Vehicle identification methods, devices and electronic equipment

Abstract

This invention relates to the field of vehicle technology and discloses a vehicle recognition method, device, and electronic device. The method includes: after entering a vehicle recognition mode, responding to a target selection command issued by a user, controlling the vehicle's onboard camera to acquire image information of the target vehicle indicated by the target selection command; identifying vehicle information of the target vehicle based on the image information; and outputting the vehicle information of the target vehicle. Applying the technical solution of this invention can solve the problem of low efficiency in acquiring vehicle information in the prior art.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, specifically to a vehicle identification method, device, and electronic device. Background Technology

[0002] As the automotive industry develops towards greater intelligence, electrification, and connectivity, vehicle brands and models are becoming increasingly diverse, leading to a growing demand from users for real-time access to vehicle information, such as the brand or model of the vehicle in front of them, while driving.

[0003] Currently, vehicle information is mostly searched through mobile devices such as smartphones, but this process is inefficient and requires users to be distracted, posing a driving safety hazard. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention provide a vehicle identification method, device and electronic device to solve the problem of low efficiency in acquiring vehicle information in the prior art.

[0005] According to one aspect of the present invention, a vehicle identification method is provided, the method being applied to a controller in a vehicle, the method comprising:

[0006] After entering vehicle recognition mode, in response to the target selection command issued by the user, the vehicle's on-board camera is controlled to collect image information of the target vehicle indicated by the target selection command.

[0007] Based on the image information, identify the vehicle information of the target vehicle;

[0008] Output the vehicle information of the target vehicle.

[0009] According to another aspect of the present invention, a vehicle identification device is provided, the device being applied to a controller in a vehicle, comprising:

[0010] The control module is used to control the vehicle's onboard camera to acquire image information of the target vehicle indicated by the target selection command after entering the vehicle recognition mode, in response to the target selection command issued by the user.

[0011] The recognition module is used to recognize the vehicle information of the target vehicle based on the image information;

[0012] The output module is used to output the vehicle information of the target vehicle.

[0013] According to another aspect of the present invention, an electronic device is provided, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0014] The memory is used to store at least one executable instruction that causes the processor to perform the operation of the vehicle identification method described above.

[0015] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction that causes an electronic device / vehicle identification device to perform the following operations:

[0016] After entering vehicle recognition mode, in response to the target selection command issued by the user, the vehicle's on-board camera is controlled to collect image information of the target vehicle indicated by the target selection command.

[0017] Based on the image information, identify the vehicle information of the target vehicle;

[0018] Output the vehicle information of the target vehicle.

[0019] In this embodiment of the invention, after entering vehicle recognition mode, the controller controls the onboard camera to acquire image information of the target vehicle in response to the user's target selection command. Based on the image information, the controller then identifies the vehicle information of the target vehicle and finally outputs the vehicle information to the user. This eliminates the need for the user to manually search for vehicle information using external devices such as mobile phones. The entire process from image acquisition to information recognition and output is automatically completed by the onboard device, simplifying the operation steps and improving the efficiency of vehicle information acquisition. At the same time, it eliminates the need for the user to manually input or search, reducing interference with driving safety and ensuring driving safety.

[0020] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0021] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0022] Figure 1 A schematic diagram of the eye data acquisition prompt provided by the present invention is shown;

[0023] Figure 2 A flowchart of a first embodiment of the vehicle identification method provided by the present invention is shown;

[0024] Figure 3 A flowchart of a second embodiment of the vehicle identification method provided by the present invention is shown;

[0025] Figure 4 A schematic diagram of an application scenario provided by the present invention is shown;

[0026] Figure 5 A schematic diagram of an embodiment of the vehicle identification device provided by the present invention is shown;

[0027] Figure 6 A schematic diagram of an embodiment of the electronic device provided by the present invention is shown. Detailed Implementation

[0028] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0029] In related technologies, relying on mobile terminals such as smartphones to search for vehicle information is not only cumbersome and inefficient, but more importantly, it requires users to manually operate the mobile terminal while driving, which can easily lead to distraction and potential driving safety hazards. Based on this, the inventors further considered building an automated recognition system based on the vehicle's own equipment. This system would use user-issued target selection commands (such as voice, eye focus, or gestures) as activation signals, and leverage the vehicle's built-in hardware—the in-vehicle camera—to automatically and accurately capture image information of the target vehicle upon receiving the command. The in-vehicle controller would then automatically recognize the image information to extract vehicle information. The entire process requires no additional user intervention, simplifying the information acquisition process, improving efficiency, and avoiding the safety risks associated with manual operation while driving.

[0030] The execution subject of this invention can be a controller in a vehicle. The controller can be an existing controller in the vehicle, such as a cockpit domain controller; or it can be a separately set controller. The controller can be an electronic control unit (ECU), a microcontroller unit (MCU), etc. This invention does not limit the scope of the invention.

[0031] The user-related data involved in this application, such as user eye data and user behavior data, were all obtained with the user's permission or consent. In other words, when this application is used in a specific product or technology, user permission is required to obtain and process the relevant data, and the processing of the relevant data must comply with the relevant laws, regulations and regulatory standards of the relevant countries and regions.

[0032] For example, Figure 1 A schematic diagram of the eye data acquisition prompt provided by the present invention is shown, such as... Figure 1As shown, when it is necessary to obtain the user's eye data, an eye data acquisition prompt can be displayed on the vehicle's screen. After receiving confirmation from the user regarding the eye data acquisition prompt, the controller can acquire the user's eye data.

[0033] Figure 2 A flowchart of a first embodiment of the vehicle identification method provided by the present invention is shown. Figure 2 As shown, the method includes the following steps:

[0034] Step 110: After entering vehicle recognition mode, in response to the target selection command issued by the user, control the vehicle's on-board camera to collect image information of the target vehicle indicated by the target selection command.

[0035] For example, vehicle recognition mode refers to the working state in which the controller enables vehicle recognition-related functions. In this mode, the controller responds to target selection commands and allocates relevant hardware resources such as the vehicle camera and interaction module. Vehicle recognition mode can be triggered by user actions such as clicking the vehicle recognition icon on the central control screen or speaking a preset wake-up word.

[0036] A target selection command is a command issued by a user to a specified vehicle to be identified, used to indicate the direction of the target vehicle. The target selection command can be any of the following forms: voice command, gaze focus command, or gesture command.

[0037] Vehicle-mounted cameras refer to image acquisition devices integrated into a vehicle. These may include front-view main cameras, left / right-side auxiliary cameras, and rear-view cameras, used to collect image data of target vehicles outside the vehicle. It is understandable that different cameras may correspond to different shooting coverage areas or shooting angles.

[0038] The image information of the target vehicle refers to image data containing the vehicle features of the vehicle to be identified, such as vehicle logos, grilles, body lines, headlights, etc. It should be noted that this application embodiment does not limit the resolution or format type of the image information of the target vehicle.

[0039] In one example, after the controller detects the user's activation of the vehicle recognition mode, it enters the vehicle recognition mode and starts monitoring the target selection command. When the target selection command issued by the user is detected, the controller first parses the target selection command. For example, voice commands are denoised and semantically recognized, gaze focusing commands are analyzed by the eye tracking module to determine the gaze direction, and gesture commands are identified by the visual sensor to determine the gesture direction. Then, the controller determines the location of the target vehicle indicated by the target selection command, determines the shooting angle of the vehicle camera, and then calls the vehicle camera in the corresponding location. According to the shooting angle, the camera is controlled to collect multiple frames of image information of the target vehicle at a preset frame rate.

[0040] Step 120: Identify the vehicle information of the target vehicle based on the image information.

[0041] For example, vehicle information is data used to characterize the attributes of the target vehicle, such as vehicle brand, model, year, manufacturer's suggested retail price, vehicle configuration, user reviews, sales data, etc. Recognition refers to the process by which the controller processes image information using a preset recognition model, extracts vehicle features, and matches them with corresponding vehicle attribute data.

[0042] It should be noted that the recognition process can be completed independently based on the pre-configured local model on the vehicle, or it can be completed in combination with the vehicle-cloud collaborative mode. In other words, the collected image information can be sent to the cloud device, and the recognition model configured on the cloud device can process the image information to obtain vehicle information. Alternatively, the controller on the vehicle can first extract features from the image information, obtain image features, and then send them to the cloud device, where the cloud device can process the image features to obtain vehicle information.

[0043] In one example, the controller first preprocesses the image information to obtain preprocessed image information. This preprocessed image information is then input into a preset recognition model, which outputs the brand and model information of the target vehicle. This information is then matched against a preset vehicle database to obtain the vehicle information corresponding to the brand and model information. Preprocessing may include, for example, Gaussian denoising, cropping, resizing, and pixel value normalization, etc., which are not limited in this embodiment. The vehicle database includes vehicle attribute information corresponding to various brand and model information. The recognition model may be, for example, a neural network model, and the type of recognition model is not limited in this embodiment.

[0044] Step 130: Output the vehicle information of the target vehicle.

[0045] For example, output refers to the process by which the controller presents the identified vehicle information to the user through the in-vehicle interaction module. The output method may be, for example, voice broadcast, display on the vehicle's screen, or display on a head-up display device. This embodiment of the invention does not limit the output method.

[0046] In one example, the controller can send a vehicle information broadcast command to the in-vehicle voice interaction module, controlling the voice module to broadcast vehicle information to the user; alternatively, the controller can send a vehicle information display command to the in-vehicle display screen, presenting the vehicle information in a visual form on the display screen.

[0047] In this embodiment, after entering vehicle recognition mode, the controller controls the onboard camera to acquire image information of the target vehicle in response to the user's target selection command. Based on this image information, the controller then identifies the vehicle information and finally outputs the vehicle information to the user. This method eliminates the need for users to manually search for vehicle information using external devices such as mobile phones. The entire process from image acquisition to information recognition and output is automatically completed by the onboard device, simplifying the operation and improving the efficiency of vehicle information acquisition. Simultaneously, it eliminates the need for users to manually input or search, reducing interference with driving safety and ensuring driving safety.

[0048] Figure 3 A flowchart of a second embodiment of the vehicle recognition method provided by the present invention is shown. Figure 4 A schematic diagram of an application scenario provided by the present invention is shown, such as... Figure 3 and Figure 4 As shown, the method includes the following steps:

[0049] Step 210: After entering vehicle recognition mode, in response to the target selection command, determine the shooting angle information of the vehicle camera.

[0050] For example, the shooting perspective information refers to the information of the camera operating parameters used to locate the target vehicle. For example, it may include the camera to be activated and shooting parameters, wherein the shooting parameters may include, for example, focal length mode, shooting angle, focus range, etc.

[0051] In one example, the controller detects that the user clicks the "Vehicle Recognition" icon on the central control screen or says a preset wake-up word (such as "Hello XX, turn on vehicle recognition"), confirms entry into vehicle recognition mode, and starts target selection command monitoring. When the target selection command issued by the user is detected, the controller performs the corresponding parsing operation according to the command type, matches the parsed location information with the coverage of each camera, determines the corresponding camera to be activated, focal length mode, shooting angle, focus range and other parameters, and obtains the shooting perspective information.

[0052] Specifically, when the target selection command is a voice command, the voice recognition process is performed on the target selection command to determine the location information of the target vehicle carried in the target selection command; based on the location information of the target vehicle, the shooting angle information of the vehicle camera is determined.

[0053] The voice command refers to the target selection instruction issued by the user to the controller through natural language pronunciation. This voice command can be an audio signal collected by the vehicle's microphone array. Voice recognition processing refers to the controller's process of noise reduction, feature extraction, and semantic parsing of the collected audio signal to convert it into understandable text information and extract key content. The voice command may include location keywords for the target vehicle, such as "front," "left," or "right rear." Optionally, the voice command may also include characteristic information of the target vehicle, such as "black sedan" or "red SUV." The target vehicle's location information refers to the information included in the voice command for locating the target vehicle, such as "front," "left," or "right rear." Optionally, the target vehicle's location information may also include relative distance information, such as keywords like "far away" or "nearby."

[0054] For example, after entering vehicle recognition mode, the controller activates the monitoring function of the in-vehicle microphone array. Upon acquiring the audio signal of the user's voice command, it first performs environmental noise reduction processing, filtering out interference signals such as engine noise, wind noise, and in-vehicle conversations using an adaptive filtering algorithm to obtain the voice feature signal. Then, the controller calls a preset voice recognition model to convert the voice feature signal into text information and performs semantic analysis on the text information to extract directional keywords, thus obtaining the location information of the target vehicle. Optionally, if the text information does not contain directional keywords, the controller can control the in-vehicle voice interaction module to issue a follow-up question, such as "Where is the black sedan located?", to obtain the user's additional location information for the target vehicle.

[0055] Table 1 illustrates a mapping table between vehicle-mounted cameras and orientations provided in an embodiment of the present invention. After determining the orientation information of the target vehicle, the controller can call the pre-stored mapping table between vehicle-mounted cameras and orientations as shown in Table 1. This mapping table pre-stores the orientation information of each camera and its coverage orientation range. The controller matches the extracted orientation information of the target vehicle with the coverage orientation range of the cameras in the mapping table to determine the camera to be activated. For example, if the orientation information of the target vehicle is "front", the front-view main camera is matched; if the orientation information of the target vehicle is "left", the left-view auxiliary camera is matched. If the orientation information of the target vehicle does not include relative distance information, preset shooting parameters such as shooting angle and focus range are used to integrate and obtain the shooting angle information. If the orientation information of the target vehicle includes relative distance information, the shooting parameters corresponding to the relative distance information can be determined based on the mapping table between vehicle-mounted cameras and orientations to integrate and obtain the shooting angle information.

[0056] Table 1 Mapping table between vehicle-mounted cameras and orientation

[0057]

[0058] When the target selection command is a gaze focus command, the system responds to the target selection command by acquiring the user's eye features; based on the eye features, it determines the user's gaze direction information; and based on the gaze direction information, it determines the shooting angle information of the vehicle camera.

[0059] The gaze-focusing command refers to the user transmitting a target selection instruction to the controller by continuously looking at a target vehicle outside the vehicle. Eye features are physiological characteristic data that characterize the user's gaze state, such as the pupil center position and eye rotation angle. The in-vehicle gaze tracking module is a hardware unit used to collect eye features, typically including an infrared binocular camera mounted in front of the steering wheel or above the dashboard. Gaze direction information refers to a three-dimensional spatial vector obtained by converting the user's eye features; for example, it may include the horizontal deflection angle, vertical deflection angle, and gaze extension distance, and can be mapped to a specific spatial area outside the vehicle. It should be noted that this embodiment of the invention does not limit whether the user wears glasses.

[0060] For example, after the controller enters the vehicle recognition mode, it automatically activates the in-vehicle gaze tracking module and starts the eye feature acquisition function. In response to the gaze focusing command, it captures the user's eye area image in real time, and extracts the two-dimensional coordinates of the pupil center in each frame of the eye area image using a pupil center localization algorithm. At the same time, by comparing the relative positional changes of the pupil center and the eyeball outline, it calculates the horizontal deflection angle (negative for leftward deflection and positive for rightward deflection) and vertical deflection angle (positive for upward deflection and negative for downward deflection) of the eyeball, and integrates them to obtain the user's eye features. The controller can pre-store a transformation matrix to convert the horizontal / vertical deflection angle of the eyeball in the eye coordinate system into the horizontal and vertical deflection angles of the gaze in the in-vehicle coordinate system. At the same time, the controller presets the correspondence between vehicle speed and gaze extension distance, and can then combine the current vehicle speed to determine the gaze extension distance and integrate the gaze direction information.

[0061] After determining the line-of-sight direction information, the controller can determine the spatial location information of the target vehicle focused on by the line of sight. For example, the line-of-sight extension distance is used as the longitudinal coordinate, the product of the line-of-sight extension distance and the tangent of the horizontal deflection angle in the line-of-sight direction information is used as the lateral coordinate, and the product of the line-of-sight extension distance and the tangent of the vertical deflection angle in the line-of-sight direction information is used as the vertical coordinate. Then, based on the preset mapping table between cameras and coverage areas in the controller, the camera to be activated can be determined, and the shooting parameters can be determined based on the line-of-sight extension distance to obtain the shooting angle information. For example, if the line-of-sight extension distance is greater than a preset distance threshold, it is considered to be far away in Table 1, and the shooting parameters shown in Table 1 can be used.

[0062] When the target selection command is a gesture command, in response to the target selection command, the system acquires the user's valid pointing gesture within the preset gesture recognition area; based on the valid pointing gesture, it determines the fingertip pointing information; and based on the fingertip pointing information, it determines the shooting angle information of the vehicle camera.

[0063] Gesture commands refer to the user's target selection command, conveying the desired vehicle location to the controller through hand pointing gestures. The preset gesture recognition area is a pre-defined three-dimensional space within the vehicle used to limit the detection range of hand movements. For example, it could be the space from above the steering wheel to in front of the central control screen. This area covers the range of hand movements during natural driving. It should be noted that this embodiment of the invention does not limit the range of the preset gesture recognition area. A valid pointing gesture refers to a hand movement that meets preset judgment criteria. For example, it could be a single-finger pointing posture (index finger extended, other four fingers bent and together), excluding invalid movements such as clenching a fist, waving, or shaking the hand. Fingerpoint pointing information is a set of core parameters extracted by the controller based on valid pointing gestures for locating the target vehicle. This could include, for example, the horizontal deflection angle of the fingertip, the vertical deflection angle of the fingertip, and the pointing distance of the fingertip, and could be mapped to a certain spatial area outside the vehicle.

[0064] For example, after the controller enters the vehicle recognition mode, it automatically activates the vehicle-mounted binocular vision sensor installed above the steering wheel. In response to gesture commands, it captures hand images in real time within the preset gesture recognition area. The captured hand images are processed sequentially by grayscale noise reduction, hand contour extraction, and key point detection. The two-dimensional coordinates of key points of the hand (finger tip, finger base, wrist) are identified through a preset convolutional neural network model. The hand posture is determined to be a single-finger pointing posture by determining that the index finger is straight and the angle between it and the other four fingers is ≥60°. If it is, it is determined to be a valid pointing gesture. If it does not meet the requirements, the action is ignored and monitoring continues.

[0065] Furthermore, after confirming that it is a valid pointing gesture, the controller can convert the two-dimensional coordinates of the index fingertip into coordinate information under the vehicle's in-vehicle coordinate system. It then calculates the horizontal deflection angle (the angle between the fingertip's X-axis coordinate and the origin's X-axis coordinate, with leftward deflection being negative and rightward deflection being positive) and the vertical deflection angle (the angle between the fingertip's Z-axis coordinate and the origin's Z-axis coordinate, with upward deflection being positive and downward deflection being negative) of the fingertip by using the coordinate difference. At the same time, the controller has a preset correspondence between vehicle speed and fingertip pointing distance, which can be combined with the vehicle's current speed to determine the fingertip pointing distance and integrate the fingertip pointing information.

[0066] After determining the fingertip pointing information, the controller can determine the spatial location information of the target vehicle pointed to by the fingertip. For example, the fingertip pointing distance is the longitudinal coordinate, the product of the fingertip pointing distance and the tangent of the horizontal deflection angle in the fingertip pointing information is the lateral coordinate, and the product of the fingertip pointing distance and the tangent of the vertical deflection angle in the fingertip pointing information is the vertical coordinate. Then, based on the preset mapping table between cameras and coverage areas in the controller, the camera to be activated can be determined, and the shooting parameters can be determined based on the fingertip pointing distance to obtain the shooting angle information. For example, if the fingertip pointing distance is greater than a preset distance threshold, it is considered to be far away in Table 1, and the shooting parameters shown in Table 1 can be used.

[0067] By using multimodal target selection commands consisting of voice, eye focus, and gestures to determine the shooting angle information of the vehicle camera, it can fully adapt to the user's contactless interaction needs in driving scenarios. It not only covers the operating habits and preferences of different users, but also reduces the risk of distraction caused by users using external devices such as mobile phones or manually operating the vehicle terminal. It ensures driving safety from the source of interaction and further optimizes the convenience and experience of users obtaining target vehicle information during driving.

[0068] Step 220: Based on the shooting angle information, control the vehicle-mounted camera to acquire image information of the target vehicle.

[0069] For example, the controller sends an activation command and a parameter configuration command to the corresponding vehicle camera based on the determined shooting angle information. After the parameter configuration is completed, the controller controls the camera to continuously acquire multiple frames of image information at a preset rate according to the set shooting parameters such as focal length mode, shooting angle, and focus range.

[0070] In some possible implementations, based on the shooting angle information, the vehicle-mounted camera is controlled to capture images and display a preview image on the vehicle's display screen; wherein, the target vehicle is selected in the preview image; if no adjustment operation is received from the user regarding the target vehicle, the vehicle-mounted camera is controlled to acquire image information of the target vehicle based on the shooting angle information; if an adjustment operation is received from the user regarding the target vehicle, a new shooting angle is determined based on the new target vehicle position information indicated by the adjustment operation, and the vehicle-mounted camera is controlled to acquire image information of the new target vehicle based on the new shooting angle information.

[0071] In this context, the preview screen refers to the real-time dynamic view of the vehicle exterior captured by the camera based on the shooting angle information. Target vehicle selection refers to the controller marking the predicted target vehicle area in the preview screen with a highlighted dynamic border using a real-time target detection algorithm. It should be noted that this embodiment of the invention does not limit the border format; the border color and thickness only need to be clearly distinguishable from the background. The vehicle's display screen can be an in-vehicle central control display screen or other display screens; this embodiment of the invention does not impose any limitations. The adjustment operation refers to the correction command issued by the user when they have objections to the target vehicle selected in the preview screen.

[0072] For example, the controller can send activation and parameter configuration commands to the matched vehicle camera based on the shooting angle information, controlling the camera to start real-time shooting according to the set focal length mode, shooting angle, and focus range. Simultaneously, the dynamic images captured by the camera are transmitted to the display screen via the vehicle bus. The controller can call a preset lightweight target detection algorithm to perform frame-level processing on the real-time preview image, identify areas in the image that match vehicle characteristics, and use a highlighted dynamic border to select the predicted target vehicle. If multiple suspected vehicles exist in the image, the controller prioritizes selecting the target vehicle that best matches the predicted location based on the spatial coordinates corresponding to the shooting angle information, and outputs a "Please confirm target vehicle" prompt. After completing the preview image and target selection, the controller starts a preset... The system employs a robust user operation monitoring mechanism that supports multimodal adjustment operations. For example, a user can issue a voice command to "select the vehicle on the right," touch another vehicle area on the central control screen, or point to another vehicle area on the central control screen with a gesture. If an adjustment operation is received, the controller parses the new target vehicle's location information from the adjustment operation, recalculates the camera's shooting angle and focus range based on the new target vehicle's location information, updates the shooting perspective information, controls the camera to adjust parameters, regenerates the preview image, and selects the new target vehicle. If no adjustment operation is received, or if the user issues a voice command to "confirm," it is determined that the user approves the currently selected target vehicle, and the controller controls the camera to continuously acquire multiple frames of image information at a preset frame rate.

[0073] This method enables the vehicle-mounted camera to generate a preview image and select the target vehicle based on the shooting angle information. This allows users to intuitively verify the accuracy of the predicted target, reducing misjudgments of target vehicles caused by factors such as multimodal command parsing deviations and external environmental interference, thus improving the accuracy of target vehicle identification. Simultaneously, it grants users adjustment permissions, updating the shooting angle information based on the new target vehicle's location information upon receiving an adjustment command. This ensures that the final captured image accurately corresponds to the target vehicle required by the user, enhancing the targeting and effectiveness of image acquisition. Furthermore, the entire process requires no complex manual operation from the user; image acquisition preparation can be completed solely through preview verification and simple adjustments, minimizing the impact on the user's driving attention, ensuring driving safety, and improving the accuracy and stability of vehicle recognition.

[0074] Step 230: Determine image feature information based on image information.

[0075] For example, image feature information refers to feature vectors extracted from an image that can characterize the attributes of the target vehicle, such as logo outline features, grille texture features, headlight shape features, body line features, etc.

[0076] In one example, the controller can first remove environmental noise from the image information using a Gaussian filtering algorithm, retain the main vehicle area in the image using a region of interest cropping algorithm, remove invalid backgrounds such as road surface and sky, scale the cropped image to a preset size, and then perform pixel value normalization processing to map pixel values ​​to the 0-1 range to obtain a standardized image. Subsequently, the controller calls a preset feature extraction network to perform feature extraction processing on the standardized image to obtain image feature information.

[0077] Step 240: Based on image feature information, determine the initial vehicle information and confidence level of the target vehicle.

[0078] For example, initial vehicle information refers to the preliminary vehicle attribute data obtained by the controller after matching image feature information through the vehicle's local model. This data may include, for example, vehicle brand, model, and other information. Confidence level refers to the quantified value of the reliability of the initial vehicle information; a higher value indicates a more reliable recognition result.

[0079] In one example, the controller inputs image feature information into a vehicle-mounted local vehicle model recognition model. This model has a feature template library of multiple vehicle models pre-stored. The model can compare the similarity of the input image feature information with the features of each vehicle model in the feature template library one by one. Based on the comparison results, the vehicle attribute data corresponding to the vehicle model feature with the highest similarity is selected as the initial vehicle information. At the same time, the model generates a confidence score based on the similarity calculation results.

[0080] Optionally, the controller can receive vehicle feature template data sent by cloud devices to update the feature template library.

[0081] Step 250: Determine whether the confidence level of the initial vehicle information is greater than a preset threshold.

[0082] For example, the preset threshold is a critical value pre-set by the controller to determine whether the initial recognition result is reliable. Its value can be adjusted according to the actual recognition accuracy requirements, for example, a value of 0.9. The core purpose of this step is to determine whether the local recognition result on the vehicle is reliable by comparing the confidence level with the preset threshold, and then decide whether to use local output or cloud-based secondary recognition for subsequent processing.

[0083] In one example, the controller compares the confidence level with a preset threshold; if the confidence level is greater than the preset threshold, the initial vehicle information is determined to be reliable, and step 260 is executed; if the confidence level is less than or equal to the preset threshold, the initial vehicle information is determined to be unreliable, and step 270 is executed.

[0084] Step 260: Use the initial vehicle information as the vehicle information.

[0085] Step 270: Send the image feature information to the cloud device and receive the vehicle information returned by the cloud device.

[0086] For example, the cloud device is a remote server with powerful computing capabilities and a massive vehicle database, and it has a built-in high-precision vehicle recognition model. The returned vehicle information is the final vehicle attribute data obtained by the cloud device through high-precision recognition, which may include information such as brand, model, year, core configuration, and manufacturer's suggested retail price.

[0087] In one example, when the controller determines that the confidence level is less than or equal to a preset threshold, it first encrypts the image feature information obtained in step 230 to prevent leakage during data transmission. Then, through the vehicle networking module, the encrypted image feature information is sent to a preset cloud device. After receiving the data, the cloud device decrypts it and inputs it into a high-precision vehicle model recognition model. It then combines the massive vehicle model database in the cloud for recognition and matching to generate complete vehicle information and return it to the controller. The controller receives the vehicle information returned by the cloud.

[0088] Optionally, cloud devices can periodically update the vehicle model database, for example, by obtaining updated vehicle model data through a preset interface to update the vehicle model database.

[0089] Step 280: Output the vehicle information of the target vehicle.

[0090] It should be noted that this step is similar to step 130 mentioned above, and will not be repeated here.

[0091] In this embodiment, after entering vehicle recognition mode, the controller responds to the user's target selection command to determine the shooting angle information of the on-board camera. Then, based on the shooting angle information, it controls the camera to acquire image information of the target vehicle. Subsequently, image feature information is extracted from the image information. Based on this image feature information, an initial vehicle information and confidence level are determined through a local model. By judging the relationship between the confidence level and a preset threshold, the system chooses to directly use the initial vehicle information as the final vehicle information, or send the image feature information to the cloud device and receive the vehicle information returned from the cloud. Finally, the final vehicle information is output to the user. This method, through a collaborative mode of local vehicle recognition and high-precision cloud recognition, ensures efficient and low-latency recognition of common vehicle models while solving the recognition challenges of niche or similar vehicle models, thus improving recognition accuracy. The entire process requires no manual operation by the user using external devices, minimizing user distraction and ensuring driving safety. Simultaneously, it simplifies the vehicle information acquisition process, improves vehicle information acquisition efficiency, and significantly enhances the user experience.

[0092] Figure 5 A schematic diagram of an embodiment of the vehicle identification device provided by the present invention is shown. Figure 5 As shown, the vehicle identification device 300 includes a control module 310, an identification module 320, and an output module 330.

[0093] The control module 310 is used to control the vehicle's onboard camera to collect image information of the target vehicle indicated by the target selection command after entering the vehicle recognition mode, in response to the target selection command issued by the user.

[0094] The recognition module 320 is used to identify vehicle information of the target vehicle based on image information;

[0095] Output module 330 is used to output vehicle information of the target vehicle.

[0096] In one alternative embodiment, the control module 310 is used for:

[0097] In response to the target selection command, determine the shooting angle information of the vehicle camera;

[0098] Based on the shooting angle information, control the vehicle-mounted camera to collect image information of the target vehicle.

[0099] In an alternative embodiment, when the target selection instruction is a voice instruction, the control module 310 is configured to:

[0100] The voice recognition process is used to determine the location information of the target vehicle carried in the target selection command.

[0101] Based on the location information of the target vehicle, determine the shooting angle information of the vehicle-mounted camera.

[0102] In an alternative embodiment, when the target selection command is a gaze-focusing command, the control module 310 is configured to:

[0103] In response to a target selection command, the user's eye features are obtained;

[0104] Determine the user's gaze direction information based on eye characteristics;

[0105] Based on the direction of the line of sight, determine the shooting angle information of the vehicle camera.

[0106] In an alternative embodiment, when the target selection instruction is a gesture instruction, the control module 310 is configured to:

[0107] In response to a target selection command, the system acquires the user's valid pointing gestures within a preset gesture recognition area.

[0108] Based on valid pointing gestures, determine the fingertip pointing information;

[0109] The shooting angle information of the vehicle camera is determined based on the fingertip pointing information.

[0110] In one alternative embodiment, the control module 310 is used for:

[0111] Based on the shooting angle information, control the vehicle-mounted camera to shoot and display a preview on the vehicle's display screen; the target vehicle is selected in the preview;

[0112] If no adjustment operation is received from the user for the target vehicle, the vehicle camera is controlled to collect image information of the target vehicle based on the shooting angle information.

[0113] If an adjustment operation is received from the user for the target vehicle, the new shooting angle information is determined based on the new target vehicle position information indicated by the adjustment operation, and the vehicle camera is controlled to acquire image information of the new target vehicle according to the new shooting angle information.

[0114] In one alternative embodiment, the identification module 320 is used for:

[0115] Based on the image information, determine the image feature information;

[0116] Based on image feature information, the initial vehicle information and confidence level of the target vehicle are determined;

[0117] If the confidence level of the initial vehicle information is determined to be greater than a preset threshold, then the initial vehicle information is used as the vehicle information.

[0118] If the confidence level of the initial vehicle information is determined to be less than or equal to a preset threshold, the image feature information is sent to the cloud device, and the vehicle information returned by the cloud device is received.

[0119] As can be seen from the above, the vehicle recognition device provided in this embodiment of the invention can automatically complete the entire process from image acquisition to information recognition and output through the in-vehicle device without requiring users to manually search for vehicle information using external devices such as mobile phones. This simplifies the operation steps and improves the efficiency of vehicle information acquisition. At the same time, it eliminates the need for users to manually input or search, reducing interference with driving safety and ensuring driving safety.

[0120] Figure 6 The diagram illustrates a structural schematic of an embodiment of the electronic device provided by the present invention. The specific embodiments of the present invention do not limit the specific implementation of the electronic device. The electronic device may be the aforementioned controller.

[0121] like Figure 6 As shown, the electronic device may include: a processor 402, a communications interface 404, a memory 406, and a communications bus 408.

[0122] The processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408. Communication interface 404 is used to communicate with other network elements such as clients or other servers. The processor 402 executes program 410, specifically performing the relevant steps described above in the vehicle identification method embodiment.

[0123] Specifically, program 410 may include program code, which includes computer-executable instructions.

[0124] Processor 402 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The electronic device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0125] Memory 406 is used to store program 410. Memory 406 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0126] Specifically, program 410 can be called by processor 402 to cause the electronic device to perform the following operations:

[0127] After entering vehicle recognition mode, in response to the target selection command issued by the user, the vehicle's on-board camera is controlled to collect image information of the target vehicle indicated by the target selection command.

[0128] Based on image information, identify vehicle information of the target vehicle;

[0129] Output the vehicle information of the target vehicle.

[0130] In one alternative approach, in response to a target selection command issued by a user, the vehicle's onboard camera is controlled to acquire image information of the target vehicle indicated by the target selection command, including:

[0131] In response to the target selection command, determine the shooting angle information of the vehicle camera;

[0132] Based on the shooting angle information, control the vehicle-mounted camera to collect image information of the target vehicle.

[0133] In one alternative approach, when the target selection command is a voice command, in response to the target selection command, the shooting angle information of the vehicle-mounted camera is determined, including:

[0134] The voice recognition process is used to determine the location information of the target vehicle carried in the target selection command.

[0135] Based on the location information of the target vehicle, determine the shooting angle information of the vehicle-mounted camera.

[0136] In one alternative approach, when the target selection instruction is a gaze-focusing instruction, in response to the target selection instruction, the shooting angle information of the vehicle-mounted camera is determined, including:

[0137] In response to a target selection command, the user's eye features are obtained;

[0138] Determine the user's gaze direction information based on eye characteristics;

[0139] Based on the direction of the line of sight, determine the shooting angle information of the vehicle camera.

[0140] In one alternative approach, when the target selection instruction is a gesture instruction, in response to the target selection instruction, the shooting angle information of the vehicle-mounted camera is determined, including:

[0141] In response to a target selection command, the system acquires the user's valid pointing gestures within a preset gesture recognition area.

[0142] Based on valid pointing gestures, determine the fingertip pointing information;

[0143] The shooting angle information of the vehicle camera is determined based on the fingertip pointing information.

[0144] In one alternative approach, based on the shooting angle information, the vehicle-mounted camera is controlled to acquire image information of the target vehicle, including:

[0145] Based on the shooting angle information, control the vehicle-mounted camera to shoot and display a preview on the vehicle's display screen; the target vehicle is selected in the preview;

[0146] If no adjustment operation is received from the user for the target vehicle, the vehicle camera is controlled to collect image information of the target vehicle based on the shooting angle information.

[0147] If an adjustment operation is received from the user for the target vehicle, the new shooting angle information is determined based on the new target vehicle position information indicated by the adjustment operation, and the vehicle camera is controlled to acquire image information of the new target vehicle according to the new shooting angle information.

[0148] In one alternative approach, vehicle information of the target vehicle is identified based on image information, including:

[0149] Based on the image information, determine the image feature information;

[0150] Based on image feature information, the initial vehicle information and confidence level of the target vehicle are determined;

[0151] If the confidence level of the initial vehicle information is determined to be greater than a preset threshold, then the initial vehicle information is used as the vehicle information.

[0152] If the confidence level of the initial vehicle information is determined to be less than or equal to a preset threshold, the image feature information is sent to the cloud device, and the vehicle information returned by the cloud device is received.

[0153] As can be seen from the above, the electronic device provided in this embodiment of the invention can automatically complete the entire process from image acquisition to information recognition and output through the in-vehicle device without requiring the user to manually search for vehicle information using external devices such as mobile phones. This simplifies the operation steps and improves the efficiency of vehicle information acquisition. At the same time, it eliminates the need for the user to manually input or search, reducing interference with driving safety and ensuring driving safety.

[0154] This invention provides a vehicle that includes an electronic device capable of performing the vehicle identification method in any of the above-described method embodiments.

[0155] This invention provides a computer-readable storage medium storing at least one executable instruction that, when executed on an electronic device / vehicle identification device, causes the electronic device / vehicle identification device to perform the vehicle identification method in any of the above method embodiments.

[0156] Specifically, the executable instructions can be used to cause the electronic device / vehicle identification device to perform the following operations:

[0157] After entering vehicle recognition mode, in response to the target selection command issued by the user, the vehicle's on-board camera is controlled to collect image information of the target vehicle indicated by the target selection command.

[0158] Based on image information, identify vehicle information of the target vehicle;

[0159] Output the vehicle information of the target vehicle.

[0160] In one alternative approach, in response to a target selection command issued by a user, the vehicle's onboard camera is controlled to acquire image information of the target vehicle indicated by the target selection command, including:

[0161] In response to the target selection command, determine the shooting angle information of the vehicle camera;

[0162] Based on the shooting angle information, control the vehicle-mounted camera to collect image information of the target vehicle.

[0163] In one alternative approach, when the target selection command is a voice command, in response to the target selection command, the shooting angle information of the vehicle-mounted camera is determined, including:

[0164] The voice recognition process is used to determine the location information of the target vehicle carried in the target selection command.

[0165] Based on the location information of the target vehicle, determine the shooting angle information of the vehicle-mounted camera.

[0166] In one alternative approach, when the target selection instruction is a gaze-focusing instruction, in response to the target selection instruction, the shooting angle information of the vehicle-mounted camera is determined, including:

[0167] In response to a target selection command, the user's eye features are obtained;

[0168] Determine the user's gaze direction information based on eye characteristics;

[0169] Based on the direction of the line of sight, determine the shooting angle information of the vehicle camera.

[0170] In one alternative approach, when the target selection instruction is a gesture instruction, in response to the target selection instruction, the shooting angle information of the vehicle-mounted camera is determined, including:

[0171] In response to a target selection command, the system acquires the user's valid pointing gestures within a preset gesture recognition area.

[0172] Based on valid pointing gestures, determine the fingertip pointing information;

[0173] The shooting angle information of the vehicle camera is determined based on the fingertip pointing information.

[0174] In one alternative approach, based on the shooting angle information, the vehicle-mounted camera is controlled to acquire image information of the target vehicle, including:

[0175] Based on the shooting angle information, control the vehicle-mounted camera to shoot and display a preview on the vehicle's display screen; the target vehicle is selected in the preview;

[0176] If no adjustment operation is received from the user for the target vehicle, the vehicle camera is controlled to collect image information of the target vehicle based on the shooting angle information.

[0177] If an adjustment operation is received from the user for the target vehicle, the new shooting angle information is determined based on the new target vehicle position information indicated by the adjustment operation, and the vehicle camera is controlled to acquire image information of the new target vehicle according to the new shooting angle information.

[0178] In one alternative approach, vehicle information of the target vehicle is identified based on image information, including:

[0179] Based on the image information, determine the image feature information;

[0180] Based on image feature information, the initial vehicle information and confidence level of the target vehicle are determined;

[0181] If the confidence level of the initial vehicle information is determined to be greater than a preset threshold, then the initial vehicle information is used as the vehicle information.

[0182] If the confidence level of the initial vehicle information is determined to be less than or equal to a preset threshold, the image feature information is sent to the cloud device, and the vehicle information returned by the cloud device is received.

[0183] As can be seen from the above, the computer-readable storage medium provided in the embodiments of the present invention stores at least one executable instruction. When the executable instruction runs on the electronic device / vehicle recognition device, it can automatically complete the entire process from image acquisition to information recognition and output through the vehicle-mounted device without the user having to manually search for vehicle information using external devices such as mobile phones. This simplifies the operation steps and improves the efficiency of vehicle information acquisition. At the same time, it eliminates the need for the user to manually input or search, reducing interference with driving safety and ensuring driving safety.

[0184] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0185] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0186] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0187] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A vehicle identification method, characterized in that, The method is applied to a controller in a vehicle, including: After entering vehicle recognition mode, in response to the target selection command issued by the user, the vehicle's on-board camera is controlled to collect image information of the target vehicle indicated by the target selection command; Based on the image information, identify the vehicle information of the target vehicle; Output the vehicle information of the target vehicle.

2. The method according to claim 1, characterized in that, The step of responding to a user-issued target selection command by controlling the vehicle's onboard camera to acquire image information of the target vehicle indicated by the target selection command includes: In response to the target selection command, the shooting angle information of the vehicle-mounted camera is determined; Based on the shooting angle information, the vehicle-mounted camera is controlled to acquire image information of the target vehicle.

3. The method according to claim 2, characterized in that, When the target selection command is a voice command, determining the shooting angle information of the vehicle-mounted camera in response to the target selection command includes: The target selection command is processed by speech recognition to determine the location information of the target vehicle carried in the target selection command; Based on the location information of the target vehicle, the shooting angle information of the vehicle-mounted camera is determined.

4. The method according to claim 2, characterized in that, When the target selection instruction is a gaze focusing instruction, determining the shooting angle information of the vehicle-mounted camera in response to the target selection instruction includes: In response to the target selection instruction, the user's eye features are acquired; Based on the described eye features, the user's gaze direction information is determined; Based on the line-of-sight information, the shooting angle information of the vehicle-mounted camera is determined.

5. The method according to claim 2, characterized in that, When the target selection instruction is a gesture instruction, determining the shooting angle information of the vehicle-mounted camera in response to the target selection instruction includes: In response to the target selection instruction, the valid pointing gesture of the user within the preset gesture recognition area is obtained; Based on the effective pointing gesture, determine the fingertip pointing information; The shooting angle information of the vehicle camera is determined based on the fingertip pointing information.

6. The method according to claim 2, characterized in that, The step of controlling the vehicle-mounted camera to acquire image information of the target vehicle based on the shooting angle information includes: Based on the shooting angle information, the vehicle-mounted camera is controlled to shoot and a preview image is displayed on the vehicle's display screen; wherein, the target vehicle is selected in the preview image; If no adjustment operation is received from the user for the target vehicle, the vehicle-mounted camera is controlled to acquire image information of the target vehicle based on the shooting angle information; If an adjustment operation is received from the user for the target vehicle, a new shooting angle is determined based on the new target vehicle's location information indicated by the adjustment operation, and the vehicle-mounted camera is controlled to acquire image information of the new target vehicle according to the new shooting angle information.

7. The method according to any one of claims 1-6, characterized in that, The step of identifying the vehicle information of the target vehicle based on the image information includes: Based on the image information, determine the image feature information; Based on the image feature information, the initial vehicle information and confidence level of the target vehicle are determined; If the confidence level of the initial vehicle information is determined to be greater than a preset threshold, then the initial vehicle information is used as the vehicle information. If the confidence level of the initial vehicle information is determined to be less than or equal to a preset threshold, the image feature information is sent to the cloud device, and the vehicle information returned by the cloud device is received.

8. A vehicle identification device, characterized in that, The device is used in a controller in a vehicle and includes: The control module is used to control the vehicle's onboard camera to acquire image information of the target vehicle indicated by the target selection command after entering the vehicle recognition mode, in response to the target selection command issued by the user. The recognition module is used to recognize the vehicle information of the target vehicle based on the image information; The output module is used to output the vehicle information of the target vehicle.

9. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation of the vehicle identification method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on the electronic device / vehicle identification device, causes the electronic device / vehicle identification device to perform the operation of the vehicle identification method as described in any one of claims 1-7.

Images