Vehicle safety reminding method, electronic equipment, storage medium and program product

Abstract

The invention provides a safety reminding method of a vehicle, electronic equipment, a storage medium and a program product, and relates to the technical field of vehicles. The environment visual data is acquired when the vehicle is in a parking state; performing risk factor identification on target image information in the environment visual data to obtain a risk identification result; the target image information at least shows a vehicle door opening range, and the vehicle door opening range comprises a space range covered by a motion trail of at least one vehicle door; and under the condition that the risk identification result indicates that any target risk factor exists in the vehicle door opening range, responding to a vehicle door opening signal, and outputting get-off safety prompt information. According to the invention, accurate and reliable safety reminding can be carried out when the user gets off the vehicle.

Description

Technical Field

[0001] This application involves vehicles. vehicle The field of technology relates, in particular to a vehicle safety alert method, electronic device, storage medium, and program product. Background Technology

[0002] With the increasing intelligence of vehicles, exit safety reminders have become a high-frequency basic interactive function. Their core value lies in mitigating the risk of "door-opening collisions"—collisions caused by users failing to notice approaching vehicles or pedestrians when exiting the vehicle—thus ensuring the safety of passengers and other road users. Therefore, traditional exit safety reminder technologies generally revolve around preventing "door-opening collisions." They use onboard sensors to identify moving objects behind the vehicle (such as vehicles, pedestrians, and non-motorized vehicles), and when the exit environment is deemed unsafe, a buzzer sounds an alarm to alert the user. However, traditional exit safety reminder technologies have poor accuracy in identifying moving objects behind the vehicle, easily resulting in missed detections (e.g., failing to notice a slow-approaching object) or misidentifications (e.g., mistaking a stationary object for a moving one). This leads to alarm signals that cannot accurately match the actual safety scenario. Summary of the Invention

[0003] The main objective of this application is to provide a vehicle safety reminder method, electronic device, storage medium, and program product, aiming to fully leverage the safety protection value of the vehicle exit safety reminder function and enhance the user experience.

[0004] To achieve the above objectives, a first aspect of this application provides a vehicle safety reminder method, the method comprising: Acquire environmental visual data of the vehicle; the environmental visual data is collected when the vehicle is in a parked state. Risk factors are identified from the target image information in the environmental visual data to obtain risk identification results; the target image information at least shows the vehicle door opening range, and the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one vehicle door. If the risk identification result indicates that there is any target risk factor within the vehicle door opening range, the system outputs a vehicle exit safety prompt message in response to the vehicle door opening signal.

[0005] In some embodiments, the method further includes at least one of the following: The environmental visual data is acquired based on a panoramic monitoring image system; The disembarkation safety information includes at least two different modalities of safety information; The modality of the exit safety warning information matches the type of the target risk factor; the target risk factor includes at least one of obstacle risk factors and road condition risk factors.

[0006] In some embodiments, the security alert information includes a first security alert in the audio modality, a second security alert in the image modality, and a third security alert in the video modality.

[0007] In some embodiments, the obstacle risk factors include static obstacle risk factors and dynamic obstacle factors.

[0008] In some embodiments, the step of identifying risk factors from the target image information in the environmental visual data to obtain risk identification results includes: Extract target image information from the environmental visual data; When the target risk factors include obstacle risk factors and road condition risk factors, the target image information is subjected to inference and recognition processing of obstacle risk factors and road condition risk factors based on a preset first visual language model and a second visual language model, respectively, to obtain the risk recognition result. When the target risk factors include obstacle risk factors, the obstacle risk factors are inferred and identified based on a preset first visual language model to obtain the risk identification result. When the target risk factors include road condition risk factors, the target image information is subjected to inference and recognition processing based on a preset second visual language model to obtain the risk recognition result.

[0009] In some embodiments, the risk factor identification of target image information in the environmental visual data includes at least one of the following: Risk factors are identified in the target image information of the environmental visual data in a cyclical manner according to a preset time window. The system acquires an initial risk identification result for the target image information in the environmental visual data, and when the initial risk identification result indicates that there are dynamic obstacle risk factors in the vehicle door opening range, it performs risk factor identification on the target image information in a cyclical manner according to a preset time window.

[0010] In some embodiments, the step of outputting a vehicle exit safety prompt information in response to a vehicle door opening signal includes: Obtain the trigger time information of the vehicle door opening signal, and obtain the target time window in which the trigger time information is located; Based on the target risk identification result corresponding to the target time window, output the vehicle exit safety prompt information; the target risk identification result includes: the risk identification result obtained by identifying risk factors of the target image information within the target time window, or the risk identification result obtained by identifying risk factors of the target image information within the previous time window of the target time window.

[0011] In some embodiments, the vehicle door opening range includes a first opening range of all vehicle doors and a second opening range of any target door among all vehicle doors. When the risk identification result indicates that any target risk factor exists within the vehicle door opening range, in response to the vehicle door opening signal, the system outputs a vehicle exit safety prompt message, including: If the risk identification result indicates that there are risk factors in the first door opening range, in response to the first opening signal, a vehicle exit safety prompt message is output; the first opening signal is a vehicle door opening signal triggered by any door of the vehicle. If the risk identification result indicates that there are risk factors in the second door opening range, a vehicle exit safety prompt message is output in response to the second opening signal; the second opening signal is a vehicle door opening signal triggered by the target door.

[0012] In some embodiments, when the risk identification result indicates that there are risk factors within the vehicle door opening range, the method further includes, before outputting exit safety prompt information in response to the vehicle door opening signal: Based on the target image information, parking scene recognition is performed to obtain the current parking scene; Determine if the current parking scenario is not a familiar parking scenario for the user of the vehicle.

[0013] In some embodiments, the method further includes: The current parking scenario is compared with the vehicle's historical parking scenarios to obtain the comparison results; In response to the comparison result indicating that the current parking scenario is the same as the target historical parking scenario of the vehicle, the cumulative parking count of the target historical parking scenario is incremented by 1 to obtain the updated cumulative parking count; If the updated cumulative number of parkings exceeds a preset parking count threshold, the target historical parking scenario is identified as a familiar parking scenario for the vehicle's user.

[0014] To achieve the above objectives, a second aspect of this application provides a vehicle safety reminder device, the device comprising: The acquisition module is used to acquire environmental visual data of the vehicle; the environmental visual data is acquired when the vehicle is in a parked state. The risk identification module is used to identify risk factors in the target image information in the environmental visual data and obtain risk identification results; the target image information at least shows the vehicle door opening range, and the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one vehicle door. The multimodal alert module is used to output a vehicle exit safety reminder message in response to a vehicle door opening signal when the risk identification result indicates that any target risk factor exists within the vehicle door opening range.

[0015] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the vehicle safety reminder method described in the first aspect.

[0016] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle safety reminder method described in the first aspect.

[0017] To achieve the above objectives, a fifth aspect of this application provides a computer program product, which includes a computer program that, when executed by a processor, implements the vehicle safety reminder method provided in the first aspect above.

[0018] The vehicle safety reminder method, device, electronic device, computer-readable storage medium, and computer program product proposed in this application acquire environmental visual data of the vehicle; the environmental visual data is collected when the vehicle is in a parked state; risk factors are identified in the target image information in the environmental visual data to obtain a risk identification result; the target image information at least shows the vehicle door opening range, the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one door; when the risk identification result indicates that any target risk factor exists in the vehicle door opening range, in response to the vehicle door opening signal, a vehicle exit safety reminder message is output.

[0019] Compared to traditional exit safety reminder technologies, this application embodiment collects environmental visual data when the vehicle is parked. Then, it identifies hazards in the target image information within this environmental visual data that at least shows the vehicle's door opening range (the spatial range covered by the movement trajectory of at least one door). This results in a risk identification result. Finally, if the risk identification result indicates the presence of any target risk factor within the vehicle's door opening range, it responds to the vehicle door opening signal and outputs an exit safety reminder. Thus, this application embodiment, by collecting environmental visual data while the vehicle is parked and accurately identifying the presence of risk factors within the target image information showing the vehicle's door opening range, outputs an exit safety reminder when a risk factor is identified. This not only effectively avoids false alarms and missed alarms in traditional exit safety reminder technologies, improving the accuracy of exit safety judgments, but also provides more reliable safety reminders when the user exits the vehicle, based on accurate exit safety judgments. This fully leverages the safety protection value of the vehicle exit safety reminder function and enhances the user experience. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the steps of the vehicle safety reminder method provided in some embodiments of this application; Figure 2a A schematic diagram of a small window prompt scenario involved in some embodiments of the vehicle safety reminder method provided in this application; Figure 2b A schematic diagram illustrating another small window notification scenario in some embodiments of the vehicle safety reminder method provided in this application; Figure 3 for Figure 1 A detailed flowchart of step S102; Figure 4 for Figure 1 A schematic diagram of another detailed step in step S102; Figure 5 for Figure 1 A detailed flowchart of step S103; Figure 6 for Figure 1 A schematic diagram of another detailed step in step S103; Figure 7 A flowchart illustrating the steps of the vehicle safety reminder method provided in this application in other embodiments; Figure 8 A flowchart illustrating the steps of the vehicle safety reminder method provided in some other embodiments of this application; Figure 9 A complete system flowchart of the vehicle safety reminder method provided in this application embodiment; Figure 10 This is a schematic diagram of the structure of the vehicle safety reminder device provided in the embodiments of this application; Figure 11 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0022] It should be noted that although functional modules are divided in the device / system schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device / system or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0024] First, the overall concept of the vehicle safety reminder method provided in the embodiments of this application will be explained.

[0025] With the increasing intelligence of vehicles, exit safety reminders have become a high-frequency basic interactive function. Their core value lies in mitigating the risk of "door-opening collisions"—collisions caused by users failing to notice approaching vehicles or pedestrians when exiting the vehicle—thus ensuring the safety of passengers and other road users. Therefore, traditional exit safety reminder technologies generally revolve around preventing "door-opening collisions." They use onboard sensors to identify moving objects behind the vehicle (such as vehicles, pedestrians, and non-motorized vehicles), and when the exit environment is deemed unsafe, a buzzer sounds an alarm to alert the user. However, traditional exit safety reminder technologies have poor accuracy in identifying moving objects behind the vehicle, easily resulting in missed detections (e.g., failing to notice a slow-approaching object) or misidentifications (e.g., mistaking a stationary object for a moving one). This leads to alarm signals that cannot accurately match the actual safety scenario.

[0026] In summary, the core problem with traditional exit safety alert technology lies in its insufficient accuracy in recognizing objects behind the vehicle. This issue directly leads to a lack of reliability in the exit safety alert function, failing to effectively prevent "door-opening kills" and potentially impacting user experience due to false alarms or missed alarms. It can even cause users to distrust the function, hindering its full realization of its safety protection value. Therefore, there is an urgent need to optimize traditional exit safety alert technology.

[0027] To address the aforementioned issues, this application proposes a vehicle safety reminder method, device, electronic device, computer-readable storage medium, and computer program product. The aim is to collect video frames from the vehicle's left and right surround views, combine this with a large model to identify obstacles (such as pillars, walls, roadblocks, vehicles, pedestrians, etc.) and muddy roads near the vehicle doors, and then, when the user opens the door, play a voice prompt and display a video image through a small window interface to remind the user to pay attention to safety when getting out of the vehicle. In this way, the safety protection value of the vehicle exit safety reminder function is fully utilized, and the user experience is improved.

[0028] In this embodiment, environmental visual data of the vehicle is acquired; the environmental visual data is collected when the vehicle is parked; risk factors are identified in the target image information in the environmental visual data to obtain a risk identification result; the target image information at least shows the vehicle door opening range, the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one door; when the risk identification result indicates that any target risk factor exists in the vehicle door opening range, a vehicle exit safety prompt message is output in response to the vehicle door opening signal.

[0029] Compared to traditional exit safety reminder technologies, this application embodiment collects environmental visual data when the vehicle is parked. Then, it identifies hazards in the target image information within this environmental visual data that at least shows the vehicle's door opening range (the spatial range covered by the movement trajectory of at least one door). This results in a risk identification result. Finally, if the risk identification result indicates the presence of any target risk factor within the vehicle's door opening range, it responds to the vehicle door opening signal and outputs an exit safety reminder. Thus, this application embodiment, by collecting environmental visual data while the vehicle is parked and accurately identifying the presence of risk factors within the target image information showing the vehicle's door opening range, outputs an exit safety reminder when a risk factor is identified. This not only effectively avoids false alarms and missed alarms in traditional exit safety reminder technologies, improving the accuracy of exit safety judgments, but also provides more reliable safety reminders when the user exits the vehicle, based on accurate exit safety judgments. This fully leverages the safety protection value of the vehicle exit safety reminder function and enhances the user experience.

[0030] Next, the vehicle safety reminder method, device, electronic device, computer-readable storage medium, and computer program product provided in this application will be specifically described through the following embodiments, and firstly, the various detailed embodiments of the vehicle safety reminder method provided in this application will be described in detail.

[0031] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0032] It should be noted that the vehicle safety reminder method provided in this application relates to the field of vehicle technology. The vehicle safety reminder method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be an in-vehicle terminal, a smartphone, tablet, laptop, desktop computer, or other electronic device associated with the vehicle and capable of communicating and interacting with the vehicle via a network. The server can be the backend server terminal device of the terminal, which can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The software can be an application implementing the vehicle safety reminder method, a computer program, and a storage medium carrying the computer program. It should be understood that, based on different design needs in practical applications, the terminal, server, and software of the vehicle safety reminder method provided in this application may also be other forms not listed here, and the vehicle safety reminder method provided in this application does not specifically limit these.

[0033] Furthermore, this application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: vehicle terminals, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, personal computers (PCs), minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0034] For ease of understanding and explanation, the following text will use the vehicle safety reminder method provided in the embodiments of this application applied to an in-vehicle terminal as an example to describe the various specific embodiments of this application in detail. The in-vehicle terminal can use the vehicle safety reminder method provided in the embodiments of this application to control the vehicle. In some descriptions, the in-vehicle terminal may be simply referred to as the terminal. The implementation of the vehicle safety reminder method provided in the embodiments of this application using any of the above-described forms of subject matter can refer to the process of applying the vehicle safety reminder method to an in-vehicle terminal as described below.

[0035] Please refer to Figure 1 , Figure 1 The flowchart illustrates the steps of the vehicle safety reminder method provided in some embodiments of this application. It should be understood that, although... Figure 1 The flowcharts illustrating subsequent steps show the execution order of some method steps. However, based on different design needs in practical applications, the vehicle safety reminder method provided in this application embodiment can, of course, employ a different execution order of method steps than shown in the figures. That is, Figure 1 The order of the method steps shown does not constitute a limitation on the execution logic order of the vehicle safety reminder method provided in the embodiments of this application. Any other order based on... Figure 1 Reasonable changes to the sequence of steps shown should be included within the protection scope of the vehicle safety reminder method provided in the embodiments of this application.

[0036] like Figure 1 As shown, in some embodiments, the vehicle safety reminder method provided in this application may include steps S101 to S103 as shown below.

[0037] Step S101: Acquire the environmental visual data of the vehicle; the environmental visual data is collected when the vehicle is in a parked state.

[0038] It should be noted that environmental visual data can be image data and / or video stream data of the vehicle's surroundings.

[0039] The vehicle-mounted terminal can use the vehicle's surround-view cameras to collect image data and / or video stream data of the vehicle's environment in real time when the vehicle is parked, thereby obtaining the vehicle's environmental visual data.

[0040] In some embodiments, the environmental visual data is acquired based on a panoramic surveillance video system.

[0041] It should be noted that the Around View Monitor (AVM) is a driver assistance system that captures environmental images through surround-view cameras (also known as ultra-wide-angle cameras) around the vehicle and stitches them together using image processing technology to create a 360° panoramic view. The AVM eliminates blind spots for the driver and provides 2D / 3D view switching, pedestrian and obstacle detection functions.

[0042] When collecting environmental visual data of a vehicle, the vehicle-mounted terminal can activate the panoramic monitoring image system (AVM) after the vehicle enters the parking state to collect image data and / or video stream data of the vehicle's environment in real time, and use this image data and / or video stream data as the vehicle's environmental visual data.

[0043] In some embodiments, when the user activates the vehicle's intelligent driving assistance system, the in-vehicle terminal can activate the vehicle's panoramic monitoring image system (AVM) and continuously collect environmental visual data of the vehicle through the AVM. Then, when the vehicle enters a parked state, the exit safety reminder function is activated to obtain the environmental visual data collected by the panoramic monitoring image system (AVM) when the vehicle is parked.

[0044] In some embodiments, the exit safety reminder function can be an intelligent agent service on the vehicle. The in-vehicle terminal can initiate the exit safety reminder agent service when the vehicle starts. Once initiated, the agent service creates a session with the surround-view cameras in the AVM (Around View Monitor). The in-vehicle terminal saves this session, thereby triggering a data acquisition process when the vehicle enters a parked state (e.g., when in P gear), to obtain environmental visual data of the vehicle using the 360° surround-view cameras through this session.

[0045] In some embodiments, after the in-vehicle terminal establishes a session for the surround-view camera, it can first capture a video frame through this session, thereby triggering the initialization of the rendering thread for vehicle environmental visual data. The initialized rendering thread can then be used for subsequent exit safety reminders based on the vehicle's environmental visual data.

[0046] Step S102: Risk factor identification is performed on the target image information in the environmental visual data to obtain risk identification results; the target image information at least shows the vehicle door opening range, and the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one door.

[0047] It should be noted that risk factors can be static, dynamic, and road conditions that affect the safety of the user getting out of the vehicle in the surrounding environment. For example, specific risk factors can include obstacles (such as pillars, walls, roadblocks, vehicles, pedestrians, etc.) and muddy roads.

[0048] After acquiring the vehicle's environmental visual data, the vehicle terminal performs real-time risk factor identification processing on the target image information that displays the vehicle's door opening range (the spatial range covered by the movement trajectory of at least one door) in the environmental visual data, thereby obtaining a risk identification result that characterizes whether there are risk factors in the vehicle's door opening range.

[0049] In some embodiments, the vehicle terminal can perform image recognition processing on target image information in environmental visual data to detect whether there are obstacles such as pillars, walls, roadblocks, stationary vehicles, moving vehicles, and pedestrians in the target image information, and to detect whether there are muddy roads in the target image information, thereby identifying risk factors in the target image information.

[0050] In some embodiments, when the vehicle terminal identifies risk factors in the target image information, if it identifies the aforementioned obstacles and / or muddy roads in the target image information, it confirms a risk identification result indicating the presence of risk factors within the vehicle door opening range. Conversely, if it identifies the aforementioned obstacles and / or muddy roads in the target image information, it confirms a risk identification result indicating the absence of risk factors within the vehicle door opening range.

[0051] Step S103: If the risk identification result indicates that there is any target risk factor within the vehicle door opening range, output a vehicle exit safety prompt message in response to the vehicle door opening signal.

[0052] After receiving the risk identification result, if the result indicates that any target risk factor exists within the vehicle door opening range (such as obstacles like pillars, walls, roadblocks, stationary vehicles, moving vehicles, pedestrians, etc., or road conditions like muddy roads), the vehicle terminal will further detect whether the user has triggered the vehicle door opening signal. If the user performs the door opening operation, thus triggering the vehicle door opening signal, the vehicle terminal will respond to the signal by outputting a vehicle exit safety warning message to remind the user to pay attention to the risk factors within the vehicle door opening range to ensure safety.

[0053] In some embodiments, the disembarkation safety prompt information includes at least two different modalities of safety prompt information.

[0054] When outputting exit safety reminders, the vehicle terminal can output at least two types of safety reminders in different modalities (such as sound, images, and videos) to remind users to pay attention to exit safety through different perception channels.

[0055] In some embodiments, the modality of the exit safety warning information matches the type of the target risk factor; the target risk factor includes at least one of obstacle risk factors and road condition risk factors.

[0056] The mode of the exit safety prompt information output by the vehicle terminal can be matched with the type of target risk factors present within the vehicle's door opening range. For example, if the target risk factors include obstacle risk factors, the vehicle terminal will output at least an audio mode of exit safety prompt information, while if the target risk factors include road condition risk factors, the vehicle terminal will output at least an image mode of exit safety prompt information.

[0057] In some embodiments, the obstacle risk factors include static obstacle risk factors and dynamic obstacle factors.

[0058] It should be noted that static obstacle risk factors include, but are not limited to, pillars, walls, roadblocks, stationary vehicles, etc., which may cause injury to users getting out of the vehicle, pedestrians, or even vehicles. Dynamic obstacle risk factors include, but are not limited to, moving vehicles, pedestrians, etc., which may cause injury to users getting out of the vehicle, pedestrians, or even vehicles.

[0059] In some embodiments, the security alert information includes a first security alert in the audio modality, a second security alert in the image modality, and a third security alert in the video modality.

[0060] It should be noted that the first safety alert information in the sound modality can be a beeping sound and / or a voice broadcast safety reminder, while the second safety alert information in the image modality can be environmental image data displaying road condition information. In addition, the second safety alert information in the image modality can be a warning light and / or a video alert displayed in a small window.

[0061] When the vehicle terminal responds to a vehicle door opening signal and outputs a safety reminder for getting out of the vehicle, based on the type of target risk factors present within the vehicle door opening range, it can output at least a first safety reminder in audio mode, a second safety reminder in image mode, and / or a third safety reminder in video mode to provide a safety reminder to the user's current action of opening the door and getting out of the vehicle. For example, when the vehicle terminal outputs multimodal safety reminders for getting out of the vehicle, if the risk identification result indicates that there are both obstacle risk factors and road condition risk factors within the vehicle door opening range, the vehicle terminal, in addition to broadcasting the first safety reminder via voice (such as "There are obstacles around the door, please observe carefully before getting out of the vehicle to ensure safety" or "The vehicle is parked on a muddy road, please observe the road surface and avoid getting dirty"), also displays a message in a small window on the vehicle's central control display screen (such as...). Figure 2a ) and / or the small window in the second-row ceiling screen of the vehicle (such as Figure 2b ), displaying a second safety warning message in the image modality (such as video frames showing obstacles or muddy roads).

[0062] In this embodiment, an in-vehicle terminal, based on an AVM (Ambient Visual Monitoring) system, collects environmental visual data of the vehicle using the vehicle's surround-view cameras when the vehicle is parked. Then, the in-vehicle terminal performs real-time risk factor identification processing on target image information within this environmental visual data that indicates the vehicle door opening range (the spatial range covered by the door's movement trajectory), obtaining a risk identification result characterizing whether any risk factors exist within that vehicle door opening range. If the risk identification result indicates the presence of any target risk factor within the vehicle door opening range, the in-vehicle terminal, in response to a user-triggered vehicle door opening signal, outputs a vehicle exit safety warning message to provide a safety reminder to the user regarding their current door opening and exiting behavior.

[0063] Compared to traditional exit safety reminder technologies, this embodiment utilizes a panoramic monitoring system to collect environmental visual data when the vehicle is parked. Then, it identifies hazards within the visual data that at least shows the vehicle's door opening range (the spatial area covered by the door's movement trajectory). If the risk identification result indicates the presence of a hazard within the door opening range, it responds to the user-triggered door opening signal and outputs multimodal exit safety reminder information. Thus, this embodiment accurately identifies the presence of hazard factors within the vehicle's door opening range by using a panoramic monitoring system. Furthermore, when a hazard factor is identified, it outputs multimodal reminder information to ensure exit safety. This effectively avoids false alarms and missed alarms common in traditional exit safety reminder technologies, improving the accuracy of exit safety assessments. It provides more reliable safety reminders when the user exits the vehicle, maximizing the safety protection value of the exit safety reminder function and enhancing the user experience.

[0064] Furthermore, this embodiment of the application improves the efficiency of extracting target image information from environmental visual data for risk factor identification by pre-creating a session for the surround-view camera and triggering the initialization of the rendering thread. For example, when the vehicle terminal captures target image information through the session, the session returns an image object, which the vehicle terminal needs to convert into an ARGB bitmap for data storage. Thus, pre-creating the session for the surround-view camera and triggering the initialization of the rendering thread can at least improve image conversion performance, thereby improving the overall efficiency of risk factor identification based on the target image information. In addition, this embodiment of the application outputs multimodal exit safety prompts to remind users of the safety of exiting the vehicle. The user can confirm the specific risk factors when exiting the vehicle through a small window display, or confirm the specific road conditions such as mud, dirt, or debris, making it easier for the user to further judge whether there is a real safety risk or a vehicle misidentification. Thus, the combination of voice broadcast and small window display prompts makes the user's perception clearer and more user-friendly.

[0065] Please refer to Figure 3 , Figure 3 for Figure 1 A detailed flowchart of step S102.

[0066] like Figure 3As shown, in some embodiments, the step of “identifying risk factors in the target image information in the environmental visual data and obtaining risk identification results” in step S102 above may include steps S301 to S303 as shown below.

[0067] Step S301: Extract target image information from the environmental visual data.

[0068] When the vehicle terminal performs risk factor identification processing based on environmental visual data, it can first extract target image information that at least shows the range of the vehicle door opening from the environmental visual data.

[0069] In some embodiments, when the vehicle terminal continuously collects environmental visual data of the vehicle from the start of the vehicle using the vehicle's surround view camera through the panoramic monitoring imaging system (AVM), it can extract real-time video frame images showing the opening range of the vehicle's left and right doors and / or tailgate from the environmental visual data collected by the surround view camera when the vehicle enters the parking state. These video frame images are the target image information in the environmental visual data that at least shows the opening range of the vehicle doors.

[0070] In some embodiments, when the vehicle terminal uses the panoramic monitoring imaging system (AVM) to collect environmental visual data of the vehicle using the vehicle's surround-view cameras while the vehicle is parked (e.g., when the vehicle is in P gear and triggers the data acquisition process to obtain the vehicle's environmental visual data using the 360° surround-view cameras through a previously created session), it can directly extract real-time video frame images showing the opening range of the vehicle's left and right doors and / or tailgate from the environmental visual data, thereby using the video frame images as target image information in the environmental visual data that at least shows the opening range of the vehicle doors.

[0071] In some embodiments, when the vehicle terminal extracts video frame images showing the opening range of the left and right vehicle doors from the environmental visual data, only one image is needed for each side of the vehicle. If an image has already been obtained from one side, subsequent video frames returned from the same side can be discarded by the vehicle terminal. This process continues until an image has been obtained from both sides. Once the vehicle terminal confirms that it has extracted video frame images showing the opening range of the left and right vehicle doors, it proceeds to the next stage.

[0072] Step S302: When the target risk factors include obstacle risk factors and road condition risk factors, based on the preset first visual language model and second visual language model, the target image information is subjected to reasoning and recognition processing for obstacle risk factors and road condition risk factors, respectively, to obtain the risk recognition result.

[0073] It should be noted that the first and second visual language models can be obtained by fine-tuning and training a Vision-Language Model (VLM). The VLM, also known as a large-scale visual model, is a cutting-edge branch of artificial intelligence that combines the capabilities of computer vision (CV) and natural language processing (NLP), enabling it to process both visual and linguistic information simultaneously. In-vehicle terminals can fine-tune and train the VLM using visual question-answering tasks, enabling the trained VLM to accurately analyze and identify obstacle risk factors (such as pillars, walls, roadblocks, stationary vehicles, moving vehicles, pedestrians, etc.) in the input image, thus using this trained VLM as the first visual language model. Similarly, in-vehicle terminals can fine-tune and train the VLM using visual question-answering tasks, enabling the trained VLM to accurately analyze and identify road condition risk factors (such as water accumulation, mud, rain, snow, ice, etc.) in the input image, thus using this trained VLM as the second visual language model.

[0074] After extracting the target image information, the vehicle-mounted terminal further inputs this information into a preset first visual language model and a second visual language model. Based on these two models, it performs obstacle risk factor and road condition risk factor inference and recognition processing on the target image information, respectively, to obtain static obstacle risk factors and / or dynamic obstacle risk factors output by the first visual language model, and / or road condition risk factors output by the second visual language model. Thus, the vehicle-mounted terminal can use the risk factors output by the first and / or second visual language models as the risk recognition result obtained from the current risk factor recognition processing of the target image information.

[0075] Step S303: When the target risk factor includes obstacle risk factors, the target image information is subjected to obstacle risk factor inference and recognition processing based on a preset first visual language model to obtain the risk recognition result.

[0076] After extracting the target image information, the vehicle-mounted terminal can input only the target image information into a preset first visual language model. Based on this model, it can then perform obstacle risk factor inference and recognition processing on the target image information, obtaining static obstacle risk factors and / or dynamic obstacle risk factors output by the first visual language model. In this way, the vehicle-mounted terminal can use the risk factors output by the first visual language model as the risk recognition result obtained from the current risk factor recognition processing of the target image information.

[0077] Step S304: When the target risk factor includes road condition risk factors, the target image information is subjected to inference and recognition processing of road condition risk factors based on a preset second visual language model to obtain the risk recognition result.

[0078] After extracting the target image information, the vehicle-mounted terminal can also input only the target image information into a preset second visual language model. Based on this second visual language model, it can then perform inference and recognition processing of road condition risk factors on the target image information, obtaining the road condition risk factors output by the second visual language model. In this way, the vehicle-mounted terminal can use the risk factors output by the second visual language model as the risk recognition result obtained from the current risk factor recognition processing of the target image information.

[0079] In some embodiments, after extracting target image information, the vehicle-mounted terminal can first perform preliminary recognition processing on the target image information to preliminarily determine the type of target risk factors contained in the target image information. Then, if it is preliminarily determined that the target risk factors include obstacle risk factors, the target image information is input into a first visual language model to perform obstacle risk factor inference recognition processing based on the first visual language model, thereby obtaining a risk recognition result. Alternatively, if it is preliminarily determined that the target risk factors include road condition risk factors, the target image information is input into a second visual language model to perform road condition risk factor inference recognition processing based on the second visual language model, thereby obtaining a risk recognition result. Or, if it is preliminarily determined that the target risk factors include both obstacle risk factors and road condition risk factors, the target image information is simultaneously input into the first visual language model and the second visual language model, performing obstacle risk factor inference recognition processing based on the first visual language model and road condition risk factor inference recognition processing based on the second visual language model, thereby obtaining a risk recognition result.

[0080] In some embodiments, the vehicle terminal can input the target image information and the prompt information indicating whether the indicator model can infer and identify the presence of obstacle risk factors in the image into the first visual language model. This allows the first visual language model to use the target image information as the input image, perform inference and identification on whether there are obstacle risk factors in the input image according to the prompt information, and output the finally identified obstacle risk factors.

[0081] In some embodiments, the vehicle terminal may also input the prompt information indicating whether the indicator model is inferring and identifying road condition risk factors in the image into the second visual language model. This allows the second visual language model to perform inference and identification processing on whether there are obstacle risk factors in the input image while taking the target image information as the input image and performing inference and identification processing on whether there are road condition risk factors in the input image according to the prompt information, and output the finally identified road condition risk factors.

[0082] In some embodiments, before inputting the target image information into the visual language model (first visual language model and / or second visual language model) for model inference, the vehicle terminal can preprocess the target image information. Specifically, it can first draw a bounding box for recognition on the image, for example, drawing a bounding box based on the distance of the car door collision (1 meter to the left and right). After drawing the bounding box, the image is cropped and scaled. The image resolution should match the optimal resolution for the visual language model's inference performance (different models have different resolution requirements), thus improving the model's inference performance and accuracy. After image preprocessing, the vehicle terminal then sends the image along with the prompt words into the visual language model for inference and recognition.

[0083] In some embodiments, considering that the real-time requirements for risk factor identification of target images are relatively high in the scenario of getting off the vehicle safety reminder, the vehicle terminal can set a high priority for model requests (used to request the first visual language model and / or the second visual language model to perform risk factor inference and identification processing on the target image information). In this way, when the visual language model has other model request tasks being processed, the model inference and identification of risk factors on the target image information can also preempt the visual language model to achieve priority processing.

[0084] In this embodiment, target image information is extracted from environmental visual data via an in-vehicle terminal. Then, obstacle risk factor inference and recognition processing is performed on the target image information based on a first visual language model, and road condition risk factor inference and recognition processing is performed on the target image information based on a second visual language model, thereby obtaining the risk recognition results output by the first and / or second visual language models. Thus, target image information within the vehicle door opening range collected by the surround-view camera is sent to the VLM to identify obstacles and muddy roads. The types of obstacles identified include static targets (such as pillars, walls, roadblocks, and vehicles), dynamic targets (pedestrians, cyclists, and tricycles), and other obstacles, as well as exit road conditions (muddy or dirty roads at the exit point). Compared to traditional sensor-based obstacle identification, this embodiment, through training, can continuously improve its recognition capabilities, thereby identifying richer obstacle information, especially muddy or dirty road surfaces not found in traditional technologies. This effectively expands the scope of scenarios for providing safety reminders to users when they exit the vehicle.

[0085] Please refer to Figure 4 , Figure 4 for Figure 1 A schematic diagram of another detailed step in step S102.

[0086] like Figure 4 As shown, in some embodiments, the step of “performing long short-term memory hierarchical storage processing on the multimodal interactive input information” in step S102 above may include at least one of steps S401 and S402 as shown below.

[0087] Step S401: Identify risk factors for the target image information in the environmental visual data in a cyclical manner according to a preset time window.

[0088] When identifying risk factors in target image information, the vehicle-mounted terminal takes into account the possibility that the identified risk factors may move. It can initiate cyclic detection, directly executing the aforementioned operation of identifying risk factors in target image information from environmental visual data repeatedly within a preset time window (e.g., 500ms). That is, within each time window, target image information is extracted from the environmental visual data and input into a first visual language model for inference and identification of obstacle risk factors (especially dynamic obstacle risk factors). In this way, the vehicle terminal device can obtain a risk identification result from the inference and identification output of the first visual language model within each time window.

[0089] Step S402: Obtain the initial risk identification result of the target image information in the environmental visual data for risk factor identification, and when the initial risk identification result indicates that there is a dynamic obstacle risk factor in the vehicle door opening range, perform risk factor identification on the target image information in a loop according to a preset time window.

[0090] When identifying risk factors in target image information, to avoid wasting resources of the first visual language model due to directly initiating cyclic detection (for example, if only static obstacle risk factors and / or road condition risk factors exist in the target image information, cyclically performing inference and recognition processing based on the first and second visual language models will always output the same risk identification result, thus wasting model resources), the vehicle terminal can first perform initial risk factor identification on the target image information after extracting it, obtaining an initial risk identification result. Then, if the initial risk identification result indicates that there are dynamic obstacle risk factors within the vehicle door opening range, the vehicle terminal will then begin cyclically identifying risk factors in the target image information according to a preset time window. In this way, the vehicle terminal device can also obtain a risk identification result output by the first visual language model in each time window, and each risk identification result can characterize the dynamic obstacle risk factors within the vehicle door opening range.

[0091] In this embodiment, the vehicle-mounted terminal initiates cyclic detection when identifying risk factors in the target image information. That is, it cyclically identifies risk factors in the target image information in the environmental visual data according to a preset time window. In this way, dynamic obstacle risk factors (such as moving vehicles, pedestrians, etc.) within the vehicle door opening range or dynamic obstacle risk factors that may appear within the vehicle door opening range can be detected in real time and accurately during their movement, thereby further improving the accuracy of the exit safety judgment.

[0092] Please refer to Figure 5 , Figure 5 for Figure 1 A detailed flowchart of step S103.

[0093] like Figure 5 As shown, in some embodiments, the step of “outputting a vehicle exit safety prompt information in response to the vehicle door opening signal” in step S103 above may include steps S501 and S502 as shown below.

[0094] Step S501: Obtain the trigger time information of the vehicle door opening signal, and obtain the target time window in which the trigger time information is located.

[0095] When the vehicle terminal detects that a user has triggered a door opening signal by performing a door opening operation, it first obtains the trigger time information of the door opening signal, and then matches the trigger time information with the time window of the cyclical risk factor time, thereby determining the target time window in which the trigger time information is located among multiple time windows.

[0096] For example, suppose the vehicle terminal starts identifying risk factors in 1-second time windows from 14:21:30, and then detects a user opening a door at 14:22:10, triggering a vehicle door opening signal. In this way, the vehicle terminal can select the time window from 14:22:09 to 14:22:10 as the target time window from among 40 time windows (14:21:30 to 14:22:10) based on the trigger time information of the vehicle door opening signal (14:22:10).

[0097] Step S502: Output a vehicle exit safety prompt based on the target risk identification result corresponding to the target time window; the target risk identification result includes: a risk identification result obtained by identifying risk factors of the target image information within the target time window, or a risk identification result obtained by identifying risk factors of the target image information within the previous time window of the target time window.

[0098] After determining the target time window, the vehicle-mounted terminal further selects the target risk identification result corresponding to the target time window from the risk identification results of multiple time windows, and outputs a vehicle exit safety prompt based on the target risk identification result. Specifically, if the vehicle-mounted terminal has already completed the risk factor identification operation on the target image information within the target time window, thus obtaining the risk identification result for that target time window, the vehicle-mounted terminal can directly use that risk identification result as the target risk identification result corresponding to that target time window. However, if the vehicle-mounted terminal has not yet completed the risk factor identification operation on the target image information within the target time window, and therefore has not yet obtained the risk identification result for that target time window, the vehicle-mounted terminal can use the risk factor identification result obtained from the previous time window as the target risk identification result corresponding to that target time window. In addition, based on the target risk identification results, the system outputs safety prompts for getting out of the vehicle, such as: when the target risk identification results indicate that there are dynamic obstacles such as vehicles / pedestrians within the vehicle's door opening range, the system broadcasts a first safety prompt in sound mode (such as "Vehicles / pedestrians are approaching the door, please observe carefully before getting out of the vehicle to ensure safety"), and simultaneously displays a second safety prompt in image mode (such as video frame images used to show the changes in the distance between the vehicle / pedestrian and the door) in a small window on the vehicle's central control display screen, and even displays a third safety prompt in video mode (such as a video stream used to show the changes in the distance between the vehicle / pedestrian and the door) in the small window.

[0099] In some embodiments, during the process of cyclically identifying risk factors in target image information from environmental visual data, the vehicle-mounted terminal can continuously detect the presence of risk factors (dynamic / static obstacles, muddy ground, etc.) within the vehicle's door opening range as soon as the vehicle enters a parking state (e.g., in P gear). Each detected risk identification result directly overwrites the previous one. In this way, there will always be only one risk identification result, allowing the vehicle-mounted terminal to directly determine that risk identification result as the target risk identification result after defining the target time window.

[0100] It should be noted that during the process of identifying risk factors in the target image information of the environmental visual data, the vehicle terminal can continue to perform loop detection until the user performs a door opening operation, triggering the vehicle door opening signal and thus triggering the exit safety reminder (i.e., outputting multimodal exit safety prompt information).

[0101] In this embodiment, when the vehicle-mounted terminal initiates cyclic detection during risk factor identification of target image information, to ensure that the user is given a safety reminder when getting out of the vehicle based on the latest detected risk identification result, the vehicle-mounted terminal can select the latest risk identification result from the risk identification results of each time window based on the trigger time information of the vehicle door opening signal to output the prompt. Thus, considering that the target risk factors collected and detected by the vehicle-mounted terminal during the exit safety reminder function may move (i.e., dynamic obstacle risk factors), cyclic detection of risk factors is initiated after the vehicle is parked (in P gear), and the exit safety reminder information is output based on the last detected risk identification result when the user triggers the vehicle door opening signal. This ensures that the exit safety reminder is based on the latest detection result when the user opens the door, thereby guaranteeing the real-time accuracy of detecting moving obstacles and outputting reminders.

[0102] In some embodiments, the vehicle door opening range includes a first opening range of all vehicle doors and a second opening range of any target door among all vehicle doors.

[0103] When the in-vehicle terminal identifies risk factors in target image information from environmental visual data, the risk factor may exist within the first opening range of all doors. For example, there may be static obstacles within the opening range of both the right front door and the right rear door, and pedestrians within the opening range of the left front door, left rear door, and tailgate. Alternatively, the risk factor may only exist within the second opening range of a single target door. For example, there may be pedestrians within the opening range of the right front door, and dynamic obstacles within the opening range of the right rear door. In this case, regardless of whether the risk factor exists within the first or second opening range, the risk identification result detected by the in-vehicle terminal directly indicates the presence of a risk factor within the vehicle's opening range, specifically indicating the presence of a risk factor within the first or second opening range. When the in-vehicle terminal provides safety reminders for a user exiting the vehicle, if only one side of the vehicle requires a safety reminder, the announcement and pop-up reminder will only be triggered when the user opens the corresponding door.

[0104] Please refer to Figure 6 , Figure 6 for Figure 1 A schematic diagram of another detailed step in step S103.

[0105] like Figure 6 As shown, in some embodiments, the step S103 above, "in response to the vehicle door opening signal, outputting a vehicle exit safety prompt information when the risk identification result indicates that there is any target risk factor within the vehicle door opening range", may include steps S601 and S602 as shown below.

[0106] Step S601: If the risk identification result indicates that there are risk factors in the first door opening range, in response to the first opening signal, output a vehicle exit safety prompt message; the first opening signal is a vehicle door opening signal triggered by any door of the vehicle.

[0107] When all vehicle doors have risk factors within their first opening range, the onboard terminal, after identifying risk factors from the target image information in the environmental visual data, determines that any target risk factor exists within that first opening range. If the user then opens any of the vehicle doors, triggering a first opening signal, the onboard terminal will respond by outputting a safety warning message to remind the user to be careful when exiting the vehicle. For example, if there are static obstacles within the opening range of the right front door and right rear door, and pedestrians within the opening range of the left front door, left rear door, and tailgate, the onboard terminal will remind the user to be careful when exiting the vehicle via voice announcement and / or a pop-up notification when opening any door.

[0108] Step S602: If the risk identification result indicates that there are risk factors in the second door opening range, in response to the second opening signal, output a vehicle exit safety prompt message; the second opening signal is a vehicle door opening signal triggered by the target door.

[0109] If a risk factor exists within the opening range of one or more target doors of the vehicle, the onboard terminal, after performing risk factor identification on the target image information in the environmental visual data, indicates the presence of any target risk factor within the second opening range. At this point, if the user opens the target door, triggering a second opening signal, the onboard terminal will respond to this signal by outputting a safety reminder to alert the user to the safety of exiting the vehicle. For example, if a pedestrian is within the opening range of the vehicle's right front door, or a dynamic obstacle is within the opening range of the vehicle's right rear door, the onboard terminal will only remind the user to be careful when exiting the vehicle via voice announcement and / or a pop-up notification when the user opens either the right front or right rear door.

[0110] Please refer to Figure 7 , Figure 7 The following are schematic flowcharts illustrating the steps of the vehicle safety reminder method provided in some other embodiments of this application.

[0111] like Figure 7As shown, in some embodiments, the vehicle safety reminder method provided in this application may further include steps S701 and S702 as shown below.

[0112] Step S701: If the risk identification result indicates that there are no risk factors in the first door opening range, abandon the exit safety reminder.

[0113] When the vehicle-mounted terminal identifies risk factors from target image information in the environmental visual data, it may find that there are no risk factors within the first opening range of all vehicle doors. In this case, the risk identification result detected by the vehicle terminal will indicate that there are no risk factors within the first opening range. Therefore, the vehicle-mounted terminal determines that the user will not encounter a safety threat when exiting the vehicle, and thus abandons the exit safety warning.

[0114] Step S702: If the risk identification result indicates that there is a risk factor in the second door opening range and a third opening signal is detected, the exit safety reminder is abandoned; the third opening signal is the opening signal of all doors of the vehicle other than the target door.

[0115] If a risk factor exists within the opening range of one or more target doors of the vehicle, the onboard terminal, after performing risk factor identification on the target image information in the environmental visual data, indicates that a risk factor exists within the second opening range. If the user does not open the target door but instead opens another door, triggering a third opening signal, the onboard terminal, upon detecting this third opening signal, also determines that the user's act of getting out of the vehicle will not pose a safety threat, and thus abandons the safety reminder for getting out of the vehicle. For example, if there is a pedestrian within the opening range of the vehicle's right front door, or a dynamic obstacle within the opening range of the vehicle's right rear door, and the user attempts to open the left front door, left rear door, or tailgate, the onboard terminal will not remind the user to be careful when getting out of the vehicle via voice announcement or pop-up notification.

[0116] In this embodiment, when the risk identification result indicates that there are risk factors within the first opening range of all doors, the vehicle terminal outputs a safety reminder message upon the user triggering the first opening signal of any door. Conversely, if the risk identification result indicates that there are risk factors within the second opening range of any target door, the vehicle terminal outputs a safety reminder message upon the user triggering the second opening signal of that target door. Furthermore, the vehicle terminal abandons the safety reminder if the risk identification result indicates that there are no risk factors within the first opening range, and also abandons the safety reminder if the risk identification result indicates that there are risk factors within the second opening range, but the user has triggered the third opening signal of another door. Thus, when the vehicle terminal provides safety reminders for the user's exit behavior, if safety reminders are needed on both sides of the vehicle (e.g., there are obstacles or muddy roads around the vehicle), the reminder will be given whenever the user opens any door. If only one side of the vehicle requires a safety reminder, the reminder will only be given when the user opens the corresponding door. This ensures accurate and reliable safety reminders for users when getting out of the vehicle, while avoiding unnecessary alarms when users get out of the vehicle but there is no safety threat. This enhances the intelligence of the vehicle exit safety reminder function and provides users with a better experience.

[0117] Please refer to Figure 8 , Figure 8 A flowchart illustrating the steps of the vehicle safety reminder method provided in some other embodiments of this application.

[0118] like Figure 8 As shown, in some embodiments, when the risk identification result indicates that there are risk factors in the vehicle door opening range, the vehicle safety reminder method provided in this application embodiment may further include the following steps S801 and S802 before the above-mentioned step of "outputting exit safety reminder information in response to vehicle door opening signal".

[0119] Step S801: Based on the target image information, perform parking scene recognition to obtain the current parking scene; Step S802: Determine a familiar parking scenario for the user of the vehicle that is not in the current parking scenario.

[0120] It should be noted that familiarity with parking scenarios refers to parking scenarios where the user has parked and exited the vehicle multiple times (e.g., 10, 15, 20 times), such as family parking spaces, private parking spaces, etc. The in-vehicle terminal can assume that the vehicle has already received multiple safety reminders for exiting the vehicle in the user's familiar parking scenario, including dynamic / static obstacles, muddy roads, and other risk factors. This allows the user to remember and avoid these risks through repeated exits, thus eliminating the need for repeated reminders.

[0121] When the detected risk identification result indicates the presence of any target risk factor within the vehicle door opening range, the in-vehicle terminal can also provide a safety reminder to the user when getting out of the vehicle, based on stored data. This stored data can be historical parking scenarios previously identified and recorded by the in-vehicle terminal that require a safety reminder. Specifically, the in-vehicle terminal performs scene recognition processing on the target image information to obtain the vehicle's current parking scenario. It then compares this current parking scenario with historical parking scenarios to determine if it is a familiar parking scenario for the user. Only if the in-vehicle terminal determines that the current parking scenario is not a familiar one will it further execute the step of "responding to the vehicle door opening signal and outputting a safety reminder message," thus providing a safety reminder to the user in an unfamiliar parking scenario.

[0122] In some embodiments, the vehicle safety reminder method provided in this application may further include the following steps: The current parking scenario is compared with the vehicle's historical parking scenarios to obtain the comparison results; In response to the comparison result indicating that the current parking scenario is the same as the target historical parking scenario of the vehicle, the cumulative parking count of the target historical parking scenario is incremented by 1 to obtain the updated cumulative parking count; If the updated cumulative number of parkings exceeds a preset parking count threshold, the target historical parking scenario is identified as a familiar parking scenario for the vehicle's user.

[0123] When determining whether the current parking scenario is a familiar parking scenario for the user, the in-vehicle terminal compares the current parking scenario with the vehicle's historical parking scenarios to obtain a comparison result indicating whether the current parking scenario is the same as a target historical parking scenario. Then, if the comparison result indicates that the current parking scenario is the same as a target historical parking scenario, the in-vehicle terminal increments the cumulative parking count for that target historical parking scenario by 1, obtaining an updated cumulative parking count. Finally, the in-vehicle terminal compares this updated cumulative parking count with a preset parking count threshold (e.g., 10, 15, 20 times, etc.). If the updated cumulative parking count is greater than the preset threshold, the target historical parking scenario is marked as a familiar parking scenario for the user. In this way, the in-vehicle terminal can determine whether the current parking scenario is a familiar parking scenario for the user.

[0124] In some embodiments, after the in-vehicle terminal performs risk factor identification on target image information in the visual environment data and obtains a risk identification result, if the risk identification result indicates that any target risk factor exists within the vehicle door opening range, the in-vehicle terminal determines that the target image information is an image that requires a safety reminder to get out of the vehicle after inference and identification, and places these images into a queue of fixed length. Thus, each time target image information is collected from the environmental visual data for risk factor inference and identification, the in-vehicle terminal can directly compare the target image information with the images in the queue to confirm whether the parking locations indicated by the two images are the same location. If they are the same location, the count is incremented by 1. Furthermore, the in-vehicle terminal sets a count limit. If the accumulated count of the parking locations indicated by the target image information currently undergoing risk factor identification exceeds this count limit, the in-vehicle terminal determines that the parking location is a familiar parking location for the user, thereby determining that the current parking scenario is a familiar parking scenario for the user. In this way, after the vehicle terminal identifies risk factors from the target image information, even if the risk identification result indicates that there are risk factors (such as obstacles or muddy roads) within the vehicle door opening range, it will not trigger a safety reminder to the user to get out of the vehicle.

[0125] In some embodiments, the vehicle terminal manages the entry and exit of images in a fixed-length queue using the principle of least-earliest deletion. That is, the image with the fewest count is deleted first, and if the counts are equal, the earliest record is deleted. This allows the number of parking spaces where users frequently park and drop off their vehicles to be incremented by 1 until the upper limit is reached.

[0126] In some embodiments, if the vehicle terminal determines that the current parking scenario is a familiar parking scenario for the user, the vehicle terminal may choose to abandon the safety reminder to the user to get out of the car in this scenario.

[0127] In this embodiment, when the risk identification result indicates that there are risk factors within the vehicle door opening range, the vehicle terminal performs parking scene recognition processing based on target image information to obtain the current parking scene. Only when it is determined that the current parking scene is not a familiar parking scene for the user, does the step of outputting a vehicle exit safety reminder in response to the vehicle door opening signal occur. If the current parking scene is determined to be a familiar parking scene for the user, the exit safety reminder is not issued. In this way, by remembering the locations where the user frequently parks and exits the vehicle, the exit safety reminder is not issued to the user in familiar parking scenes to avoid excessive reminders and disturbing the user. This further enhances the intelligence of the vehicle exit safety reminder function while ensuring accurate and reliable exit safety reminders, thus improving the user experience.

[0128] Please refer to Figure 9 , Figure 9 The overall system flowchart of the vehicle safety reminder method provided in the embodiments of this application in a complete embodiment.

[0129] like Figure 9 As shown, in one embodiment, the overall process of the vehicle terminal applying the vehicle safety reminder method provided in this application embodiment to remind the user to get out of the vehicle includes the following stages 1) to 6): Phase 1) Start the vehicle disembarkation safety agent service.

[0130] When the vehicle starts, the exit safety reminder agent service is activated. After the service starts, it creates and saves a session with the surround-view camera. Creating the session in advance allows for direct image capture during the subsequent data collection phase, thereby improving the speed of image capture.

[0131] After creating a session for the surround-view camera, a video frame will be captured once, which triggers the initialization of the rendering thread. This can improve the performance of converting the image to an ARGB bitmap during the data acquisition stage (the native Android camera session returns an image object, which needs to be converted to an ARGB bitmap for data storage).

[0132] Phase 2) Data collection.

[0133] When the vehicle is put into Park (P) gear, triggering the data acquisition process, the data acquisition module uses the session of the surround-view camera (stored in step 1) to capture real-time video frames from the left and right sides of the vehicle respectively, and caches the original images in memory. Only one image is needed from each side; once an image is captured from one side, subsequent frames from the same side are discarded until images from both sides are captured, at which point the process proceeds to the next stage.

[0134] Phase 3) Image preprocessing and model inference.

[0135] Before the image is sent to the model for inference, it needs to be preprocessed: First, draw the recognition region bounding box on the image, based on the distance of the collision with the car door (generally limiting the recognition range to within 1 meter on the left and right); after drawing the region bounding box, the image is cropped and scaled (the image resolution should match the optimal resolution for the model's inference performance to improve inference performance and accuracy; different models with different parameters have different resolution requirements).

[0136] After image preprocessing, it is sent along with prompts to the Visual Language Model (VLM) for inference and recognition. Because this scenario has high real-time requirements, a high priority can be set in the model request, allowing it to preemptively process other model requests.

[0137] The model identifies two categories of obstacles and muddy roads. Obstacles include: freestanding roadblocks, row of roadblocks, vehicles, two-wheeled vehicles, three-wheeled vehicles, pedestrians, people riding bicycles, walls, pillars, and other obstacles; muddy roads include: roads with water or mud that may cause wet shoes or slipping.

[0138] The recognition result is returned in the form of a string token, which describes the recognition results on the left and right sides of the vehicle. The results on the left and right sides are separated by special characters (such as newline characters). For example, "with pillar - muddy road surface" indicates that there is a pillar on the left side of the vehicle, while the road surface on the right side of the vehicle is muddy road surface.

[0139] The process of collecting images and inferring from the model is a loop. Once the vehicle is in park, it continuously detects surrounding obstacles, muddy ground, and other obstacles. The results of each subsequent detection overwrite the results of the previous one, and the loop stops only when the door is opened and a user notification is triggered.

[0140] Stage 4) Recognition and memory.

[0141] Images identified as requiring alerts are placed in a fixed-length queue. During the next image collection and inference process, the queued images are compared to determine if they represent the same location. If they do, the count is incremented by 1, with a maximum count limit. Images entering and leaving the queue follow a least-first-delete principle: images with the lowest counts are deleted first, and if counts are equal, the oldest record is deleted. Frequently used parking spaces accumulate counts until the limit is reached. Parking spaces that have reached the limit are considered familiar locations, and even if obstacles or muddy roads are detected, user alerts will not be triggered.

[0142] Phase 5) User reminders.

[0143] After the model inference is completed, the user will not be immediately reminded to be careful when getting out of the car. The user reminder is triggered by combining the model inference result and the door opening signal. This includes the following cases A), B), and C): Scenario A) The model's inference results show no obstacles or muddy road surfaces, and no warnings are given; Scenario B) If the model's inference indicates an obstacle and the car door is not currently open, no warning will be issued until the door is opened. If there are obstacles or muddy / muddy surfaces on both the left and right sides, opening either door will trigger a warning; if there are only obstacles or muddy / muddy surfaces on the left or right side, a warning will only be issued when the corresponding door is opened. Scenario C) If the model inference result indicates an obstacle, and the car door is already open, a warning will be issued directly. The left and right warning rules are the same as in B), that is, if there are obstacles or muddy roads on both the left and right sides, opening either car door will trigger a warning; if there are obstacles or muddy roads only on the left or right side, a warning will only be issued when the corresponding side door is opened. The alert is delivered via both a voice message and a small window on the screen. The voice message is similar to: "There are obstacles and water accumulation on the left and right sides of the vehicle. Please be careful." Small windows are displayed on the left and right sides respectively, showing the original images captured by the 360° surround-view camera cached in step 2). After a period of time (5 seconds), the small windows automatically close.

[0144] Phase 6) Reset the reasoning results.

[0145] If the model inference results in the presence of obstacles or muddy road surfaces, the alert will not be triggered if the car door remains closed. This is because the vehicle may have started moving or the user may have exited and locked the car (if there are obstacles on the other side of the driver's side). Therefore, when the vehicle shifts to a gear other than P or the vehicle's usage mode changes, the inference result needs to be reset to indicate no obstacles or muddy road surfaces.

[0146] In this embodiment, by using fisheye images captured by a surround-view camera, collision objects or unsafe road surfaces (such as muddy or muddy surfaces) on both sides of the vehicle after parking (at least including the driver's and passenger's door opening areas) are identified. Compared to traditional methods that primarily identify potential collision risks from moving objects behind the vehicle (such as capturing rear-side images using a wide-angle intelligent driving camera), this method improves the accuracy of identifying risks around the vehicle by recognizing obstacles and unsafe road surfaces in different locations and application scenarios, avoiding missed or false identifications. This fully leverages the safety protection value of the vehicle exit safety reminder function and enhances the user experience. Furthermore, compared to traditional sensor-based obstacle identification, this embodiment uses a trained visual language model for risk factor inference and identification, continuously improving the ability to identify risks around the vehicle. It can identify richer obstacle information and, notably, includes the identification of muddy or muddy road surfaces, which is not present in traditional technologies. This expands the range of safety reminder scenarios. Additionally, this embodiment avoids excessive reminders by remembering frequently parked locations and not providing reminders. Furthermore, compared to the traditional method of using alarm lights / beep sounds for single-modal output prompts, this embodiment combines voice broadcasts and small window displays to output multi-modal exit safety prompts, which can make users' perception of exit safety reminders clearer and more user-friendly.

[0147] Please refer to Figure 11 This application also provides a vehicle safety reminder device, which can implement the above-described vehicle safety reminder method.

[0148] like Figure 11 As shown, the vehicle safety reminder device provided in this application embodiment may include: The acquisition module is used to acquire environmental visual data of the vehicle; the environmental visual data is acquired when the vehicle is in a parked state. The risk identification module is used to identify risk factors in the target image information in the environmental visual data and obtain risk identification results; the target image information at least shows the vehicle door opening range, and the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one vehicle door. The multimodal alert module is used to output a vehicle exit safety reminder message in response to a vehicle door opening signal when the risk identification result indicates that any target risk factor exists within the vehicle door opening range.

[0149] In some embodiments, the environmental visual data is acquired based on a panoramic monitoring image system; the exit safety prompt information includes at least two types of safety prompt information in different modalities; the modality of the exit safety prompt information matches the type of the target risk factor; the target risk factor includes at least one of obstacle risk factors and road condition risk factors.

[0150] In some embodiments, the security alert information includes a first security alert in the audio modality, a second security alert in the image modality, and a third security alert in the video modality.

[0151] In some embodiments, the obstacle risk factors include static obstacle risk factors and dynamic obstacle factors.

[0152] In some embodiments, the risk identification module is further configured to extract target image information from the environmental visual data; when the target risk factors include obstacle risk factors and road condition risk factors, the module performs inference and identification processing on the target image information based on a preset first visual language model and a preset second visual language model to obtain the risk identification result; when the target risk factors include obstacle risk factors, the module performs inference and identification processing on the target image information based on a preset first visual language model to obtain the risk identification result; when the target risk factors include road condition risk factors, the module performs inference and identification processing on the target image information based on a preset second visual language model to obtain the risk identification result.

[0153] In some embodiments, the risk identification module is further configured to perform risk factor identification on the target image information in the environmental visual data in a cyclical manner according to a preset time window; obtain the initial risk identification result of the risk factor identification on the target image information in the environmental visual data, and, when the initial risk identification result indicates that there is a dynamic obstacle risk factor in the vehicle door opening range, perform risk factor identification on the target image information in a cyclical manner according to the preset time window.

[0154] In some embodiments, the multimodal alert module is further configured to acquire the trigger time information of the vehicle door opening signal and acquire the target time window in which the trigger time information is located; output a vehicle exit safety reminder based on the target risk identification result corresponding to the target time window; the target risk identification result includes: a risk identification result obtained by identifying risk factors of the target image information within the target time window, or a risk identification result obtained by identifying risk factors of the target image information within the previous time window of the target time window.

[0155] In some embodiments, the vehicle door opening range includes a first opening range of all vehicle doors and a second opening range of any target door among all vehicle doors. The risk identification module is further configured to, in response to a first opening signal, output a vehicle exit safety warning message when the risk identification result indicates that there is a risk factor in the first door opening range; the first opening signal is a vehicle door opening signal triggered by any door of the vehicle; and in response to a second opening signal, output a vehicle exit safety warning message when the risk identification result indicates that there is a risk factor in the second door opening range; the second opening signal is a vehicle door opening signal triggered by the target door.

[0156] In some embodiments, the risk identification module is further configured to identify parking scenarios based on the target image information to obtain the current parking scenario; and determine that the current parking scenario is not a familiar parking scenario for the user of the vehicle.

[0157] In some embodiments, the multimodal reminder module is further configured to compare the current parking scenario with the vehicle's historical parking scenarios to obtain a comparison result; in response to the comparison result indicating that the current parking scenario is the same as the vehicle's target historical parking scenario, increment the cumulative parking count of the target historical parking scenario by 1 to obtain an updated cumulative parking count; if the updated cumulative parking count is greater than a preset parking count threshold, determine the target historical parking scenario as a familiar parking scenario for the vehicle's user.

[0158] It should be noted that the specific implementation of the vehicle safety reminder device provided in this application is basically the same as the specific implementation of the vehicle safety reminder method described above, and will not be repeated here.

[0159] Please see Figure 11 This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-mentioned vehicle safety reminder method.

[0160] In some embodiments, the electronic device can be any smart terminal such as an in-vehicle terminal, an in-vehicle hardware platform (e.g., an in-vehicle computer), a tablet computer, a smartphone, and a wearable device.

[0161] like Figure 11 As shown, the electronic device provided in this application embodiment may include: The processor 1101 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1102 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1102 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1102 and is called by the processor 1101 to execute the vehicle safety reminder method of the embodiments of this application. Input / output interface 1103 is used to implement information input and output; The communication interface 1104 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1105 transmits information between various components of the device (e.g., processor 1101, memory 1102, input / output interface 1103, and communication interface 1104); The processor 1101, memory 1102, input / output interface 1103 and communication interface 1104 are connected to each other within the device via bus 1105.

[0162] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned vehicle safety reminder method.

[0163] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0164] This application also provides a computer program product, including a computer program. The steps implemented by the computer program when executed by a processor are basically the same as those in the specific embodiments of the vehicle safety reminder method described above, and will not be repeated here.

[0165] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0166] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0167] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0168] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0169] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification 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 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.

[0170] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0171] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0172] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0173] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0174] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0175] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A vehicle safety reminder method, characterized in that, The method includes: Acquire environmental visual data of the vehicle; the environmental visual data is collected when the vehicle is in a parked state. Risk factors are identified from the target image information in the environmental visual data to obtain risk identification results; the target image information at least shows the vehicle door opening range, and the vehicle door opening range includes the spatial range covered by the movement trajectory of at least one vehicle door. If the risk identification result indicates that there is any target risk factor within the vehicle door opening range, the system outputs a vehicle exit safety prompt message in response to the vehicle door opening signal.

2. The method according to claim 1, characterized in that, The method further includes at least one of the following: The environmental visual data is acquired based on a panoramic monitoring image system; The disembarkation safety information includes at least two different modalities of safety information; The modality of the exit safety warning information matches the type of the target risk factor; the target risk factor includes at least one of obstacle risk factors and road condition risk factors.

3. The method according to claim 2, characterized in that, The security alert information includes a first security alert in the audio modality, a second security alert in the image modality, and a third security alert in the video modality.

4. The method according to claim 2, characterized in that, The obstacle risk factors include static obstacle risk factors and dynamic obstacle factors.

5. The method according to claim 1, characterized in that, The step of identifying risk factors from target image information in the environmental visual data to obtain risk identification results includes: Extract target image information from the environmental visual data; When the target risk factors include obstacle risk factors and road condition risk factors, the target image information is subjected to inference and recognition processing of obstacle risk factors and road condition risk factors based on a preset first visual language model and a second visual language model, respectively, to obtain the risk recognition result. When the target risk factors include obstacle risk factors, the obstacle risk factors are inferred and identified based on a preset first visual language model to obtain the risk identification result. When the target risk factors include road condition risk factors, the target image information is subjected to inference and recognition processing based on a preset second visual language model to obtain the risk recognition result.

6. The method according to claim 1, characterized in that, The risk factor identification of target image information in the environmental visual data includes at least one of the following: Risk factors are identified in the target image information of the environmental visual data in a cyclical manner according to a preset time window. The system acquires an initial risk identification result for the target image information in the environmental visual data, and when the initial risk identification result indicates that there are dynamic obstacle risk factors in the vehicle door opening range, it performs risk factor identification on the target image information in a cyclical manner according to a preset time window.

7. The method according to claim 6, characterized in that, The response to the vehicle door opening signal, outputting exit safety prompt information, includes: Obtain the trigger time information of the vehicle door opening signal, and obtain the target time window in which the trigger time information is located; Based on the target risk identification result corresponding to the target time window, output the vehicle exit safety prompt information; the target risk identification result includes: the risk identification result obtained by identifying risk factors of the target image information within the target time window, or the risk identification result obtained by identifying risk factors of the target image information within the previous time window of the target time window.

8. The method according to claim 1, characterized in that, The vehicle door opening range includes the first opening range of all vehicle doors, and the second opening range of any target door among all vehicle doors. When the risk identification result indicates that any target risk factor exists within the vehicle door opening range, in response to the vehicle door opening signal, the system outputs a vehicle exit safety prompt message, including: If the risk identification result indicates that there are risk factors in the first door opening range, in response to the first opening signal, a vehicle exit safety prompt message is output; the first opening signal is a vehicle door opening signal triggered by any door of the vehicle. If the risk identification result indicates that there are risk factors in the second door opening range, a vehicle exit safety prompt message is output in response to the second opening signal; the second opening signal is a vehicle door opening signal triggered by the target door.

9. The method according to any one of claims 1 to 8, characterized in that, If the risk identification result indicates that there are risk factors within the vehicle door opening range, the method further includes, before outputting exit safety prompt information in response to the vehicle door opening signal: Based on the target image information, parking scene recognition is performed to obtain the current parking scene; Determine if the current parking scenario is not a familiar parking scenario for the user of the vehicle.

10. The method according to claim 9, characterized in that, The method further includes: The current parking scenario is compared with the vehicle's historical parking scenarios to obtain the comparison results; In response to the comparison result indicating that the current parking scenario is the same as the target historical parking scenario of the vehicle, the cumulative parking count of the target historical parking scenario is incremented by 1 to obtain the updated cumulative parking count; If the updated cumulative number of parkings exceeds a preset parking count threshold, the target historical parking scenario is identified as a familiar parking scenario for the vehicle's user.

11. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the vehicle safety reminder method according to any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the vehicle safety alert method as described in any one of claims 1 to 10.

13. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the vehicle safety reminder method as described in any one of claims 1 to 10.

Images