Control method and device of vehicle tailgate, storage medium, electronic device

Abstract

This application discloses a control method and device for a vehicle tailgate, a storage medium, an electronic device, and a computer program product, relating to the field of vehicle control. The control method for the vehicle tailgate includes: acquiring a palm image of a target object; performing correction processing on the palm image to adjust it to a standard posture; using a pre-trained detection model to detect the palm image to obtain a palmprint image; extracting feature information from the palmprint image; and controlling the opening of the vehicle tailgate by correcting the posture of the palmprint image to ensure that the palmprint image can reach a standardized posture, thereby improving the accuracy of subsequent palmprint recognition and comparison, and improving the accuracy of vehicle tailgate control.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and more specifically, to a method and apparatus for controlling a vehicle tailgate, a storage medium, an electronic device, and a computer program product. Background Technology

[0002] With the rapid development of the automotive industry, people have increasingly higher demands for the intelligence of vehicles. Automatic tailgate opening has become a standard feature in many high-end models, designed to provide convenience and enhance the user experience.

[0003] In related technologies, automatic tailgates are usually opened by kicking, which means that a kick sensor or proximity sensor is designed on the tailgate to enable automatic unlocking and opening.

[0004] However, the solutions in the relevant technologies require users to perform a specific action below the tailgate to trigger its opening, which is not convenient enough and is prone to misidentifying the actions of other unrelated creatures behind the vehicle, causing the tailgate to open unexpectedly and be easily triggered. Summary of the Invention

[0005] This application provides a method and apparatus for controlling a vehicle tailgate, a storage medium, and an electronic device.

[0006] According to one aspect of the embodiments of this application, a method for controlling a vehicle tailgate is provided. The method includes: acquiring a palm image of a target object; performing correction processing on the palm image to adjust the palm image to a standard posture; detecting the palm image using a pre-trained detection model to obtain a palmprint image; extracting feature information from the palmprint image; and controlling the opening of the vehicle tailgate when, based on the feature information, it is determined that the matching degree between the palmprint image and palmprints in a preset palmprint database meets the standard.

[0007] In one embodiment of this application, the step of correcting the palm image to adjust it to a standard pose includes: determining multiple key points in the palm image; establishing a first coordinate system for the palm image based on the multiple key points; determining the transformation relationship between the first coordinate system and the standard coordinate system; and adjusting the angle of the palm image according to the transformation relationship to transform the palm image to the standard coordinate system, wherein the pose of the palm image in the standard coordinate system is the standard pose.

[0008] In one embodiment of this application, before detecting the palm image using a pre-trained detection model to obtain a palmprint image, the method further includes: acquiring a training sample set, wherein the training sample set includes palm images of multiple people in various poses; labeling the training sample set to mark the palmprint region and multiple key points of each palm image in the training sample set; and training a preset model using the training sample set to obtain the pre-trained detection model.

[0009] In one embodiment of this application, before acquiring the palm image of the target object, the method further includes: when a target key is detected, controlling the vehicle to power on and unlock, and activating the tailgate module; detecting the location information of the target object; and when it is determined that the target object is within a preset range of the vehicle tailgate, controlling the vehicle's vision module to start in order to search for the palm image of the target object.

[0010] In one embodiment of this application, the method for controlling the tailgate of a vehicle further includes: continuously searching for a palm image of a target object when the tailgate of the vehicle is in an open state; and controlling the tailgate of the vehicle to close when a palm image of the target object is searched and acquired, and the palm print image corresponding to the palm image matches the palm print in a preset palm print database.

[0011] In one embodiment of this application, the method for controlling the tailgate of a vehicle further includes: pre-collecting multiple palm images of different poses to be recorded; performing correction processing on the multiple palm images of different poses to adjust them to a standard pose; using a pre-trained detection model to locate the palmprint region in the corrected palm images, and then using a pre-trained feature extraction model to extract features from the cropped palmprint image to determine the feature information of the multiple palmprint images to be recorded; storing the multiple palmprint images to be recorded and the feature information of the multiple palmprint images to be recorded to obtain the preset palmprint library.

[0012] In one embodiment of this application, extracting feature information from a palmprint image includes: determining the center point of the corrected palmprint image; and extracting image information of a region within a preset range of the center point of the palmprint image as feature information.

[0013] According to another aspect of the embodiments of this application, a control device for a vehicle tailgate is also provided, comprising: an acquisition module for acquiring a palm image of a target object; a correction module for correcting the palm image to adjust it to a standard posture; a detection module for detecting the palm image using a pre-trained detection model to obtain a palmprint image; an extraction module for extracting feature information from the palmprint image; and a control module for controlling the opening of the vehicle tailgate when the matching degree between the palmprint image and a preset palmprint database is determined to be satisfactory based on the feature information.

[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, and the computer program is configured to execute the above-described vehicle tailgate control method when running.

[0015] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described vehicle tailgate control method through the computer program.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of the methods described in the various embodiments of this application. The above-described vehicle tailgate control method first acquires an image of the target object's palm. Then, the palm image is corrected to adjust it to a standard posture, thereby ensuring that the palm image achieves a standardized posture and improving the accuracy of subsequent palmprint recognition and comparison. Then, a pre-trained detection model is used to detect the palm image, obtaining a palmprint image, and extracting feature information from the palmprint image. If, based on the feature information, the matching degree between the palmprint image and the palmprints in a preset palmprint database meets the standard, the vehicle tailgate is controlled to open. By using the method of this application, correcting the palm image before detection improves the accuracy of vehicle tailgate control. Furthermore, using image recognition for vehicle tailgate opening control eliminates the need for additional sensors, saving costs. Since palmprint features are complex and difficult to replicate, using palmprint recognition for tailgate control also improves the security of tailgate control. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a hardware structure block diagram of the vehicle terminal of the vehicle tailgate control method according to an embodiment of this application;

[0020] Figure 2 This is a flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0021] Figure 3 This is a second flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0022] Figure 4 This is a third flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0023] Figure 5 This is a flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0024] Figure 6 This is the fifth flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0025] Figure 7 This is a flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0026] Figure 8 This is the seventh flowchart of a vehicle tailgate control method according to an embodiment of this application;

[0027] Figure 9 This is a structural block diagram of an optional vehicle tailgate control device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data used in this way can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] The methods and embodiments provided in this application can be executed in an in-vehicle terminal or a similar computing device. Taking running on an in-vehicle terminal as an example, Figure 1 This is a hardware structure block diagram of the computer terminal for the vehicle tailgate control method according to an embodiment of this application. (See diagram below.) Figure 1 As shown, a computer terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor (MPU) or a programmable logic device (PLD)) and a memory 104 for storing data are also shown. In one exemplary embodiment, the vehicle-mounted terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned vehicle-mounted terminal. For example, the vehicle-mounted terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 Equivalent functions or ratios shown Figure 1 The functions shown have more different configurations.

[0031] The memory 104 can be used to store computer programs and a palmprint database, such as application software programs and modules, like the computer program corresponding to the vehicle tailgate control method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the aforementioned method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the vehicle terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0032] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the vehicle terminal. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0033] This embodiment provides a method for controlling a vehicle tailgate, applied to an in-vehicle terminal. Figure 2 This is a flowchart of an optional vehicle tailgate control method according to an embodiment of this application, the process including the following steps S200-S240:

[0034] Step S200: Acquire an image of the target object's palm.

[0035] Specifically, a camera module installed in the vehicle, such as a fisheye camera in an Around View Monitor (AVM) system, can be used to capture an image of the target object's palm. This camera module can be integrated into the vehicle's tailgate. The target object refers to anyone near the vehicle's tailgate when the vehicle is powered on; it is not necessarily the vehicle owner, and further palmprint recognition is required to determine this.

[0036] For example, the target can spread their palm with fingers outstretched, the flat side facing the fisheye camera, and place it 10-15cm away from the camera, perpendicular to the camera's optical axis. This ensures clear palm prints while reducing distortion. When the target moves in this way, it indicates that the target intends to open the tailgate.

[0037] Step S210: Correct the palm image to adjust it to a standard pose.

[0038] Specifically, since the user's hand posture is uncontrollable, the original captured hand image may not be in a standard pose, which will affect subsequent feature extraction and palmprint matching. Therefore, correction processing is needed to adjust the hand in the image to a standard pose that is easy to recognize. The standard pose can be a preset pose that is easy to recognize.

[0039] Step S220: Use a pre-trained detection model to detect the palm image to obtain a palm print image.

[0040] The pre-trained detection model can be a deep learning model (object detection model) such as the YOLO series, SSD, Faster R-CNN, etc. These models have been trained on a large amount of labeled data and can accurately detect and locate key points in palm images. The acquired palm image is input into this detection model, which can then identify the palmprint region and obtain the palmprint image. In other words, in this application, the initial input is a palm image, and the palmprint image is obtained only after key point detection and pose correction steps.

[0041] Step S230: Extract feature information from the palm print image.

[0042] Specifically, based on the palmprint image after posture correction, feature information that can be used for identity verification is extracted. For example, palmprint features, such as creases and lines, are biometric identifiers of an individual's identity.

[0043] Feature extraction can be performed using a pre-trained feature extraction model, such as a model with MobileFaceNet as the backbone (i.e., the feature extraction network) and MagFace / ArcFace / AdaFace as the loss function. The training process can be as follows: 1) Obtain a training sample set, which includes palmprint images of multiple people in various poses, categorized by person and left / right hand. 2) Collect at least five palm images of each person and each hand in different poses and environments, annotate them to mark the palmprint region, and then uniformly adjust them to the same size, such as 224*224 pixels. 3) Categorize the annotated images by person and left / right hand, dividing them into training and validation sets in a 4:1 ratio. For example, if a sample set of 200 people is annotated, 160 people are used as the training set and 40 people as the validation set. 4) Train the pre-set recognition network model using the training and validation sets. Once the required accuracy is achieved, use the feature extraction network as the pre-trained feature extraction model.

[0044] By employing a feature extraction model, an image of a region within a preset range of the center point of the palm print can be extracted from a palm image after pose correction. This image is then resized to the same size, for example, 224*224 pixels. The palm print feature vector can then be output using the feature extraction model.

[0045] Step S240: If the matching degree between the palm print image and the palm print in the preset palm print database meets the standard based on the feature information, the tailgate of the vehicle is controlled to open.

[0046] Specifically, the extracted feature information can be compared with the feature information of all preset palm prints stored in the palm print database. The degree of matching can be calculated using methods such as Euclidean distance or cosine similarity, and can be quantified as a matching score. If the matching score exceeds a preset threshold, it means that the identity authentication of the target object is successful, thereby controlling the opening of the tailgate.

[0047] In this embodiment, a palm image of the target object is first acquired. Then, the palm image is corrected to adjust it to a standard posture. This posture correction ensures the palm image achieves a standardized posture, thereby improving the accuracy of subsequent palmprint recognition and comparison. A pre-trained detection model is then used to detect the palm image, obtaining a palmprint image, and extracting its feature information. If, based on the feature information, the palmprint image matches a pre-set palmprint database, the vehicle's tailgate is opened. This approach, by correcting the palm image before detection, improves the accuracy of tailgate control. Furthermore, using image recognition for tailgate control eliminates the need for additional sensors, saving costs. Since palmprint features are complex and difficult to replicate, using palmprint recognition for tailgate control also enhances security.

[0048] In one embodiment, such as Figure 3 As shown, step S210 involves correcting the hand image to adjust it to a standard pose, including steps S300-S330. Wherein:

[0049] Step S300: Determine multiple key points in the palm image.

[0050] Among them, the key points in the palm image may include, but are not limited to, the gaps between the thumb and index finger, the index and middle finger, the middle and ring finger, the ring and little finger, and the center point of the palm.

[0051] For example, in palm recognition, the finger gaps and various creases in the palm are key features due to their high uniqueness and stability. For instance, the detection of the finger gap P1 between the thumb and index finger, P2 between the index and middle finger, P3 between the middle and ring finger, and P4 between the ring and little finger is crucial. After detecting these four key points, the regions corresponding to the point pairs (P2, P3) and (P3, P4) can be defined as the finger gap target category. These regions contain important finger gap information, which helps improve the accuracy of palm print region localization. The region of interest (ROI) of the entire palm print image is defined as the target category for palm print extraction. This region contains creases and other palm print details due to their high uniqueness and stability. Therefore, the two target categories of the detection model are defined as: the finger gap target category and the palm print target category.

[0052] Step S310: Establish the first coordinate system of the palm image based on multiple key points in the palm image.

[0053] Specifically, after determining multiple key points in the palm image, a first coordinate system can be established based on the positions of these key points. This first coordinate system can represent the current pose of the palm image. For example, by performing mathematical operations on the key points, such as calculating the center point and the line connecting point pairs (P2, P3) and (P3, P4), coordinate axes and an origin can be defined. The line connecting the center points of point pairs (P2, P3) and (P3, P4) is used as part of the X-axis, and the Y-axis is constructed with the center point as the origin, thus establishing the first coordinate system.

[0054] For example, the center point (A, B) of each double-finger slit target frame can be calculated, and the center point C0 of the initial palm print target frame is (x c ,y c Due to detection model errors, C0 may not be accurate; therefore, subsequent palmprint image correction processing is performed. Then, the straight line l passing through point AB... AB As the X-axis, define positive and negative directions; for example, with the palm pointing vertically upwards, the positive X-axis direction is to the right. Then calculate the center point O of A and B = (O x O y Given the length ||AB|| and the modulus, we define point O as the origin. Next, we will... (The sentence is incomplete and requires more context to translate accurately.) AB straight line l C The Y-axis is defined with positive and negative directions. The positive Y-axis direction is defined by rotating 90° counterclockwise from the positive X-axis direction; for example, upwards is the positive Y-axis direction. Based on the spatial distribution characteristics of palm prints, the center point of the palm print lies on the Y-axis. The establishment of this first coordinate system provides a basis for subsequent palm print image correction and feature point localization.

[0055] Step S320: Determine the transformation relationship between the first coordinate system and the standard coordinate system.

[0056] The standard coordinate system is a reference coordinate system defined for an ideal pose of a hand image. Transforming the first coordinate system to the standard coordinate system can eliminate inconsistencies in the original image caused by different poses or angles. For example, the standard coordinate system is the world coordinate system, which is constructed based on the palm pointing vertically upwards.

[0057] Specifically, by calculating the rotation parameters between the first coordinate system and the standard coordinate system, a transformation relationship is determined, which contains the coordinate transformation information required to map a point in the first coordinate system to the standard coordinate system.

[0058] For example, calculate the line that passes through point C0 and is perpendicular to line l. AB The foot of the perpendicular is T, and the unit vector is... It is the direction vector along the -Y axis. When constructing the first coordinate system, C0 is the initial center point of the palm print, and the straight line l AB This is the X-axis. To determine the positive direction of the Y-axis, we first need to calculate the distance from point C0 to line l. AB The foot of the perpendicular is T, which determines the perpendicular direction of the Y-axis. After calculating T, the vector... The vector pointing from T to C0, with its unit vector (i.e., the normalized vector), represents the direction of the -Y axis. This is because, based on the characteristics of palmprint spatial distribution, the center point C of the palmprint should be located on the Y-axis in a normal posture. The vector direction from C0 to T helps determine the Y-axis of the first coordinate system. Then, through rotation correction, the Y-axis of the first coordinate system is made to coincide with the Y-axis of the standard coordinate system, thus aligning the center point of the palmprint with the Y-axis of the standard coordinate system. Next, the rotation angle θ from the local coordinate system to the world coordinate system is calculated. The rotation angle θ needs to satisfy the following condition:

[0059]

[0060] The calculation of the rotation angle θ depends on the unit vector. The direction is (x, y). When x > 0 and y < 0, x represents the cosine of the angle with the positive direction of the standard coordinate system's x-axis. This means that a counter-clockwise rotation of a certain angle is needed to convert the negative direction of the Y-axis to the positive direction of the Y-axis in the standard coordinate system. When x < 0 and y < 0, θ = π / 2 + arccos(-x), indicating that a clockwise rotation of a certain angle is needed to achieve the same purpose. When x > 0 and y ≥ 0, θ = arccos(x) - π / 2, meaning that the counter-clockwise rotation angle is different to accommodate different hand postures. For other cases (i.e., x < 0 and y ≥ 0 or x = 0, y = 0, etc.), θ = π / 2 - arccos(-x), which is also to ensure that the center point of the palm print is accurately aligned with the Y-axis in the standard coordinate system regardless of the initial posture.

[0061] Step S330: Adjust the angle of the palm image according to the transformation relationship, and transform the palm image to the standard coordinate system.

[0062] The pose of the hand image in the standard coordinate system is the standard pose.

[0063] Specifically, after calculating the rotation angle θ, a pose correction operation is performed on the palmprint image, that is, the palmprint image is rotated according to the calculated rotation angle θ to ensure that the center point C of the palmprint is accurately located on the Y-axis of the standard coordinate system. This correction process corrects the position of the center point of the palmprint, thereby ensuring the correct extraction and matching of palmprint features.

[0064] In this embodiment, the establishment of the first coordinate system provides the necessary spatial positioning reference for the posture correction of the palm image, enabling the palm image to be uniformly represented and processed under different postures, thus enhancing the robustness of the recognition algorithm. The determination of the transformation relationship ensures that regardless of the palm's posture during image acquisition, the transformed palm image will conform to the standard posture, ensuring the correct extraction and matching of palmprint features, eliminating recognition errors caused by changes in the user's palm posture, and ensuring consistent palmprint features even under different shooting conditions, thereby improving the accuracy of palmprint recognition.

[0065] In one embodiment, such as Figure 4 As shown, before obtaining the palm print image by using a pre-trained detection model in step S300, the vehicle tailgate control method further includes steps S400-S430. Wherein:

[0066] Step S400: Obtain the training sample set.

[0067] The training sample set includes hand images of multiple people in various poses.

[0068] Specifically, fisheye cameras can be used to collect images of different users' hands from various angles, under different lighting conditions, and in different background environments. Typically, hundreds or thousands of images are needed to cover all possible hand poses. These images can include different hand poses such as palm up, down, tilted, and rotated, as well as hand features from different groups of people. These images serve as a training sample set.

[0069] Step S410: The training sample set is labeled to mark the palm print area and multiple key points of each palm image in the training sample set.

[0070] Specifically, the collected training sample set is labeled, that is, the key points in each palm image are identified and marked. Image annotation tools can be used by professionals or through automated programs to accurately mark the location of key points in each image, such as the gaps between the thumb and index finger, index and middle finger, middle and ring finger, ring and little finger, and the center point of the palm print. Then, the marked key points are converted into corresponding bounding boxes, such as the double-finger gap bounding box and the ROI region bounding box of the palm image.

[0071] Step S420: Train the preset model using the training sample set to obtain the pre-trained detection model.

[0072] Specifically, a pre-defined deep learning model is trained using labeled training and validation sets, enabling it to identify and locate key points in palm images and convert the labeled key points into corresponding bounding boxes, such as the double-finger gap bounding box and the ROI region bounding box of the palm image.

[0073] In this embodiment, a diverse training sample set enables the model to maintain high recognition rates and robustness across various real-world usage scenarios, reducing recognition errors caused by pose changes. The annotation of key points provides the model with a learning basis, helping it accurately grasp palmprint features and improving training efficiency and recognition accuracy. The trained detection model can automatically identify and locate key points in input palm images, maintaining a high recognition rate even under complex backgrounds and pose variations. This forms the foundation for palmprint recognition and pose correction, ensuring the accuracy and efficiency of subsequent palmprint feature extraction and matching.

[0074] In one embodiment, such as Figure 5 As shown, before acquiring the palm image of the target object in step S200, the vehicle tailgate control method further includes steps S500-S520. Wherein:

[0075] In step S500, if the target key is detected, the vehicle is powered on and unlocked, and the tailgate module is activated and available.

[0076] Specifically, when the vehicle power is OFF, if a user approaches the vehicle with a digital key (including but not limited to Ultra Wide Band (UWB) digital keys, Bluetooth keys, etc.), the vehicle will detect the digital key (target key), authenticate and pair it with the vehicle, and if the pairing is successful, the vehicle will be powered on, the system will control the vehicle to power on and unlock, and the tailgate module will be activated and available.

[0077] Step S510: Detect the location information of the target object.

[0078] Specifically, after the vehicle is powered on, it can locate the target object to determine whether the user is within the preset range of the tailgate. This can be achieved by using UWB, Bluetooth channel detection and positioning, or other high-precision positioning technologies. The vehicle can track the signal of the digital key in real time and calculate the relative position and distance between the target object and the vehicle's tailgate.

[0079] Step S520: If it is determined that the target object is within a preset range of the vehicle's tailgate, the vehicle's vision module is activated to search for the target object's palm image.

[0080] Specifically, once the vehicle's system confirms the location information of the target object and determines that the location is indeed within the preset range of the tailgate (e.g., a certain distance near the tailgate, such as 100cm to 150cm), the vehicle's computing unit module will send an instruction to the vehicle's vision module (such as the high-definition fisheye camera in the AVM surround view system) to start the image stream and prepare to capture the palm image.

[0081] In this embodiment, by energizing the vehicle's tailgate upon detecting a target key, the vision module is ensured to activate immediately when the user approaches the tailgate, improving the smoothness and immediacy of the user experience. By detecting the location information of the target object, it can determine whether the user is ready to open the tailgate. By combining digital key detection and the target object's location information, the system can intelligently determine when to activate the vision module, avoiding unnecessary power consumption and resource waste when no one is approaching the tailgate. Simultaneously, this step also ensures user privacy, as the vision module only activates when the user clearly approaches the tailgate and is ready to use it, reducing unnecessary image capture.

[0082] In one embodiment, such as Figure 6 As shown, the control method for the vehicle tailgate also includes steps S600-S610.

[0083] in:

[0084] In step S600, while the tailgate of the vehicle is open, the search continues for the palm image of the target object.

[0085] Specifically, when the tailgate is opened, the vehicle's system continuously activates the vision module to search for images of the target object's hand. This can be achieved by automatically keeping the AVM (Advanced Visual Monitoring System) high-definition fisheye camera (or other vision sensors) active after the tailgate is opened, continuously capturing images to locate the target object's hand. During this process, the computing unit module continuously runs a hand recognition algorithm, identifying and locating the hand from the continuous image stream.

[0086] Specifically, since the AVM's rear-view fisheye camera is located on the trunk, when the tailgate is open, the rear-view fisheye camera flips up with the trunk, changing the field of view of the rear-view fisheye camera. At this time, the left / right fisheye camera can be used to search for the palm image of the target object.

[0087] Step S610: If the palm image of the target object is searched and acquired, and the matching degree between the palm print image corresponding to the palm image and the palm print in the preset palm print library meets the standard, control the tailgate of the vehicle to close.

[0088] Specifically, if both conditions are met—that the tailgate is open and that the identity of the target is successfully verified based on palm print recognition—it is assumed that the user intends to close the tailgate, and the vehicle's tailgate is controlled to close.

[0089] In this embodiment, continuously searching for palm images ensures that the system can immediately respond to the user's intention to close the tailgate, improving the immediacy and convenience of the user experience. Palmprint recognition can serve as both user authentication and determination of the intention to open and close the tailgate, achieving two goals at once and ensuring the security of tailgate opening and closing.

[0090] In one embodiment, such as Figure 7 As shown, the control method for the vehicle tailgate also includes steps S700-S730.

[0091] in:

[0092] Step S700: Collect multiple images of the hand to be recorded in different poses in advance.

[0093] Specifically, it is necessary to collect images of the hand to be recorded beforehand (e.g., images of the car owner's hand). These images should include pictures of the hand in different postures to ensure that the system can adapt to various hand postures in actual use. Under the guidance of the system, the user (car owner) places their hand in the field of view of a high-definition fisheye camera at different angles and postures, and takes multiple images. To obtain high-quality palmprint data, the user may need to place their hand under specific lighting conditions to ensure that the palmprint features are clearly visible.

[0094] Step S710: Correct multiple images of the hand to be recorded in different poses are processed to adjust the multiple images of the hand to be recorded in different poses to the standard pose.

[0095] Specifically, the correction method has been described in the above embodiments and will not be repeated here.

[0096] Step S720: A pre-trained detection model is used to locate the palm print region in multiple palm print images to be recorded. Then, a pre-trained feature extraction model is used to extract features from the cropped palm print images to determine the feature information of the multiple palm print images to be recorded.

[0097] Specifically, the collected palm images are input into a pre-trained detection model. This model can locate and detect key points (such as finger gaps) in the palm images and construct a first coordinate system for the image based on these points. Then, pose correction is performed. Finally, the corrected palm print image is extracted and input into a pre-trained feature extraction model to obtain the palm print feature information, such as wrinkles and lines. The feature information is usually represented as a feature vector, which is a mathematical description that can uniquely identify each palm print image.

[0098] Step S730: Store multiple palm print images to be recorded and their feature information to obtain a preset palm print database.

[0099] Specifically, each palmprint image to be entered and its corresponding feature vector are stored in a database, which is called the palmprint database. To improve system efficiency, only the feature vector can be stored instead of the entire image.

[0100] In this embodiment, collecting diverse palmprint images helps improve the accuracy and robustness of the system's recognition. The feature vectors extracted through the deep learning model possess high discriminative power and stability, maintaining a high matching degree even under different lighting, angles, or poses, greatly improving the accuracy and efficiency of palmprint recognition. Creating a palmprint database provides a data foundation for subsequent palmprint recognition. By storing feature information from multiple images, the system can more comprehensively identify users; even if a user's palmprint changes slightly due to environmental factors, the system can still find the corresponding user through feature matching, further enhancing the system's stability and security.

[0101] In one embodiment, such as Figure 8 As shown, step S230, extracting feature information from the palmprint image, includes steps S800-S810. Wherein:

[0102] Step S800: Determine the center point of the corrected palmprint image.

[0103] Specifically, after pose correction is completed on the palmprint image, a reference point needs to be determined, namely the center point of the palmprint image. This is the basis for subsequent extraction of the core feature regions of the palmprint. Determining the center point of the palmprint image helps to focus on the key parts of the palmprint, namely the areas with dense lines and rich information, thereby improving the accuracy and efficiency of feature extraction.

[0104] Step S810: Extract image information of the region within a preset range of the center point of the palm print image as feature information.

[0105] Specifically, the system defines a preset region based on the center point information, such as a circular region centered on the palm print's center point, or a square region. The side length and position of this region are set based on statistical analysis or experience to ensure coverage of the most recognizable part of the palm print. Then, the image information of this region is cropped from the corrected palm print image.

[0106] For example, based on the spatial distribution characteristics of palm prints and empirical information, a square ROI region can be extracted with the palm print center point C as the center. The side length of the square can be... AB can be the distance between two key points on the palm image. Select... The reason is that, through experiments and experience, it has been found that, based on the finger gap points, setting the side length of the ROI to 2.5 times the finger gap point distance can cover a sufficient palm print feature area, while avoiding unnecessary background interference, thus improving the efficiency and accuracy of feature extraction.

[0107] In this embodiment, extracting the core feature region can significantly improve the accuracy and reliability of palmprint recognition because this region contains the richest ridge information, which can focus on the core feature region, that is, the region that more accurately reflects an individual's biometric characteristics. At the same time, it also reduces the amount of computation and speeds up the matching and verification process.

[0108] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0109] This embodiment also provides a vehicle tailgate control device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0110] Figure 9 This is a structural block diagram of an optional vehicle tailgate control device according to an embodiment of this application. Figure 9 As shown, it includes:

[0111] Acquisition module 901 is used to acquire images of the palm of the target object;

[0112] The correction module 902 is used to correct the palm image to adjust the palm image to a standard posture.

[0113] The detection module 903 is used to detect palm images using a pre-trained detection model to obtain palm print images;

[0114] Extraction module 904 is used to extract feature information from palmprint images;

[0115] The control module 905 is used to control the opening of the vehicle's tailgate when the palm print image meets the matching degree standard with the palm print in the preset palm print library based on the feature information.

[0116] The aforementioned device first acquires an image of the target object's palm. Then, the palm image is corrected to a standard posture, ensuring a standardized hand posture and improving the accuracy of subsequent palmprint recognition and comparison. A pre-trained detection model is then used to detect the palm image, obtaining a palmprint image and extracting its feature information. If the palmprint image matches a pre-set palmprint database based on the feature information, the vehicle's tailgate is opened. This approach, by correcting the palm image before detection, improves the accuracy of tailgate control. Furthermore, using image recognition for tailgate control eliminates the need for additional sensors, saving costs. Since palmprint features are complex and difficult to replicate, using palmprint recognition for tailgate control also enhances security.

[0117] In one exemplary embodiment, the correction module 902 includes:

[0118] The recognition unit is used to determine multiple key points in a palm image.

[0119] The coordinate system establishment unit is used to establish the first coordinate system of the palm image based on multiple key points in the palm image.

[0120] The transformation relationship determination unit is used to determine the transformation relationship between the first coordinate system and the standard coordinate system.

[0121] The transformation unit is used to adjust the angle of the palm image according to the transformation relationship and transform the palm image to the standard coordinate system, wherein the pose of the palm image in the standard coordinate system is the standard pose.

[0122] In one exemplary embodiment, the above-described apparatus further includes:

[0123] The sample acquisition module is used to acquire a training sample set, which includes hand images of multiple people in various poses.

[0124] The annotation module is used to annotate the training sample set, marking the palm print region and multiple key points of each palm image in the training sample set.

[0125] The training module is used to train a pre-set model using a training sample set to obtain a pre-trained detection model.

[0126] In one exemplary embodiment, the above-described apparatus further includes:

[0127] The power-on control module is used to control the vehicle's power-on and unlocking when a target key is detected, and the tailgate module is activated and available.

[0128] The location detection module is used to detect the location information of the target object.

[0129] The activation control module is used to activate the vehicle's vision module to search for the target object's palm image when the target object is determined to be within a preset range of the vehicle's tailgate.

[0130] In one exemplary embodiment, the above-described apparatus further includes:

[0131] The search module is used to continuously search for images of the hand of a target object when the tailgate of the vehicle is open.

[0132] The shutdown control module is used to close the vehicle's tailgate when a palm image of the target object is searched and acquired, and the palm print image corresponding to the palm image matches the palm print in the preset palm print library.

[0133] In an exemplary embodiment, the above-described apparatus is further configured to: pre-collect multiple palm images of different poses to be recorded; perform correction processing on the multiple palm images of different poses to adjust them to a standard pose; use a pre-trained detection model to locate the palm print region in the corrected palm images; then use a pre-trained feature extraction model to extract features from the cropped palm print image to determine the feature information of the multiple palm print images to be recorded; and store the multiple palm print images to be recorded and their feature information to obtain a preset palm print database.

[0134] In one exemplary embodiment, the extraction module 903 includes:

[0135] The center determination unit is used to determine the center point of the corrected palmprint image.

[0136] The extraction unit is used to extract image information of a region within a preset range of the center point of the palm print image as feature information.

[0137] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.

[0138] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:

[0139] S1, acquire the palm image of the target object.

[0140] S2, corrects the palm image to adjust it to a standard pose.

[0141] S3 uses a pre-trained detection model to detect the palm image and obtain the palm print image.

[0142] S4, extract feature information from the palm print image.

[0143] S5, if the palmprint image matches the palmprint in the preset palmprint database according to the feature information, then control the tailgate of the vehicle to open.

[0144] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0145] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0146] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0147] S1, acquire the palm image of the target object.

[0148] S2, corrects the palm image to adjust it to a standard pose.

[0149] S3 uses a pre-trained detection model to detect the palm image and obtain the palm print image.

[0150] S4, extract feature information from the palm print image.

[0151] S5, if the palmprint image matches the palmprint in the preset palmprint database according to the feature information, then control the tailgate of the vehicle to open.

[0152] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0153] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium storing the computer program product, wherein the computer program, when executed by a processor, implements the steps of the methods in various embodiments of this application.

[0154] Optionally, in this embodiment, the computer program described above can be configured to perform the following steps when executed by a processor:

[0155] S1, acquire the palm image of the target object.

[0156] S2, corrects the palm image to adjust it to a standard pose.

[0157] S3 uses a pre-trained detection model to detect the palm image and obtain the palm print image.

[0158] S4, extract feature information from the palm print image.

[0159] S5, if the palmprint image matches the palmprint in the preset palmprint database according to the feature information, then control the tailgate of the vehicle to open.

[0160] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0161] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0162] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for controlling a vehicle tailgate, characterized in that, The method includes: Acquire an image of the target object's palm; The hand image is corrected to adjust it to a standard pose; A pre-trained detection model is used to detect the palm image to obtain a palmprint image; Extract the feature information from the palmprint image; If, based on the aforementioned feature information, the palmprint image is determined to match the palmprint in a preset palmprint database to a certain degree, the tailgate of the vehicle is opened.

2. The vehicle tailgate control method according to claim 1, characterized in that, The step of correcting the hand image to adjust it to a standard pose includes: Identify multiple key points in the palm image; Based on multiple key points in the palm image, a first coordinate system for the palm image is established; Determine the transformation relationship between the first coordinate system and the standard coordinate system; According to the transformation relationship, the angle of the palm image is adjusted to transform the palm image into the standard coordinate system, wherein the posture of the palm image in the standard coordinate system is the standard posture.

3. The vehicle tailgate control method according to claim 2, characterized in that, Before using a pre-trained detection model to detect the palm image and obtain a palmprint image, the method further includes: acquiring a training sample set, wherein the training sample set includes palm images of multiple people in various poses; The training sample set is labeled to mark the palm print region and multiple key points of each palm image in the training sample set. The pre-trained detection model is obtained by training the pre-set model using the training sample set.

4. The vehicle tailgate control method according to claim 1, characterized in that, Before acquiring the palm image of the target object, the method further includes: Upon detecting the target key, the vehicle is powered on and unlocked, and the tailgate module is activated and available. Detect the position information of the target object; If the target object is determined to be within a preset range of the vehicle's tailgate, the vehicle's vision module is activated to search for an image of the target object's hand.

5. The vehicle tailgate control method according to any one of claims 1-4, characterized in that, The method further includes: With the tailgate of the vehicle open, the search continues for the palm image of the target object; If a palm image of the target object is searched and acquired, and the matching degree between the palm print image corresponding to the palm image and the palm print in the preset palm print library meets the standard, the tailgate of the vehicle is controlled to close.

6. The vehicle tailgate control method according to any one of claims 1-4, characterized in that, The method further includes: Multiple images of the hand to be recorded in different poses were collected in advance; The multiple images of the hand to be recorded in different poses are corrected to adjust them to a standard pose. A pre-trained detection model is used to locate the palm print region in multiple palm print images to be recorded after correction. Then, a pre-trained feature extraction model is used to extract features from the cropped palm print images to determine the feature information of multiple palm print images to be recorded. Multiple palmprint images to be entered and their feature information are stored to obtain the preset palmprint library.

7. The vehicle tailgate control method according to any one of claims 1-4, characterized in that, The extraction of feature information from the palmprint image includes: Determine the center point of the corrected palmprint image; The image information of the region within a preset range of the center point of the palm print image is extracted as the feature information.

8. A control device for a vehicle tailgate, characterized in that, include: The acquisition module is used to acquire images of the target object's palm. The correction module is used to correct the palm image to adjust it to a standard pose. The detection module is used to detect the palm image using a pre-trained detection model to obtain a palmprint image; The extraction module is used to extract feature information from the palmprint image; The control module is used to control the opening of the vehicle's tailgate when the matching degree between the palmprint image and the palmprint in the preset palmprint database meets the standard based on the feature information.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-7.

Images