Vehicle front hood unlocking method, device, equipment and medium

Abstract

The present application provides a method, device, equipment and medium for unlocking a vehicle hood, the method comprising: obtaining vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, and predicting a target person's need to open the hood based on the vehicle maintenance data and the vehicle abnormality data; if the prediction result shows that the target person has a potential need to open the hood, the image information is identified to determine the target person's behavioral intention based on the identification result; when the target person's behavioral intention is to open the vehicle hood, the vehicle hood is controlled to be unlocked; by judging the target person's potential need and actual behavioral intention, the target person's actual need to open the vehicle hood is accurately identified, and the hood is controlled to be automatically unlocked, thereby avoiding the trouble of not being able to open the hood normally due to forgetting to unlock it manually, and providing convenience for users.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a method, device, equipment and medium for unlocking a vehicle hood. Background Art

[0002] With the rapid advancement of modern automotive technology, the design of vehicle hoods has evolved far beyond basic functionality. Convenience and safety have become two key priorities for the majority of users. As a vital component of the vehicle, the opening method of the hood (commonly referred to as the engine hood or bonnet) directly impacts the user's daily experience. For a long time, unlocking the hood has relied primarily on manual intervention by the driver, such as using an unlock button inside the vehicle or remote control using the key fob.

[0003] However, this traditional unlocking method can sometimes be inadequate in practical applications, lacking sufficient intelligence and user-friendly considerations. This is especially true when users forget to manually unlock the hood due to carelessness, busy schedules, or other reasons. This often leads to the awkward situation of being unable to open the hood in time, causing unnecessary inconvenience and potentially affecting their travel plans and mood. Summary of the Invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a vehicle front hood unlocking method, device, equipment and medium to solve the above-mentioned technical problems.

[0005] The present application provides a method for unlocking a vehicle hood, the method comprising: acquiring vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, the vehicle maintenance data comprising at least one of maintenance data of preset maintenance items and vehicle fluid maintenance data, the image information comprising portrait information of at least one target person; predicting the target person's need to open the hood based on the vehicle maintenance data and the vehicle abnormality data; if the prediction result shows that the target person has a potential need to open the hood, identifying the image information to determine the target person's behavioral intention based on the identification result; and controlling the vehicle hood to be unlocked when the target person's behavioral intention is to open the vehicle hood.

[0006] In one embodiment of the present application, the target person's need to open the hood is predicted based on the vehicle maintenance data and the vehicle abnormality data, including: obtaining the current time, the current remaining amount of vehicle fluid, and the door signal; comparing the current time with the maintenance data of the preset maintenance item to determine whether the current time is at a maintenance time point; based on the vehicle fluid maintenance data, predicting the fluid replenishment demand state under the current vehicle fluid remaining amount state; judging whether the vehicle has an accident or failure within a preset time based on the abnormal data, and evaluating the target person's intention to repair in combination with the door signal; if the current time is at a maintenance time point, or there is a need for fluid replenishment, or the target person has an intention to repair, it is determined that the target person has a potential need to open the hood.

[0007] In one embodiment of the present application, the portrait information is identified to determine the behavioral intention of the target person based on the identification result, including: extracting facial feature points, body feature points, and hand feature points from the portrait information of the target person; identifying the relative position relationship between each facial feature point to determine the orientation of the target person; when the target person is facing the front hood of the vehicle, detecting whether the target person has a bending action based on the change in the angle formed by the body feature points; if a bending action is detected, identifying the target person's gesture based on the hand feature points, and determining the target person's behavioral intention in combination with the target person's orientation, bending state and gesture.

[0008] In one embodiment of the present application, before identifying the relative positional relationship between each facial feature point to determine the orientation of the target person, it also includes: obtaining the initial coordinates of each facial feature point; with the initial coordinates as the center, cropping out a first cropped image area containing all the initial coordinates, and preliminarily positioning each facial feature point in the first cropped image area to obtain the first updated coordinates; with the first updated coordinates as the new center point, cropping out a cropped image area containing all the updated coordinates again, and performing a second positioning of each facial feature point in the cropped image area again to obtain the again updated coordinates; repeating the above cropping and positioning process until the obtained facial feature point coordinates meet the preset accuracy conditions.

[0009] In one embodiment of the present application, based on the change in the angle formed by the body feature points, detecting whether the target person has a bending action includes: obtaining the coordinates of each body feature point, the body feature points including at least a waist feature point, a chest feature point, and a leg feature point; taking the waist feature point as the vertex, calculating the angle formed between the chest feature point and the leg feature point; if the angle is less than a preset angle threshold, it is determined that the target person has a bending action.

[0010] In one embodiment of the present application, the gesture of the target person is identified based on the hand feature points, and the behavioral intention of the target person is determined in combination with the orientation, bending state and gesture of the target person, including: when the target person is facing the hood of the vehicle, and the target person is in a bending state, and the target person's gesture matches at least one preset gesture, then it is determined that the target person has the intention to open the hood of the vehicle, and the preset gesture includes at least one of the following: knocking on the hood, a preset specific hood opening gesture, and the hand touching the hood sensor at the front of the vehicle.

[0011] In one embodiment of the present application, before controlling the unlocking of the vehicle's hood, it also includes: pre-logging into the vehicle owner's account and entering the authorization information, wherein the authorization information includes at least one of facial information and image information; based on the authorization information, authenticating the target person who has the behavioral intention to open the vehicle's hood; if the verification is passed, sending a request to unlock the hood to the body domain controller to trigger the body domain controller to identify the operating status of the vehicle's hood; when the operating status of the hood is an openable state, executing the unlocking operation of the hood.

[0012] The present application provides a vehicle hood unlocking device, which includes: a data acquisition module for acquiring vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, wherein the vehicle maintenance data includes at least one of maintenance data of preset maintenance items and vehicle fluid maintenance data, and the image information includes at least one human portrait; a demand prediction module for predicting a target person's demand for opening the hood based on the vehicle maintenance data and the vehicle abnormality data; a behavior intention recognition module for recognizing the image information when the prediction result shows that the target person has a potential demand to open the hood, so as to determine the target person's behavior intention based on the recognition result; and an unlocking control module for controlling the vehicle hood to unlock when the target person's behavior intention is to open the vehicle hood.

[0013] The present application provides an electronic device, comprising a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the vehicle hood unlocking method as described above.

[0014] The present application provides a computer-readable storage medium, characterized in that a computer program is stored thereon, and the computer program is used to enable a computer to execute the vehicle front hood unlocking method as described above.

[0015] Beneficial effects of this application: The vehicle hood unlocking method proposed in this application, through comprehensive analysis of vehicle maintenance data, vehicle abnormality data, and real-time external image information related to the use of the vehicle hood, can intelligently predict the target person's need to open the hood, and further confirm the target person's actual behavioral intention through image recognition technology, and then open the hood based on the target person's actual behavioral intention. Among them, demand-based recognition greatly reduces the inconvenience encountered by users due to forgetting to manually unlock the vehicle, making the opening of the vehicle hood smoother and more natural, significantly improving the user's daily car experience; in addition, its dual verification mechanism effectively avoids the risk of misoperation or illegal unlocking, providing additional security for the vehicle and user property. In summary, this method not only solves the pain point of users forgetting to unlock the hood due to negligence, but also improves the convenience, safety, and efficiency of vehicle use through intelligent and precise technical means.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are incorporated into and constitute a part of the specification, illustrating embodiments consistent with the present application and, together with the specification, serving to explain the principles of the present application. It is obvious that the drawings described below are merely some embodiments of the present application, and a person of ordinary skill in the art can derive other drawings based on these drawings without inventive effort. In the drawings:

[0018] Figure 1 1 is a schematic diagram of an implementation environment of a vehicle front hood unlocking method according to an exemplary embodiment of the present application;

[0019] Figure 2 1 is a schematic diagram of the implementation steps of a method for unlocking a vehicle front hood according to an exemplary embodiment of the present application;

[0020] Figure 3 1 is a schematic diagram of a target person's facial orientation recognition step in a vehicle front hood unlocking method according to an exemplary embodiment of the present application;

[0021] Figure 4 is a block diagram of a vehicle front hood unlocking device shown in an exemplary embodiment of the present application;

[0022] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0023] The following will describe the embodiments of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand the other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for the purpose of illustrating the present application and are not intended to limit the scope of protection of the present application.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0025] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0026] First of all, it should be noted that CNN is a type of feedforward neural network (Feedforward Neural Networks) that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning.

[0027] DCNN, an extension of CNN, increases model depth by stacking multiple convolutional, pooling, and fully connected layers, thereby improving image classification accuracy. DCNN simulates the workings of the human visual system and achieves efficient feature extraction and classification by performing multi-level convolution and pooling operations on images.

[0028] OpenPose,OpenPose is an open source library for real-time detection of body, feet, hands, and facial key points.

[0029] Figure 1 It is a schematic diagram of an implementation environment of a vehicle front hood unlocking method shown in an exemplary embodiment of the present application.

[0030] like Figure 1As shown, the implementation environment of the vehicle hood unlocking method primarily includes a data acquisition module 101 and a computer module 102. Data acquisition module 101 is responsible for collecting multi-dimensional data related to vehicle hood unlocking. This data includes vehicle hood usage and maintenance data, vehicle anomaly data, and real-time image information from the vehicle's exterior. Specifically, vehicle hood usage and maintenance data may be derived from historical vehicle usage records, user preferences, etc.; vehicle anomaly data is provided in real time by status monitoring sensors for various vehicle systems, such as engine status and door status; and image information from the vehicle's exterior is captured by high-definition cameras installed around the vehicle. Data acquisition module 101 can be an integration of various sensors and cameras built into the vehicle, or it can be a remote device connected via an onboard communication network. This application does not impose any restrictions on the specific configuration and connection method of data acquisition module 201.

[0031] Computer module 102 is responsible for efficiently processing the various types of data collected by data acquisition module 101. It first uses advanced algorithmic models to predict whether the target person has a potential need to open the hood, based on vehicle maintenance data and vehicle anomaly data. If the prediction indicates such a need, computer module 102 immediately performs deep recognition of image information outside the vehicle to accurately determine the target person's actual behavioral intentions. During this process, the computer module uses image recognition technology to analyze the target person's movements and posture to determine whether they actually intend to open the vehicle's hood. Once it confirms that the target person's behavioral intention matches the prediction, computer module 102 quickly generates an unlock command, automatically unlocking the vehicle's hood. This design not only avoids the inconvenience caused by users forgetting to manually unlock the vehicle, but also greatly improves the user experience through intelligent and automated methods.

[0032] It is worth noting that the computer module 102 can be a high-performance computing unit built into the vehicle, such as an onboard computer or embedded system, or a remote server connected to the vehicle via cloud computing technology. This application also does not impose any restrictions on the specific type and deployment method of the computer module 102 to ensure the flexibility and wide applicability of the solution.

[0033] Figure 2 It is a schematic diagram of the implementation steps of a method for unlocking a vehicle front hood shown in an exemplary embodiment of the present application.

[0034] like Figure 2 As shown, in an exemplary embodiment, the vehicle front hood unlocking method includes at least steps S210 to S240, which are described in detail as follows:

[0035] Step S210, obtaining vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, wherein the vehicle maintenance data includes at least one of maintenance data of preset maintenance items and vehicle fluid maintenance data, and the image information includes portrait information of at least one target person.

[0036] In one embodiment of the present application, vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle are collected and processed respectively, as follows:

[0037] Regarding obtaining vehicle maintenance data: Maintenance data for pre-set maintenance items includes the vehicle's scheduled maintenance cycles and the specific time of each maintenance. Pre-set maintenance items include some that require opening the hood for maintenance. These pre-set maintenance items vary depending on the factory settings of different vehicle models, and this application does not impose any specific restrictions on the specific content of these pre-set maintenance items. Vehicle fluids include windshield washer fluid, coolant, and engine oil. Vehicle fluid maintenance data includes the remaining amount of each vehicle fluid added. Therefore, by connecting to the vehicle manufacturer's cloud service, the vehicle's scheduled maintenance records are automatically synchronized. These records include, but are not limited to, timestamps for key maintenance items such as oil changes, brake system inspections, and fluid changes. Additionally, the owner or driver can manually enter or confirm maintenance times through the in-vehicle app to ensure data accuracy. Built-in sensors in the vehicle continuously monitor the consumption of fluids in the fuel tank, fluid tank, brake fluid reservoir, and other fluids, and record the time of each refill or replacement, as well as the remaining amount at the time of each refill.

[0038] Regarding obtaining vehicle abnormality data: The vehicle is equipped with a comprehensive condition monitoring system, including engine fault monitoring, electrical system fault monitoring, tire pressure monitoring, etc. When the system detects any abnormality (such as engine fault code, low tire pressure, etc.), it will immediately record the abnormality data, including the type of abnormality, time of occurrence, duration, etc., and may be accompanied by an alarm prompt.

[0039] Regarding the acquisition of image information: Multi-angle high-definition cameras are installed on the outside of the vehicle, especially on the front and sides of the vehicle, to ensure that the target person approaching the vehicle can be captured. The camera has night vision and wide-angle functions to ensure clear imaging under different lighting conditions and viewing angles. Then, advanced facial recognition technology is used to analyze the images captured by the camera in real time to identify the portrait information of at least one target person. The recognition results include the facial features, position, movements, etc. of the target person, providing key data for subsequent behavioral intention judgment. Then, all the collected data (vehicle maintenance data, abnormal data, image information) will be sent to the vehicle's built-in computer module or remote server for fusion processing. Through the comprehensive analysis of the data by the algorithm model, the system can more accurately predict the target person's need to open the front hood, and control the unlocking of the front hood based on their behavioral intention.

[0040] It can be understood that through the above-described embodiment, the system can comprehensively and accurately obtain vehicle maintenance data, vehicle anomaly data, and vehicle exterior image information containing the target person's portrait information. This data provides the system with a rich basis for decision-making, enabling it to intelligently determine the target person's potential needs and actual behavioral intentions, thereby implementing the automatic unlocking function of the vehicle's hood, greatly improving user convenience and safety.

[0041] Step S220 : predicting the hood opening demand of the target person based on the vehicle maintenance data and the vehicle abnormality data.

[0042] In one embodiment of the present application, the target person's need to open the hood is predicted based on vehicle maintenance data and vehicle abnormality data, including: obtaining the current time, the current remaining amount of vehicle fluid, and the door signal; comparing the current time with the maintenance data of the preset maintenance item to determine whether the current time is at a maintenance time point; based on the vehicle fluid maintenance data, predicting the fluid replenishment demand state under the current vehicle fluid remaining amount state; judging whether the vehicle has an accident or failure within the preset time based on the abnormal data, and evaluating the target person's intention to repair in combination with the door signal; if the current time is at a maintenance time point, or there is a need for fluid replenishment, or the target person has an intention to repair, it is determined that the target person has a potential need to open the hood.

[0043] In one specific embodiment of the present application, the system first accurately obtains factory information for the battery pack and motor from the vehicle manufacturer's database, including key parameters such as production date, expected lifespan, and maintenance cycle. It then receives real-time data calculated by the battery pack and motor sensors and the intelligent diagnostic system based on a lifespan prediction model. This data is closely integrated with maintenance reminder information from the maintenance reminder system to accurately determine the scheduled maintenance time for the battery pack and motor. Furthermore, the system continuously records the user's windshield washer fluid and coolant usage and cycle times through the vehicle's built-in high-sensitivity sensors, providing real-time information on the current remaining levels of these two fluids, providing a solid data foundation for subsequent demand forecasting. Furthermore, the system monitors the operating status of the motor and battery pack in real time, and any fault signals detected are immediately recorded and reported. Furthermore, through the vehicle's safety system or collision sensors, the system can quickly detect vehicle accident signals, ensuring a comprehensive understanding of the vehicle's status.

[0044] Based on this collected data, the system first obtains the current time and accurately compares it with the scheduled maintenance times for the battery pack and motor to determine whether it is currently at or approaching a maintenance period. Next, based on the user's past windshield washer fluid and coolant usage habits and the current remaining amount of windshield washer fluid or coolant, it predicts whether additional windshield washer fluid or coolant will be needed within the next usage cycle. In the event of a vehicle breakdown or accident, this abnormal data is immediately recorded and combined with door signals (such as a door being opened after an accident) to assess whether the target person intends to repair the vehicle.

[0045] Finally, based on all the above information, the target person's need to open the front hood is determined: if the current time is the maintenance time point, it is determined that there is a need for regular maintenance; if there is a need for fluid replenishment, it is determined that there is a need to add windshield washer fluid or coolant; if the vehicle breaks down or has an accident and the door signal indicates an intention to inspect, it is determined that there is a need to open the front hood for inspection.

[0046] As can be understood, the above embodiment can comprehensively consider multiple dimensions of information, including maintenance data for pre-set maintenance items, user vehicle fluid maintenance data, vehicle abnormality data, and door signals, to accurately predict the target person's hood opening needs. This significantly improves user convenience and automatically adjusts vehicle settings based on the prediction results, such as enabling premature hood opening, enhancing the user experience.

[0047] In step S230 , if the prediction result shows that the target person has a potential need to open the front hood, the image information is recognized to determine the target person's behavioral intention based on the recognition result.

[0048] In one embodiment of the present application, portrait information is recognized to determine the behavioral intention of the target person based on the recognition result, including: extracting facial feature points, body feature points, and hand feature points from the portrait information of the target person; identifying the relative position relationship between each facial feature point to determine the orientation of the target person; when the target person is facing the front hood of the vehicle, detecting whether the target person has a bending action based on the change in the angle formed by the body feature points; if a bending action is detected, identifying the gesture of the target person based on the hand feature points, and determining the behavioral intention of the target person in combination with the orientation, bending state and gesture of the target person.

[0049] In a specific embodiment of the present application, advanced facial recognition technology is first used to accurately extract the facial feature points of the target person from the image captured by the camera. These key points include the eyes, nose, mouth, and facial contours, which provide a solid foundation for subsequent analysis. Then, through the human posture recognition algorithm, the system further identifies and extracts the target person's body feature points, such as shoulders, elbows, knees and other key parts, to construct a human skeleton model, which makes it possible to monitor the target person's movements. It is particularly noteworthy that the system pays special attention to the hand area and extracts hand feature points in detail, including fingertips, palm centers, etc., which lays a solid foundation for subsequent gesture recognition.

[0050] After acquiring this critical portrait information, the system begins analyzing the relative positional relationships between facial landmarks, particularly the arrangement of the eyes, nose, and mouth, as well as the orientation of the facial contours, to accurately determine the target person's orientation. When the system detects that the target person is facing the vehicle's hood, it continuously monitors changes in the angles formed by the body's landmarks, particularly the angle between the shoulders and waist, and the angle between the knees and the ground. When these angles undergo specific changes—a gradual decrease in the angle between the shoulders and waist, and an increase in the angle between the knees and the ground—the system can accurately determine that the target person is bending. Once a bending motion is detected, the system further analyzes the relative position of the fingertips and the center of the palm, as well as the degree of finger bending, to identify the target person's gestures, such as grasping, pointing, and waving.

[0051] Finally, the target person's behavioral intention is comprehensively judged based on their orientation, bending position, and recognized gestures. For example, if the target person is facing the vehicle's hood, bending over, and making a grasping gesture, it is determined that the target person intends to open the hood or perform some operation. If they make a pointing gesture, they may be indicating or inspecting a component inside the hood.

[0052] Figure 3 It is a schematic diagram of the target person's facial orientation recognition steps of the vehicle front hood unlocking method shown in an exemplary embodiment of the present application.

[0053] like Figure 3 As shown in the figure, the recognition of facial feature points requires multiple positioning to obtain the precise positioning of each feature point. Specifically: first, the image containing the portrait information of the target person is extracted to obtain multiple facial feature points of the target person, and the coarse positioning of each facial feature point is determined. Then, based on the coarse positioning, the facial feature points are further processed to obtain the corresponding initial positioning and precise positioning. Finally, the orientation of the target person is obtained by processing the precise positioning of each feature point.

[0054] In one embodiment of the present application, identifying the relative positional relationship between facial feature points to determine the orientation of a target person includes: upon detecting a signal indicating the driver has opened a vehicle door, starting a timer and estimating the time required for the driver to reach a preset location based on the driver's dynamic parameters, which include at least the driver's stride length and gait; during this time, capturing a facial image of the driver from the driver's portrait information to obtain a plurality of facial feature points; obtaining initial coordinates for each facial feature point; cropping a primary cropped image region containing all initial coordinates with the initial coordinates as the center, and performing preliminary positioning of each facial feature point within the primary cropped image region to obtain initial updated coordinates; cropping a secondary cropped image region containing all updated coordinates with the primary updated coordinates as the new center point, and performing secondary positioning of each facial feature point within the secondary cropped image region to obtain further updated coordinates; repeating the aforementioned cropping and positioning process until the obtained facial feature point coordinates meet a preset accuracy requirement. Finally, based on the precisely positioned coordinates of each facial feature point, calculating the relative positional relationship between the facial feature points to determine the facial orientation of the target person.

[0055] In one embodiment of the present application, the vehicle's built-in camera or other sensors are first used to accurately capture the driver's gait information, including key dynamic parameters such as stride length and gait pattern. These parameters are crucial for the system because they are used to estimate the time required for the driver to move from the vehicle door to a predetermined position (such as in front of the driver's seat).

[0056] Next, based on the driver's dynamic parameters, a timer is started to accurately estimate the time window required for the driver to reach the preset location. This time window is crucial for subsequent portrait information collection, as it ensures that the system can obtain sufficient and high-quality facial data before the driver reaches the preset location. During the estimated movement time, the vehicle's internal camera continuously and rapidly captures the driver's facial images. These images undergo preprocessing, such as grayscale conversion and binarization, so that the subsequent facial recognition algorithm can more accurately identify facial feature points and record their initial coordinates.

[0057] Then, with the initial coordinates as the center, a cropped image region containing all the initial coordinates is cropped, and the facial landmarks are initially located within this region to obtain the initial updated coordinates. Next, a smaller cropped image region containing all the updated coordinates is cropped again, using the initial updated coordinates as the new center point, and a second localization is performed within this region to obtain the third updated coordinates. This process is repeated, with each iteration using more precise coordinates as the center point, cropping a smaller image region, and performing more precise localization, until the obtained facial landmark coordinates meet the preset accuracy requirements.

[0058] Finally, based on the precise positioning coordinates of each facial feature point, the system calculates the relative positional relationships between these feature points, such as the relative positions and distances of key feature points such as the eyes, nose, and mouth. Using these relative positional relationships, the system uses a preset algorithm or model to accurately determine the driver's facial orientation. This process may involve comparing a preset facial orientation template or using a machine learning algorithm to identify different orientation patterns, thereby ensuring that the system can accurately and quickly identify the driver's facial orientation. This application does not impose any restrictions on the specific calculation steps.

[0059] In a specific embodiment, in order to accurately determine the driver's facial orientation, a facial recognition algorithm is first used to identify feature points in the driver's facial images continuously captured by the vehicle's internal camera. These feature points cover key parts such as the eyes, nose, mouth, and eyebrows, and the precise coordinates of each identified feature point in the image are recorded as the basis for subsequent calculation of relative position relationships.

[0060] Next, the relative positional relationships of the feature points are calculated. First, one or more reference points are selected that are relatively stable when the facial orientation changes, such as the center of the eyes or the apex of the nose. Then, for each other feature point, its coordinates relative to the reference points are calculated. This typically involves simple geometric operations, such as calculating the horizontal and vertical distances between two points. Furthermore, to eliminate the effects of varying driver facial sizes and camera distances on the calculation results, these relative coordinates can be normalized by dividing the coordinate values ​​by a reference value (such as facial width or height) to obtain a relative ratio.

[0061] After obtaining the relative position relationship of the feature points, the facial orientation is determined. To this end, one or more facial orientation models are pre-established, which describe the relative position relationship of facial feature points under different orientations. For example, when looking straight ahead, the relative positions of the eyes and nose will show a specific proportional relationship. The calculated relative position relationship of the feature points is compared with these pre-built orientation models, and the best matching orientation model is found by calculating the similarity or distance metric (such as Euclidean distance, cosine similarity, etc.). Finally, the best matching orientation model is selected as the driver's facial orientation. If the calculated relative position relationship of the feature points is closest to the model looking straight ahead, the system will determine that the driver's facial orientation is looking straight ahead.

[0062] Finally, to improve the accuracy of the judgment, iterative optimization can be performed, repeatedly performing feature point recognition, relative position relationship calculation, and facial orientation judgment, adjusting parameters or optimizing the algorithm based on the previous results. Furthermore, other information sources can be integrated, such as the driver's body posture and head rotation angle, which can be obtained from other sensors or cameras inside the vehicle.

[0063] In one specific implementation of this application, DCNN facial feature points are used to predict the facial orientation of a person. The facial position and subsequent posture are described based on the five facial feature points (corners of the eyes, nose, and mouth). A multi-stage cascade method is usually used to gradually approximate the actual feature point positions from coarse to fine. The following are the three specific stages of DCNN prediction of the five facial feature points, as well as its basic process and calculation formula:

[0064] The first stage: coarsely locate facial feature points. Use a convolutional neural network (such as the F1 network) to extract features from the entire face image. The network outputs a feature vector containing the predicted coordinates of the five feature points, as shown below:

[0065] Output = f(Input) Formula (1)

[0066] Among them, f() represents the forward propagation process of the network, Input is the input face image, and Output is the feature point coordinate vector output by the network.

[0067] During the entire process from getting out of the car to walking to the front of the car to open the front hood, the facial features collected by the camera placed at the windshield are from the initial side face (that is, only one side of the eye corner, mouth corner and nose, a total of 3 feature points) to the face completely facing the windshield (5 feature points); therefore, the timing starts after the driver's door opening signal is recognized. The timing time determines the approximate time required for the user to walk to the front of the car based on the user's walking parameters such as stride length and gait; during this time, the number of facial feature points is continuously collected and input into the image to calculate the coordinate vector.

[0068] In the second stage, the facial feature points are initially located. Based on the output of the first stage, a small area of ​​the image containing the feature points is cropped with the predicted feature points as the center. Multiple CNNs (such as EN1 and NM1 networks) are used to initially locate the feature points in these small area images.

[0069] New_Coordinates=CNN(Cropped_Image) Formula (2)

[0070] Where CNN represents a specific convolutional neural network, Cropped_Image is the cropped image area, and New_Coordinates is the updated feature point coordinates.

[0071] In the third stage, facial landmarks are precisely located. Based on the output of the second stage, a smaller image region is cropped. In the CNN for fine-grained localization, a series of computational steps, including convolution, activation, and pooling, are repeated in each convolutional layer until the network outputs the final landmark coordinates.

[0072] The convolution operation is as follows:

[0073] (I*K)[x,y]=∑m∑nI[xm.yn]·k[m,n] Formula (3)

[0074] Where I is the input image, K is the convolution kernel, * indicates that convolution is done, [x, y] is the coordinate of the output feature map, m is the number of steps the convolution kernel slides in the horizontal direction, and n is the number of steps the convolution kernel slides in the vertical direction.

[0075] The activation function is specifically such as ReLU:

[0076] ReLU(z)=max(0,z) Formula (4)

[0077] Here, z is an input value in the neural network.

[0078] It can be understood that in the simplest case, z can be a single input value in the neural network, such as the sum of the weighted inputs of a neuron; or in other common cases, z can be a vector or matrix, in which case the ReLU function is applied element by element, for example, for each element zi in the vector or matrix, the ReLU function calculates max(0,zi) and combines all the results into a new vector or matrix.

[0079] Pooling operations such as max pooling:

[0080] P[x, y] = max m，n (I[xm,yn]) Formula (5)

[0081] Among them, P is the feature map after pooling, I is the feature map, x is the horizontal coordinate of the target position of the pooling operation, y is the vertical coordinate of the target position of the pooling operation, and m and n are the window sizes that define the pooling operation.

[0082] It can be understood that in a convolutional neural network, the input feature map (Input Feature Map) will be processed by the convolution layer to obtain one or more feature maps, and I refers to one of these feature maps. In addition, m and n are the window sizes that define the pooling operation. In maximum pooling, m and n usually take the same value, indicating the size of the pooling window in the horizontal and vertical directions. For example, if a 2x2 pooling window is used, then m = 2 and n = 2.

[0083] In one embodiment of the present application, based on the change of the angle formed by the body feature points, it is detected whether the target person has a bending action, including: obtaining the coordinates of each body feature point, the body feature points including at least a waist feature point, a chest feature point, and a leg feature point; taking the waist feature point as the vertex, calculating the angle formed between the chest feature point and the leg feature point; if the angle is less than a preset angle threshold, it is determined that the target person has a bending action.

[0084] In a specific embodiment of the present application, after detecting that the user is facing the windshield, the user's distance from the front of the vehicle is detected in real time; when the distance between the user and the front of the vehicle reaches a threshold, OpenPose is used to detect whether the user has bent over: the human skeleton key point (and body feature point) data is extracted from the image information collected above. This data includes the coordinates of various parts of the body, such as the head, shoulders, elbows, wrists, hips, knees, etc., and a function is defined to calculate the angle formed by three key points (for example, chest, waist, knees), and the function is specifically as follows:

[0085]

[0086] Among them, θ represents the final angle formed, A and C represent one of the key points (and one of the chest feature point and knee feature point), and B represents the vertex forming the angle (and the waist feature point).

[0087] In addition, a bending threshold is set. For example, when the waist angle (the angle between the chest and knees) is less than 150 degrees, it can be considered that the person is bending. By comparing the calculated angle with the threshold, it is determined whether bending has occurred. If the angle is less than the threshold, bending is detected.

[0088] In one embodiment of the present application, the gesture of the target person is identified based on the hand feature points, and the behavioral intention of the target person is determined in combination with the orientation, bending state and gesture of the target person, including: when the target person is facing the hood of the vehicle, and the target person is in a bending state, and the gesture of the target person matches at least one preset gesture, it is determined that the target person has the intention to open the hood of the vehicle, and the preset gesture includes at least one of the following: knocking on the hood, a preset specific hood opening gesture, and the hand touching the hood sensor at the front of the vehicle.

[0089] In one embodiment of the present application, a high-definition camera accurately captures an image of the target person's hand. A deep learning algorithm is then used to meticulously extract the hand's various feature points, including key information such as the precise position of the fingertips, the outline of the palm, and the degree of bend in the finger joints. Simultaneously, the target person's orientation and bending state are continuously monitored. By analyzing the relative positional relationship between facial and body feature points, as well as monitoring changes in the angles between body feature points such as the shoulders and waist, the target person's body posture is accurately determined.

[0090] When entering the gesture recognition phase, the system has already stored a series of preset gesture models closely related to opening the vehicle's hood. These models cover the simulated action of tapping the hood, specific hood opening gestures (such as simulating the action of pulling the door or pressing the switch), and the behavior of a hand touching the hood sensor at the front of the vehicle. When the target person's hand feature points match any of these preset gesture models, it will be preliminarily determined that the target person may be performing an action related to opening the hood.

[0091] Finally, during the behavioral intention determination phase, the system comprehensively considers the target person's orientation, bent position, and recognized gestures. If the target person is facing the vehicle's hood, bent over, and their hand features match at least one pre-defined gesture model, the system determines that the target person has a clear intention to open the hood. Once this intention is determined, the system immediately triggers a corresponding response mechanism, such as rapidly unlocking the hood, thereby accurately identifying and promptly responding to the target person's behavioral intention.

[0092] In one specific embodiment, as a target person approaches a vehicle, the system captures their facial and torso features through a camera, determining that they are leaning toward the vehicle's hood and hunched over. The system then observes that the target person's gesture matches a pre-defined "specific hood opening gesture," simulating the action of pulling a door or pressing a hood switch. Based on this combined information, the system quickly determines that the target person intends to open the hood, immediately unlocks the hood, and simultaneously sends a notification to the user, notifying them that the hood is ready.

[0093] Step S240: When the target person's behavioral intention is to open the vehicle's hood, the vehicle's hood is controlled to be unlocked.

[0094] In one embodiment of the present application, before controlling the unlocking of the vehicle's hood, it also includes: pre-logging into the vehicle owner's account and entering the authorization information, where the authorization information includes at least one of facial information and image information; based on the authorization information, authenticating the target person who has the behavioral intention to open the vehicle's hood; if the verification is successful, sending a request to unlock the hood to the body domain controller to trigger the body domain controller to identify the operating status of the vehicle's hood; if the current hood operating status is an openable state, executing the unlocking operation of the hood.

[0095] In one embodiment of the present application, the vehicle owner must first log in to their account through the vehicle's central control system or mobile phone app to ensure the legitimacy of subsequent operations. After successfully logging in, the vehicle owner can pre-enter the authorized person's information under the account, which includes at least facial information and / or image information. The system will save this information and use it for subsequent identity verification.

[0096] When the vehicle is powered on, the SRR radar controller and camera module monitor the vehicle's surroundings in real time to ensure safety. The camera module uses advanced facial recognition technology to match the faces of people approaching the vehicle and quickly determine whether they are authorized. Furthermore, the module uses gesture recognition technology to carefully analyze the target person's behavior and accurately determine whether they have a clear intention to open the hood, such as bending over or reaching out.

[0097] Once the target person's facial information perfectly matches the stored authorized user information and their behavior matches the pre-set hood opening pattern, the system deems authentication successful. At this point, the camera module immediately sends a command to the vehicle's vehicle interface unit (VIU) to unlock the hood. Upon receiving this command, the VIU immediately performs a series of security checks, including critical information such as the vehicle's power mode, gear position, and anti-theft status, to ensure the vehicle is operational and safe. The VIU then meticulously identifies the hood's operational status to confirm its readiness for opening. Once the VIU confirms that everything is in place and the hood is indeed ready to open, it immediately unlocks the hood, allowing the authorized user to easily open it for subsequent operation or maintenance. The entire process is efficient, intelligent, secure, and reliable, fully demonstrating the advanced and convenient nature of modern vehicle intelligent management systems.

[0098] In one specific embodiment, assume that the vehicle owner has pre-entered their facial information and certain specific behavioral patterns into the system. One day, the vehicle owner approaches the vehicle and prepares to open the hood. The system captures the vehicle owner's face through the camera module and matches it with the pre-stored information. The system also recognizes that the vehicle owner is bending over and reaching toward the hood, which aligns with the pre-set hood-opening behavior. The system then quickly completes authentication, and the camera module sends an unlock request to the VIU. After confirming that the vehicle is safe and the hood can be opened, the VIU immediately executes the unlock operation, allowing the vehicle owner to successfully open the hood.

[0099] It can be understood that the method proposed in the above embodiment not only ensures that only authorized personnel can open the front hatch, but also improves the safety and convenience of operation through real-time environment detection and behavior recognition.

[0100] Figure 4 This is a block diagram of a vehicle front hood unlocking device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1 The device may also be applicable to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0101] like Figure 4 As shown, the exemplary vehicle hood unlocking device includes: a data acquisition module 410 , a demand prediction module 420 , a behavior intention recognition module 430 , and an unlocking control module 440 .

[0102] Among them, the data acquisition module 410 is used to obtain vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle. The vehicle maintenance data includes at least one of the maintenance data of preset maintenance items and vehicle fluid maintenance data, and the image information includes at least one human portrait; the demand prediction module 420 is used to predict the target person's demand for opening the front hood based on the vehicle maintenance data and vehicle abnormality data; the behavior intention recognition module 430 is used to identify the image information when the prediction result shows that the target person has a potential demand to open the front hood, so as to determine the target person's behavior intention based on the recognition result; the unlocking control module 440 is used to control the unlocking of the vehicle front hood when the target person's behavior intention is to open the vehicle front hood.

[0103] In a specific embodiment of the present application, the vehicle hood unlocking device can be further divided into a first module, a second module, and a third module. The first module is a command acquisition module, which is used to acquire hood control commands input by the user inside the vehicle through the human-computer interaction module; the second module is a radar module, which is used to detect whether there are dynamic commands from the user outside the vehicle; and the third module is a camera module, which is used to predict static gestures and behaviors of the user outside the vehicle, match them with pre-stored behaviors, and control the hood unlocking command based on the recognized behaviors.

[0104] In one specific implementation, the radar module (radar controller SRR) operates continuously. The vehicle's front radar module detects objects within a preset area extending outward from the front of the vehicle to detect whether a user has made a dynamic action in front of the vehicle within a specified time period. If the user's dynamic action matches any of the pre-stored dynamic action databases, the corresponding hood control command is output. When the radar module SSR detects a user's behavior outside the vehicle, it calls the camera module (vehicle camera module MRC) to capture and identify the user and action. After the camera module detects the user's facial recognition outside the vehicle and passes authorized facial recognition, the camera module collects and compares static actions captured outside the vehicle within a preset time period in real time to see if they match the preset actions. If so, the camera module extracts features from the captured static user actions outside the vehicle and compares them with the actions in the pre-stored action database to determine if the captured action matches the hood unlocking action. The pre-stored gesture database contains multiple data points corresponding to the same gesture for people of different genders, ages, weights, and heights.

[0105] It should be noted that the vehicle hood unlocking device provided in the above-mentioned embodiment and the vehicle hood unlocking method provided in the above-mentioned embodiment are based on the same concept. The specific manner in which the various modules and units perform their operations has been described in detail in the method embodiments and will not be repeated here. In actual applications, the vehicle hood unlocking device provided in the above-mentioned embodiment can, as needed, allocate the aforementioned functions to different functional modules, i.e., divide the internal structure of the device into different functional modules to perform all or part of the aforementioned functions, and this is not a limitation herein.

[0106] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle hood unlocking method provided in the above-mentioned embodiments.

[0107] Figure 5 The following is a schematic diagram showing the structure of a computer system suitable for implementing an electronic device according to an embodiment of the present application. Figure 5 The computer system 500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0108] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503, such as executing the method described in the above embodiment. Various programs and data required for system operation are also stored in the RAM 503. The CPU 501, ROM 502 and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0109] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, and the like; an output section 507 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 508 including a hard disk; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. Removable media 511, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 510 as needed, so that computer programs read therefrom can be installed into the storage section 508 as needed.

[0110] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 509, and / or installed from a removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the various functions defined in the system of the present application are executed.

[0111] It should be noted that the computer-readable medium shown in the embodiments of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable computer program. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. A computer program embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0112] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. Among them, each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0113] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. In some cases, the names of these units do not constitute limitations on the units themselves.

[0114] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon. When executed by a computer processor, the computer program causes the computer to perform the vehicle hood unlocking method described above. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist independently and not be incorporated into the electronic device.

[0115] Another aspect of the present application provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the vehicle hood unlocking method provided in each of the above embodiments.

[0116] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, any equivalent modifications or alterations accomplished by a person of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A method for unlocking a vehicle hood, characterized in that: The method comprises: Acquiring vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, wherein the vehicle maintenance data includes at least one of maintenance data of preset maintenance items and vehicle fluid maintenance data, and the image information includes portrait information of at least one target person; Predicting a target person's need to open the hood based on the vehicle maintenance data and the vehicle abnormality data includes obtaining the current time, the current amount of vehicle fluid remaining, and a door signal; if the current time is a maintenance time point, or there is a need for fluid replenishment, or the target person has an intention to perform maintenance, then determining that the target person has a potential need to open the hood; If the prediction result shows that the target person has a potential need to open the front hatch, the image information is recognized to determine the target person's behavioral intention based on the recognition result; When the target person's behavioral intention is to open the vehicle's front hood, the vehicle's front hood is controlled to unlock.

2. The vehicle front hood unlocking method according to claim 1, characterized in that: Predicting a hood opening demand of a target person based on the vehicle maintenance data and the vehicle abnormality data includes: Comparing the current time with the maintenance data of the preset maintenance item to determine whether the current time is at a maintenance time point; Based on the vehicle fluid maintenance data, predicting a fluid replenishment demand state under the current vehicle fluid remaining state; The abnormal data is used to determine whether the vehicle has had an accident or malfunction within a preset time, and the target personnel's willingness to perform maintenance is evaluated in combination with the door signal.

3. The vehicle front hood unlocking method according to claim 2, characterized in that: Identify the portrait information to determine the target person's behavioral intention based on the identification result, including: Extracting facial feature points, body feature points, and hand feature points from the portrait information of the target person; Identifying the relative positional relationship between facial feature points to determine the orientation of the target person; When the target person is facing the front hood of the vehicle, detecting whether the target person is bending over based on the change in the angle formed by the body feature points; If a bending action is detected, the target person's gesture is identified based on the hand feature points, and the target person's behavioral intention is determined in combination with the target person's orientation, bending state and gesture.

4. The vehicle front hood unlocking method according to claim 3, characterized in that: Before identifying the relative position relationship between facial feature points and determining the orientation of the target person, the following steps are also included: Get the initial coordinates of each facial feature point; With the initial coordinates as the center, a first cropped image region including all the initial coordinates is cropped, and each facial feature point is preliminarily located within the first cropped image region to obtain the first updated coordinates; Using the initially updated coordinates as the new center point, the cropped image region containing all updated coordinates is cropped again, and each facial feature point is relocated within the cropped image region to obtain the updated coordinates. Repeat the above cropping and positioning process until the obtained facial feature point coordinates meet the preset accuracy conditions.

5. The vehicle front hood unlocking method according to claim 3, characterized in that: Detecting whether the target person is bending over based on the change in the angle formed by the body feature points includes: Obtaining coordinates of each body feature point, wherein the body feature points include at least a waist feature point, a chest feature point, and a leg feature point; Taking the waist feature point as a vertex, calculating the angle formed between the chest feature point and the leg feature point; If the angle is smaller than a preset angle threshold, it is determined that the target person is bending over.

6. The vehicle front hood unlocking method according to claim 3, characterized in that: The target person's gesture is identified based on the hand feature points, and the target person's behavioral intention is determined based on the target person's orientation, bending state, and gesture, including: When the target person is facing the hood of the vehicle and is in a bent-over state, and the target person's gesture matches at least one preset gesture, it is determined that the target person has the intention to open the hood of the vehicle. The preset gesture includes at least one of the following: knocking on the hood, a preset specific hood opening gesture, and hand touching the hood sensor at the front of the vehicle.

7. The vehicle front hood unlocking method according to any one of claims 1 to 6, characterized in that: Before controlling the vehicle's front hood unlocking, it also includes: Log in to the vehicle owner's account in advance and enter the authorizer information, wherein the authorizer information includes at least one of facial information and image information; Based on the authorized person information, identity verification is performed on the target person who has the behavioral intention to open the front hood of the vehicle; If the verification is successful, a request to unlock the front hood is sent to the body domain controller to trigger the body domain controller to identify the vehicle's front hood operation status; When the front hatch operation state is the openable state, the front hatch unlocking operation is performed.

8. A vehicle front hood unlocking device, characterized in that: The device comprises: a data acquisition module, configured to acquire vehicle maintenance data, vehicle abnormality data, and image information outside the vehicle, wherein the vehicle maintenance data includes at least one of maintenance data of preset maintenance items and vehicle fluid maintenance data, and the image information includes at least one human image; a demand prediction module for predicting a target person's need to open the front hood based on the vehicle maintenance data and the vehicle abnormality data, including obtaining a current time, a current remaining amount of vehicle fluid, and a door signal; and determining that the target person has a potential need to open the front hood if the current time is a maintenance time point, or there is a need for fluid replenishment, or the target person has an intention to perform maintenance; a behavioral intention recognition module, configured to recognize the image information when the prediction result shows that the target person has a potential need to open the front hatch, so as to determine the behavioral intention of the target person based on the recognition result; The unlocking control module is used to control the unlocking of the vehicle's front hood when the target person's behavioral intention is to open the vehicle's front hood.

9. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute the computer program stored in the memory to implement the vehicle front hood unlocking method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and the computer program is used to enable a computer to execute the vehicle front hood unlocking method according to any one of claims 1 to 7.

Images