage frequency can be understood as the actual frequency with which a user applies a recommended scenario template. Usage frequency reflects the user's actual need for and acceptance of the template. This application can achieve this by recording each instance of a user applying a recommended scenario template and accumulating the usage count of the corresponding template in the database.

[0204] For example, whenever a user activates a scene template through intelligent recommendation or manual selection, the system increments its usage count by one.

[0205] In addition, this application can also maintain a detailed scene template activation log, which includes the timestamp and template identifier for each activation. When calculation is needed, the number of uses can be obtained by querying and statistically analyzing the activation records within a specific time period (such as the most recent week, month, or total).

[0206] The first acceptance count can be understood as the number of times a user actively accepts a specific scenario template recommended by the system. Compared to the number of uses, the act of actively accepting recommendations better reflects the user's trust in the system's recommendations and their clear preference for that template. This application can increment the first acceptance count for each scenario template recommended to the user and after receiving a clear application instruction from the user.

[0207] Alternatively, this application can also achieve this by analyzing recommendation feedback logs, which can record each recommended scenario template and its corresponding user response (acceptance, rejection, or no response), thereby filtering and counting the number of times the user explicitly accepted the recommendation.

[0208] The fifth weight is used to adjust the importance of the first acceptance count in the preference score calculation, highlighting the value of the user's proactive acceptance of the recommendation. This application can make proactively accepted preference signals occupy a more significant position in the final preference evaluation by assigning weight to the first acceptance count.

[0209] The fifth weight can be a preset fixed value, for example, it can be set to 1.5 or 2.0, so that the preference intensity of one active acceptance is equivalent to 1.5 or 2 ordinary uses.

[0210] In addition, the fifth weight can also be dynamically adjusted based on factors such as users' historical behavior data, the overall accuracy of the recommendation system, or user activity, in order to more flexibly adapt to different user preference patterns.

[0211] In this embodiment, the formula for calculating the preference score can be:

[0212] Where k is the total number of all scenario templates to be recommended; The user's preference score for the k-th category of recommended scenario templates. The number of times the k-th category of the recommended scenario template has been used; This represents the first acceptance count for the k-th scenario. As the fifth weight, Let i be the number of times the template for the i-th type of scenario to be recommended is used. This represents the first acceptance count for the i-th scenario.

[0213] For example, suppose a user's usage data is as follows: The highway driving scenario template was used 40 times, with 12 recommendations accepted; the city parking scenario template was used 25 times, with 8 recommendations accepted; the mountain road curve scenario template was used 5 times, with 2 recommendations accepted; the rainy / foggy weather scenario template was used 15 times, with 6 recommendations accepted; and the nighttime driving scenario template was used 15 times, with 5 recommendations accepted. k=5. The preference scores for each scenario template are shown below.

[0214] Preference score for high-speed driving scenario template for:

[0215] Urban parking scenario preference score for:

[0216] Mountain road curve scene preference score for:

[0217] Rainy / foggy weather scene preference score for:

[0218] Nighttime driving scenario preference score for:

[0219] Furthermore, it can be seen that Mr. Zhang prefers high-speed driving scenarios (38.8%), followed by urban parking scenarios (24.8%).

[0220] In some embodiments, after receiving an application instruction from a user regarding recommended information, the electronic rearview mirror of the vehicle is adjusted for field of view based on the configuration parameters corresponding to the target scene template. This includes: if the application instruction is a first instruction, obtaining the configuration parameters corresponding to the target scene template and adjusting the field of view of the electronic rearview mirror according to the configuration parameters corresponding to the target scene template; if the application instruction is a second instruction, reducing the recommendation weight of the target scene template; and if the application instruction is a third instruction, adjusting the user activity level of the target scene template.

[0221] In this application, the first instruction can be understood as a feedback signal indicating the user's acceptance or agreement to the recommended scenario template. The first instruction can be input by the user in various ways, such as by clicking the "Apply" button on the in-vehicle display, issuing a specific voice command (such as "Accept Recommendation"), or performing a preset gesture operation. Simultaneously, upon receiving the first instruction, it is confirmed that the user wishes to adopt the currently recommended field-of-view configuration.

[0222] The configuration parameters corresponding to the target scene template can be understood as a set of specific configuration data associated with the accepted target scene template, used to define the display state of the electronic rearview mirror. The configuration parameters corresponding to the target scene template may include, but are not limited to, the boundary values ​​of the field of view cropping area (such as pixels or percentages in the top, bottom, left, and right), image enhancement parameters (such as brightness, contrast, and saturation adjustment values), display modes (such as normal mode, wide-angle mode, and split-screen mode), and the enabled status of auxiliary functions (such as blind spot warning, lane line overlay, and defogging algorithm).

[0223] In the process of adjusting the field of view of the electronic rearview mirror, this application can adjust the display screen of the electronic rearview mirror in real time according to the configuration parameters corresponding to the target scene template. Specifically, it can adopt a direct switching method, that is, immediately apply all new parameters; or adopt a smooth transition method, that is, gradually and progressively adjust various parameters within a certain time window, so as to avoid sudden changes in the field of view causing discomfort or safety hazards to users.

[0224] The second instruction can be understood as a feedback signal indicating that the user rejects or does not accept the recommended scenario template. The second instruction can be input by the user in various ways, such as clicking the "ignore" button on the in-vehicle display, issuing a specific voice command (such as "reject recommendation"), or performing a preset gesture. Upon receiving the second instruction, it is confirmed that the user does not wish to adopt the currently recommended field-of-view configuration.

[0225] In the process of reducing the recommendation weight of the target scenario template, this application can adjust the priority of the target scenario template in future recommendation decisions based on user feedback of rejection. This can be achieved in various ways, such as reducing the weight coefficient of the template in the recommendation algorithm, or updating the historical acceptance rate of the template (reducing the number of acceptances and increasing the number of rejections), thereby reducing the likelihood of the template being recommended again in similar driving situations.

[0226] The third instruction can be understood as a feedback signal that the user does not give any clear response (neither accepting nor rejecting) within a preset time after the system sends out recommendation information. It can be manifested as the recommendation information disappearing automatically after being displayed for a period of time, without the user taking any action.

[0227] In adjusting the user activity score of a target scenario template, the activity score of the template or the user can be modified based on the user's lack of response to recommendation information. For example, the user's activity score in the recommendation system can be reduced, or the activity factor in the recommendation confidence calculation of the template in a specific user context can be reduced to reflect the user's low response frequency to the recommendation system, thereby affecting subsequent recommendation strategies.

[0228] On the in-vehicle terminal provided in this application, the displayed recommendation card includes two buttons: "Apply" and "Ignore." Clicking the "Apply" button immediately adjusts the electronic rearview mirror's field of view to the configuration parameters corresponding to the target scene template, and records acceptance feedback in the system log. This feedback data is used for subsequent AI learning to enhance the recommendation weight of the scene template in similar situations. Clicking the "Ignore" button cancels the recommendation, keeping the current field of view configuration unchanged, and records rejection feedback in the system log. This feedback data is used to reduce the recommendation weight of the scene template in similar situations, avoiding repeated recommendations of inappropriate scene templates. After no operation, i.e., if the user does not perform any operation on the recommendation card within a certain period of time (usually 5 seconds), it is automatically determined to be ignored, the recommendation card automatically disappears, and no scene switching is performed. This behavior is recorded as "unresponsive" in the system log, and the number of unresponsive events will affect the user's activity score.

[0229] In this application, as Figure 5 As shown, this application allows for the setting of an intelligent recommendation mode and a one-click switching mechanism, i.e., a dual-mode switching mechanism, to meet the interaction needs in different scenarios. Specifically, the intelligent recommendation mode dynamically optimizes the electronic rearview mirror's field of view adjustment by specifically processing different types of user responses to recommended information, thus solving the problem of insufficient user feedback processing.

[0230] Specifically, in the intelligent recommendation mode, upon receiving the first instruction, it indicates that the user accepts the recommendation. The system then retrieves the configuration parameters corresponding to the target scenario template and adjusts the rearview mirror accordingly to ensure the correct application of the field of view configuration, preventing the user from neglecting actual parameter adjustments simply because they accept the recommendation. Upon receiving the second instruction, it indicates that the user rejects the recommendation. The system can then reduce the recommendation weight of the target scenario template based on this response, preventing the repeated recommendation of inappropriate templates in similar situations. Upon receiving the third instruction, it indicates that the user ignores the recommendation. The system adjusts the user activity level of the target scenario template based on this response, reflecting the user's response habits, thereby optimizing future recommendation strategies. This allows the system to adaptively adjust the recommendation mechanism based on user feedback, improving the accuracy and personalization of recommendations, and significantly enhancing the user experience and driving safety.

[0231] During the one-click manual switching process, users can actively open the scene template library interface, browse all available scene templates (including preset scene templates, user-defined and smart scene templates), and click on any scene template to switch instantly.

[0232] The one-click manual switching process is as follows: the user clicks the scene template icon on the central control screen or the shortcut entry on the electronic rearview mirror display, and the system pops up the scene template list interface. The list interface displays all available scene templates in the form of icons and names, arranged according to usage frequency or a custom order.

[0233] After the user clicks on the target scene template, the system immediately reads the configuration parameters of that scene template and initiates the view switching process. The switching process uses a gradual parameter transition (the transition time is usually 300 to 1000 milliseconds, dynamically adjusted according to the parameter change range). Parameters such as cropping area, brightness, and contrast change smoothly within the transition time window, and a transition animation effect is superimposed on the displayed screen to indicate to the user that the view is switching.

[0234] Furthermore, this application also enables dual-mode collaborative operation, namely, intelligent recommendation mode and one-click switching can work together. For example, in a high-speed driving scenario, the system intelligently recommends a high-speed driving scenario template. After the user clicks the application, if they subsequently enter a tunnel, they can manually switch to the tunnel scenario template to obtain a more suitable field of view configuration for the tunnel environment. The switching records of both the intelligent recommendation mode and the one-click switching method are recorded by the AI ​​learning module and used to optimize future scene recognition and recommendation strategies.

[0235] For example, after identifying the target scene template as a high-speed driving scene template, this application can pop up a recommendation card in the upper right corner of the central control display screen. The card size can be 300×150 pixels, and the background can be semi-transparent black (60% transparency).

[0236] The card content may include: a scene icon (car) and the scene name of the high-speed driving scene displayed at the top, the reason for recommending that a high-speed scene (vehicle speed 95km / h, highway) was detected displayed in the middle of the screen, and a blue button for the application and a gray button for ignoring it displayed at the bottom.

[0237] If the driver clicks the application's blue button within 2 seconds of seeing the recommended card, the following actions will be performed immediately: The configuration parameters of the high-speed driving scenario template will be read; a smooth transition mechanism will be activated, setting the transition time (e.g., 500 milliseconds) based on the parameter change magnitude; the difference between the current and target parameters will be calculated, for example, brightness needs to be adjusted from 0% to +5%, and contrast needs to be adjusted from 0% to +10%; the parameters will be updated frame by frame at a rate of 30 frames per second, completing the transition in 15 frames (500ms × 30fps / 1000 = 15 frames); a prompt bar will be displayed at the top of the screen, indicating that the system is switching to a high-speed driving scenario; after the transition is complete, the prompt bar will automatically disappear, and the field of view configuration will be updated to the high-speed driving scenario; the timestamp, scenario type (high-speed driving), trigger method (intelligent recommendation), driver response (acceptance), and response time (2 seconds) will be recorded in the log.

[0238] If the driver clicks the gray "ignore" button, the scene switch is canceled, the recommended card disappears immediately, and the field of view configuration remains unchanged. The log records the timestamp, scene type (high-speed driving), trigger method (intelligent recommendation), and driver response (rejection). Simultaneously, the AI ​​learning module incorporates this rejection feedback into the training data, reducing the weight of recommending high-speed scenes in the current context.

[0239] If the driver does not perform any action on the recommendation card within 5 seconds, it will be automatically considered ignored, and the recommendation card will fade out; the driver's response (no response) will be recorded in the log.

[0240] In some embodiments, adjusting the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template includes: obtaining the current field of view parameters of the electronic rearview mirror, and determining the frame parameter increment for adjusting the field of view of the electronic rearview mirror based on the current field of view parameters and the configuration parameters corresponding to the target scene template; and adjusting the field of view of the electronic rearview mirror based on the frame parameter increment and the configuration parameters corresponding to the target scene template.

[0241] Specifically, in the process of obtaining the current field of view parameters of the electronic rearview mirror, this application can obtain the field of view configuration information currently being used by the electronic rearview mirror. This can be obtained either by reading the currently active field of view configuration data stored inside the electronic rearview mirror control unit (ECU), or by querying the currently effective field of view parameters in real time from the electronic rearview mirror display module via the vehicle bus (such as the CAN bus).

[0242] Current field of view parameters may include field of view cropping parameters (such as the top, bottom, left and right boundary positions), image enhancement parameters (such as brightness adjustment values, contrast adjustment values, saturation adjustment values), display mode parameters (such as normal mode, wide-angle mode, split-screen mode), and auxiliary function switch parameters (such as blind spot warning, lane line overlay, defogging algorithm, etc.).

[0243] In determining the frame parameter increment for adjusting the field of view of the electronic rearview mirror based on the current field of view parameters and the configuration parameters corresponding to the target scene template, the amount of parameters that need to be adjusted for each frame within a preset transition time can be calculated, given the current field of view configuration of the electronic rearview mirror and the configuration parameters corresponding to the target scene template. Here, the frame parameter increment can be understood as the parameter increment for each frame.

[0244] Specifically, this application calculates the difference between the current field of view parameters and the configuration parameters corresponding to the target scene template, and then divides the difference by the preset transition time (e.g., 300 milliseconds to 1000 milliseconds) and the display frame rate to obtain the parameter increment for each frame.

[0245] Alternatively, this application can employ an interpolation algorithm (such as linear interpolation or curve interpolation) to calculate the parameter values ​​to be adjusted for each frame based on the current parameters, target parameters, and preset number of transition frames, thereby obtaining the frame parameter increment.

[0246] Assuming the current parameter value is The target parameter value is The system frame rate is If the frame rate is 1 frame / second, then the parameter increment per frame is:

[0247] In the process of adjusting the field of view of the electronic rearview mirror, this application can update the current parameters according to each frame update. Adjust gradually until the target parameter value is reached. For situations where multiple parameters change simultaneously, the increment for each parameter can be calculated independently and updated synchronously, ensuring that all parameters complete the transition within the same time window.

[0248] In addition, to allow users to clearly perceive the scene transition process, this application can overlay visual cues on the display screen during the transition period.

[0249] Simultaneously, a semi-transparent prompt bar is displayed at the top or bottom of the screen, indicating that the scene is transitioning to a high-speed driving scene, accompanied by a brief animation effect (such as a progress bar or fade-in / fade-out), until the prompt bar disappears automatically after the transition is complete. Additionally, this application can briefly display a scene icon (such as a car icon for a high-speed driving scene) during the transition to further enhance visual recognition.

[0250] To prevent scene switching from adversely affecting driving safety, this application can also set multiple safety protection measures. First, when the vehicle speed exceeds a certain threshold (e.g., 100 km / h), large-scale field of view adjustments are prohibited, allowing only minor adjustments to image enhancement parameters. Second, upon detecting emergency braking or a collision warning signal, the scene switching process can be immediately terminated, locking the current field of view configuration to prevent sudden changes in field of view from distracting the user. Finally, this application can record user behavior data (such as gaze shifts, steering wheel operations, etc.) before and after each scene switch, analyze the impact of scene switching on driving behavior, promptly identify potential safety risks, and adjust the switching strategy accordingly.

[0251] In this application, as Figure 9 As shown, this application solves the problem of abrupt changes in field of vision through a parameter smoothing transition mechanism and a visual feedback mechanism, ensuring a smooth and continuous process of changing the field of vision, thereby improving driving safety and visual comfort. Specifically, this application can obtain the current field of vision parameters of the electronic rearview mirror, providing a clear starting point for adjustment, ensuring that the adjustment is based on the actual field of vision rather than switching out of thin air, thus avoiding abrupt changes caused by an unclear starting point.

[0252] Based on this, the frame parameter increment can be determined according to the current field of view parameters and the configuration parameters corresponding to the target scene template. By calculating the difference between the current and target parameters and decomposing it into small changes frame by frame, the parameters are gradually transitioned rather than switched instantaneously, which significantly reduces the visual impact on users and the risk of distraction.

[0253] Finally, this application adjusts the field of view of the electronic rearview mirror based on the frame parameter increment and the configuration parameters corresponding to the target scene template. The field of view is gradually adjusted by applying increments while anchoring the target parameters, ensuring the controllability of the change process and ultimately achieving the target configuration. This maintains the continuity of vision, allowing users to experience seamless and non-intrusive changes in field of view, making the intelligent recommendation function more easily accepted by users, and further improving safety during driving.

[0254] In some embodiments, the electronic rearview mirror field of view adjustment method further includes: receiving a scene mode switching command input by a user; if the scene mode switching command is a preset fourth command, displaying at least one of a preset scene template, a custom scene template, and an intelligent scene template; if a target command is received, determining a target scene template from at least one of the preset scene template, the custom scene template, and the intelligent scene template; if the scene mode switching command is a preset fifth command, determining the vehicle's target scene template from a scene template library based on at least one of driving status information, road information, and environmental information.

[0255] In this embodiment, the scene mode switching command can be a signal actively issued by the user to change the electronic rearview mirror's field of view adjustment mode. The scene mode switching command serves as an entry point for the user to actively intervene in and control the electronic rearview mirror's field of view adjustment process. Simultaneously, this application allows entry into a one-click manual switching mode via the scene mode switching command.

[0256] Specifically, users can use virtual buttons or menu options on the in-vehicle central control screen for touch input, or use physical buttons on the steering wheel or voice recognition system for voice command input, or even use gestures or clicks on the touch area of ​​the electronic rearview mirror display to trigger scene mode switching.

[0257] If the scene mode switching command is the fourth command, at least one of the preset scene template, custom scene template, and smart scene template will be displayed. The fourth command is a specific user input predefined by the system to trigger the scene template display function. Display can be understood as presenting the available scene templates to the user in a visual way, thereby allowing the user to browse and select the available scene templates and achieve manual switching.

[0258] For example, the fourth command could be a click on a specific icon (such as a scene mode icon) on the central control screen. After the system responds, a list or grid view pops up on the central control screen, displaying all available preset scene templates, custom scene templates, and smart scene templates. Alternatively, the fourth command could be a voice command (such as displaying scene templates). After the system responds, the names and icons of some or all scene templates are displayed on the electronic rearview mirror display or the central control screen in the form of floating cards or sidebars.

[0259] If a target instruction is received, the system determines the target scene template from at least one of the preset scene templates, custom scene templates, and intelligent scene templates. The target instruction is the user's instruction to select a specific scene template on the display interface, while determining the target scene template means that the system identifies and confirms the scene template selected by the user to ensure that the user's selection is accurately recognized by the system and used for subsequent field of view adjustments.

[0260] For example, the target instruction could be that the user clicks on the name or icon of a scene template in the display list or grid, and after receiving the click event, the system uses the corresponding scene template as the target scene template; or, the target instruction could also be that the user directly specifies a displayed scene template through voice command (e.g., select high-speed mode), and the system determines the corresponding target scene template through voice recognition and semantic understanding.

[0261] If the scene mode switching command is the fifth command, the system determines the vehicle's target scene template from the scene template library based on at least one of the driving status information, road information, and environmental information. The fifth command is another specific user input predefined by the system to trigger the automatic scene recognition function. Determining the vehicle's target scene template from the scene template library based on at least one of the driving status information, road information, and environmental information allows the system to revert to a mode based on intelligent recognition and recommendation using sensor data. This allows users to switch between a one-click manual switching mode and an intelligent recommendation mode, providing greater flexibility.

[0262] In this application, by adding a user-input scene mode switching instruction processing mechanism, the system's flexibility and user control are improved. At the same time, through the dual-mode collaborative mechanism, the efficiency of intelligent recommendation is retained while giving users active control, effectively making up for the shortcomings of the original solution. This makes the field of view adjustment of the electronic rearview mirror more in line with the user's personalized needs and driving habits, thereby improving the driving experience and safety.

[0263] In some embodiments, generating and displaying recommended information for the target scene template includes: identifying the user in the vehicle by using at least one of seat memory button recognition, facial recognition, Bluetooth device recognition, and car key recognition to obtain the user's identity information; and generating and displaying recommended information for the target scene template based on the identity information.

[0264] In this embodiment, user identification is the foundation for personalized recommendations. This application accurately identifies the identity of the current user, thereby loading the user's exclusive configuration and preference data to generate more targeted recommendation information.

[0265] Meanwhile, this application also enables personalized configuration management for multiple users, solving the problems of configuration isolation and automatic switching in scenarios where multiple people share a vehicle. It also allows for the maintenance of an independent scenario template library and usage habit data for each user. In addition, user identification technology can also achieve automatic loading and switching of configurations.

[0266] Specifically, such as Figure 7 As shown, this application can employ multiple methods for user identification. For example, the priority order from highest to lowest is: seat memory button recognition, facial recognition, Bluetooth device recognition, and car key recognition.

[0267] For seat memory button recognition, when a user presses a vehicle seat memory button (usually button 1, 2, 3, etc.), the seat memory position can be associated with the user's identity. For example, seat memory 1 corresponds to user A, and seat memory 2 corresponds to user B.

[0268] For facial recognition, the user's facial image is captured by the in-vehicle camera of the user monitoring system, and the facial recognition algorithm is used for identity matching.

[0269] For Bluetooth device identification, it can detect Bluetooth devices (such as mobile phones) connected in the car and identify the user's identity through the device's MAC address. It is suitable for scenarios where users habitually use a fixed mobile phone to connect to the car's Bluetooth.

[0270] For car key recognition, for vehicles equipped with multiple smart keys, the unique ID information of the currently used car key can be read and associated with a preset user identity to achieve user identification.

[0271] Furthermore, after obtaining the user's identity information, this application can extract personalized preferences from the user's exclusive data based on that identity information, thereby generating highly customized recommendation information to ensure that the recommended content matches the user's actual needs and habits.

[0272] In addition, after identifying a specific user, this application can prioritize loading the custom scene template created by that user, and combine it with at least one of the current driving status information, road information, and environmental information to determine the target scene template from the scene template library, thereby generating recommendation information that better meets the user's personalized needs. For example, it can recommend the user's exclusive "My Commuting Mode" instead of the general city road mode.

[0273] In addition, each user can have an independent AI recommendation engine instance. This application can call the corresponding user's recommendation engine based on the identified identity information, and calculate the recommendation confidence based on the scene matching degree of the scene template, the user's historical acceptance rate, spatiotemporal similarity, and user activity, thereby generating highly personalized recommendation information.

[0274] In some embodiments, this application may also employ a multi-user profile isolation architecture, providing each user with an independent configuration data space. For example... Figure 7 As shown, the configuration file contains basic information, a scene template library, usage habit data, preference settings, and an independent AI recommendation engine instance.

[0275] Basic information includes user name, height, seat memory position, etc.; the scene template library includes user-created user-defined templates and AI-generated learning templates for the user. Preset scene templates can be shared resources and are accessible to all users; usage habit data can record the user's scene switching history, manual adjustment records, recommendation feedback records, etc., for AI learning and statistical analysis; preference settings can record the user's preference values ​​for parameters such as brightness, contrast, and transparency, for fine-tuning the default parameters of scene templates; independent AI recommendation engine instances can include each user's independent scene preference vector, historical acceptance rate, spatiotemporal similarity model, and user activity, ensuring that the recommendation strategy is completely personalized.

[0276] In addition, this application can also set up an isolation mechanism for multi-user recommendation strategies, thereby ensuring that when multiple people share a vehicle, each user can receive personalized recommendations based on their own driving habits without interfering with each other.

[0277] The isolation mechanism allows each user to maintain an independent recommendation engine instance, ensuring that when multiple people share a vehicle, each user receives personalized recommendations based on their own driving habits without interference. The specific implementation is as follows: 1. User A's scenario preference vector User B's scenario preference vector Completely independent; 2. Historical acceptance rate According to individual user statistics, user A's rejection of a certain scenario does not affect user B's recommendations; 3. Spatiotemporal similarity Calculations are based on each individual's historical usage records, and historical data is not shared. 4. User activity It reflects each user's own response habits.

[0278] Furthermore, when a user switches identities (e.g., from user A to user B), the recommendation engine instance and historical data of user A can be unloaded, while the recommendation engine instance and historical data of user B can be loaded. Additionally, the recommendation confidence of all scenario templates can be recalculated based on user B's preferences. .

[0279] In addition, this application can also set up a seat memory linkage mechanism. When the user presses the seat memory button, not only will the seat position, physical rearview mirror angle and steering wheel height be adjusted, but the user's electronic rearview mirror configuration will also be loaded simultaneously, including scene template library, default field of view parameters, AI recommendation engine status, etc.

[0280] The trigger window for the seat memory linkage mechanism is usually before the seat adjustment is completed. This application can start loading the configuration as soon as the seat begins to move, ensuring that the configuration is also loaded when the seat adjustment is complete, achieving a seamless switch. Simultaneously, this application can also display a welcome message on the central control display screen, such as "Welcome back, Zhang San! Your scene configuration has been loaded," thereby enhancing the user experience.

[0281] In the electronic rearview mirror field of view adjustment method provided in this embodiment of the invention, a target scene template is determined from the scene template library by using at least one of the vehicle's current driving status information, road information, and environmental information. This allows the user to be recommended to use the configuration parameters corresponding to the target scene template to adjust the field of view of the vehicle's electronic rearview mirror. This solves the technical problem of rigid field of view configuration of electronic rearview mirrors, ensuring that the electronic rearview mirror can provide the user with the optimal field of view under different operating conditions. This effectively reduces the user's blind spots, meets the diverse needs of users in different driving scenarios, and improves driving safety.

[0282] In some embodiments, the present invention also provides a field-of-view adjustment device 300 for an electronic rearview mirror, which is used to perform any of the aforementioned field-of-view adjustment methods for electronic rearview mirrors.

[0283] Specifically, please refer to Figure 11 , Figure 11 This is a schematic block diagram of the field of view adjustment device 300 for an electronic rearview mirror provided in an embodiment of the present invention.

[0284] like Figure 11 As shown, the electronic rearview mirror field of view adjustment device 300 provided in this application includes: an acquisition unit 310, a determination unit 320 and an adjustment unit 330.

[0285] The acquisition unit 310 is used to acquire the vehicle's current driving status information, road information, and environmental information; the determination unit 320 is used to determine a target scene template from the scene template library based on at least one of the driving status information, road information, and environmental information; the scene template library includes at least one preset scene template; the adjustment unit 320 is used to generate and display recommendation information for the target scene template, and after receiving the user's application instruction for the recommendation information, adjust the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0286] The electronic rearview mirror field of view adjustment device 300 provided in this application embodiment can determine a target scene template from a scene template library by using at least one of the vehicle's current driving status information, road information, and environmental information. This allows the user to be recommended to use the configuration parameters corresponding to the target scene template to adjust the field of view of the vehicle's electronic rearview mirror. This solves the technical problem of rigid field of view configuration of electronic rearview mirrors, ensuring that the electronic rearview mirror can provide the user with the optimal field of view under different operating conditions. It effectively reduces the user's blind spots, meets the diverse needs of users in different driving scenarios, and improves driving safety.

[0287] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned electronic rearview mirror field of view adjustment device and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0288] The aforementioned electronic rearview mirror's field-of-view adjustment device can be implemented as a computer program, which can, for example... Figure 12 It runs on the electronic device shown.

[0289] Please see Figure 12 , Figure 12 This is a schematic block diagram of an electronic device provided in an embodiment of the present invention.

[0290] See Figure 12 The device 400 includes a processor 402, a memory, and a network interface 405 connected via a system bus 401, wherein the memory may include a storage medium 403 and internal memory 404.

[0291] The storage medium 403 may store an operating system 4031 and a computer program 4032. When the computer program 4032 is executed, it causes the processor 402 to execute a method for adjusting the field of view of the electronic rearview mirror.

[0292] The processor 402 provides computing and control capabilities to support the operation of the entire device 400.

[0293] The internal memory 404 provides an environment for the operation of the computer program 4032 in the non-volatile storage medium 403. When the computer program 4032 is executed by the processor 402, the processor 402 can perform the field of view adjustment method of the electronic rearview mirror.

[0294] This network interface 405 is used for network communication, such as providing data transmission. Those skilled in the art will understand that... Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the device 400 to which the present invention is applied. The specific device 400 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0295] The processor 402 is used to run a computer program 4032 stored in the memory to perform the following functions: acquiring the vehicle's current driving status information, road information, and environmental information; determining a target scene template from a scene template library based on at least one of the driving status information, road information, and environmental information; the scene template library includes at least one preset scene template; generating and displaying recommendation information for the target scene template; and adjusting the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template after receiving the user's application instruction for the recommendation information.

[0296] Those skilled in the art will understand that Figure 12 The embodiments of device 400 shown do not constitute a limitation on the specific configuration of device 400. In other embodiments, device 400 may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, in some embodiments, device 400 may include only memory and processor 402. In such embodiments, the structure and function of memory and processor 402 are similar to those shown. Figure 12 The embodiments shown are consistent and will not be described again here.

[0297] It should be understood that, in this embodiment of the invention, the processor 402 may be a Central Processing Unit (CPU), or it may be another general-purpose processor 402, a digital signal processor 402 (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 402 may be a microprocessor 402, or it may be any conventional processor 402, etc.

[0298] According to one aspect of this application, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the following steps: acquiring current driving state information, road information, and environmental information of the vehicle; determining a target scene template from a scene template library based on at least one of the driving state information, road information, and environmental information; the scene template library includes at least one preset scene template; generating and displaying recommendation information for the target scene template; and, upon receiving an application instruction from a user regarding the recommendation information, adjusting the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0299] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0300] In another embodiment of the present invention, a computer storage medium is provided. This storage medium may be a non-volatile computer-readable storage medium or a volatile storage medium. The storage medium stores a computer program 4032, which, when executed by a processor 402, performs the following steps: acquiring current driving state information, road information, and environmental information of the vehicle; determining a target scene template from a scene template library based on at least one of the driving state information, road information, and environmental information; the scene template library includes at least one preset scene template; generating and displaying recommendation information for the target scene template; and, upon receiving an application instruction from a user regarding the recommendation information, adjusting the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

[0301] The storage medium can be any computer-readable storage medium that can store program code, such as a USB flash drive, external hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0302] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0303] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0304] The steps in the methods of this application embodiment can be adjusted, merged, or deleted according to actual needs. The units in the apparatus of this application embodiment can be merged, divided, or deleted according to actual needs. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0305] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods provided in the various embodiments of this application.

[0306] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for adjusting the field of view of an electronic rearview mirror, characterized in that, include: Acquire current vehicle driving status information, road information, and environmental information; A target scene template is determined from the scene template library based on at least one of the driving status information, the road information, and the environmental information. The scene template library includes at least one preset scene template; The system generates and displays recommended information for the target scene template, and after receiving the user's application instruction for the recommended information, adjusts the field of view of the vehicle's electronic rearview mirror based on the configuration parameters corresponding to the target scene template.

2. The method for adjusting the field of view of an electronic rearview mirror according to claim 1, characterized in that, The scene template library also includes custom scene templates, and the method further includes: Based on the user's operation instructions, determine the first triggering condition and the first field of view adjustment parameters corresponding to the creation of the custom scene template; The custom scene template is generated based on the first triggering condition and the first field of view adjustment parameters.

3. The method for adjusting the field of view of an electronic rearview mirror according to claim 2, characterized in that, The scene template library also includes intelligent scene templates, and the method further includes: Obtain a preset number of the user's most recent adjustment records, each of which includes scene information and field of view condition parameters; Cluster analysis was performed on the adjustment records to confirm the second triggering condition and the second field of view condition parameters; The intelligent scene template is generated based on the second triggering condition and the second field of view condition parameters.

4. The method for adjusting the field of view of an electronic rearview mirror according to any one of claims 1-3, characterized in that, The step of determining a target scene template from the scene template library based on at least one of the driving state information, the road information, and the environmental information includes: The scene matching degree of each scene template in the scene template library is determined based on at least one of the driving status information, the road information, and the environmental information. The target scene template is determined from the scene template library based on the scene matching degree of each scene template.

5. The method for adjusting the field of view of an electronic rearview mirror according to claim 4, characterized in that, The step of determining the target scene template from the scene template library based on the scene matching degree of each scene template includes: The target scene template is determined from scene templates whose scene matching degree reaches a preset matching degree threshold.

6. The method for adjusting the field of view of an electronic rearview mirror according to claim 5, characterized in that, Determining the target scene template from scene templates whose scene matching degree reaches a preset matching degree threshold includes: Scene templates that meet a preset matching threshold are identified as scene templates to be recommended. Determine the recommendation confidence level of each of the proposed scenario templates, and determine the target scenario template from the plurality of proposed scenario templates based on the recommendation confidence level; Determining the recommendation confidence of each of the proposed scenario templates includes: The historical acceptance rate of each of the proposed scenario templates is determined, the spatiotemporal similarity between the spatiotemporal environment in which the vehicle is located when the user uses each of the proposed scenario templates and the spatiotemporal environment in which the vehicle is currently located, and the user activity level of the user responding to the vehicle recommending each of the proposed scenario templates; based on the scenario matching degree, the first weight corresponding to the scenario matching degree, the historical acceptance rate, the second weight corresponding to the historical acceptance rate, the spatiotemporal similarity, the third weight corresponding to the spatiotemporal similarity, the user activity level, and the fourth weight corresponding to the user activity level, a recommendation confidence level for each of the proposed scenario templates is generated.

7. The method for adjusting the field of view of an electronic rearview mirror according to claim 6, characterized in that, The method further includes: Determine the temporal similarity between the time when the vehicle is in each of the recommended scene templates and the current time of the vehicle, and the spatial similarity between the location of the vehicle when the user uses each of the recommended scene templates and the current location of the vehicle; The spatiotemporal similarity is obtained by weighted summation of the temporal similarity and the spatial similarity.

8. The method for adjusting the field of view of an electronic rearview mirror according to claim 7, characterized in that, The step of determining the target scene template from multiple scene templates to be recommended based on the recommendation confidence includes: If the confidence difference between the recommendation confidence of each of the proposed scenario templates meets the preset conditions, the user's preference score for each of the proposed scenario templates is determined. Based on the preference score, the target scene template is determined from a plurality of scene templates to be recommended.

9. The method for adjusting the field of view of an electronic rearview mirror according to claim 8, characterized in that, The preference score is calculated as follows: The usage count of each of the proposed scenario templates, the first acceptance count of each of the proposed scenario templates, and the fifth weight corresponding to the first acceptance count are determined. Based on the number of uses, the first number of acceptances, and the fifth weight corresponding to the first number of acceptances, a preference score is generated for the user for each of the recommended scenario templates.

10. The method for adjusting the field of view of an electronic rearview mirror according to claim 4, characterized in that, The step of determining the scene matching degree of each scene template in the scene template library based on at least one of the driving state information, the road information, and the environmental information includes: For each scene template, if the scene template is the preset scene template, then the matching degree calculation formula for the scene template is: ; Let be the matching degree of the i-th preset scene template. Let i be the weight of the i-th preset scene template in the j-th dimension. The dimension data of the i-th preset scene template in the j-th dimension; If the scene template is a custom scene template or an intelligent scene template, determine and calculate multiple dimensions of the scene template, and calculate the corresponding scene matching degree based on the multiple dimensions of the scene template using a logical expression tree method; each dimension of the scene template is one of the driving state information, the road information, and the environmental information.

11. The method for adjusting the field of view of an electronic rearview mirror according to claim 10, characterized in that, After receiving the user's application instruction regarding the recommendation information, it also includes: Generate a reward signal for the target scene template; Based on the reward signal, the weight of the preset scene template in the j-th dimension is dynamically adjusted.

12. The method for adjusting the field of view of an electronic rearview mirror according to claim 5, characterized in that, The method further includes: Determine the first rejection rate of the user in the current period for rejecting the vehicle recommendation scenario template; The preset matching threshold is determined based on the preset target rejection rate and the first rejection rate.

13. The method for adjusting the field of view of an electronic rearview mirror according to claim 6, characterized in that, After receiving the user's application instruction regarding the recommendation information, it also includes: Determine the number of times the user has refused and the number of times the user has accepted in the current period; Based on the number of rejections and the number of acceptances, the first weight, the second weight, the third weight, and the fourth weight are adjusted to obtain the adjusted first weight, second weight, third weight, and fourth weight.

14. An electronic device, characterized in that, The device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the field of view adjustment method of the electronic rearview mirror according to any one of claims 1 to 13.

Images