Electronic mirror-based blind area warning method and related device

Abstract

The application discloses a blind area early warning method based on an electronic rearview mirror and a related device. The method comprises the following steps: acquiring an original image of at least one predetermined collection direction through an image collection device, and preprocessing the original image to obtain a target image; determining a target early warning area in the target image according to the running state of a vehicle; classifying a reference moving object in the target early warning area to obtain a classification result; determining a ranging range of the reference moving object according to the classification result; screening the reference moving object according to the ranging range to obtain a target moving object; determining a target risk level of the target moving object; early warning the vehicle according to the target risk level, and marking the target moving object and the target risk level on the target image displayed on the display screen. The application can improve the accuracy and reliability of blind area early warning.

Description

Technical Field

[0001] This application relates to the field of vehicle electronics technology, and in particular to a blind spot warning method and related device based on an electronic rearview mirror. Background Technology

[0002] Traditional optical rearview mirrors are limited by mirror size, vehicle structure, and installation angle, making it difficult to achieve full coverage of the area around the vehicle and creating blind spots. These blind spots can seriously affect the driver's judgment of the surrounding traffic conditions. For example, the driver may not be able to see a motorcycle approaching rapidly from the side or rear when changing lanes, a pedestrian may be blocked by the A-pillar when turning, or a vehicle may suddenly merge into the adjacent lane at high speed. The lack of vision in these scenarios can lead to delayed or misjudgment by the driver, greatly increasing the risk of traffic accidents.

[0003] However, while current blind spot warning systems can detect the presence of targets within blind spots, they cannot accurately determine the true threat level of the targets, leading to frequent false alarms in non-dangerous scenarios and causing driver fatigue; or they may react too slowly in truly dangerous scenarios, making it difficult to provide effective warnings. Summary of the Invention

[0004] This application provides a blind spot warning method and related device based on an electronic rearview mirror to improve the accuracy and reliability of blind spot warning.

[0005] In a first aspect, embodiments of this application provide a blind spot warning method based on an electronic rearview mirror, applied to a control chip of an electronic rearview mirror system. The electronic rearview mirror system further includes an image acquisition device installed outside the vehicle and a display screen installed inside the vehicle. The control chip is connected to the image acquisition device and the display screen respectively. The method includes:

[0006] The image acquisition device acquires at least one original image in a predetermined acquisition direction, and preprocesses the original image to obtain the target image.

[0007] The target warning area in the target image is determined based on the vehicle's operating status. The target warning area is the area in the target image that is associated with the occurrence of the accident.

[0008] The reference moving objects in the target warning area are classified to obtain the classification results;

[0009] The ranging range of the reference moving object is determined based on the classification results;

[0010] The reference moving objects are filtered according to the ranging range to obtain the target moving object;

[0011] Determine the target risk level of the moving object;

[0012] The vehicle is given a warning based on the target risk level, and the target moving object and the target risk level are marked on the target image displayed on the screen.

[0013] The step of determining the target warning area in the target image based on the vehicle's operating status includes:

[0014] Determine the direction of the early warning based on the described operating status;

[0015] The shape of the target warning area is determined based on the warning direction;

[0016] Obtain the first type, first speed, and road conditions of the vehicle;

[0017] The target warning area is determined based on the shape, the first type, the first driving speed, and the driving conditions.

[0018] The step of determining the target warning area based on the shape, the first type, the first driving speed, and the road conditions includes:

[0019] Determine the warning distance based on the first driving speed;

[0020] The field of view width and field of view angle are determined according to the first type;

[0021] Referring to the shape, the initial warning area is determined based on the field of view width, the field of view angle, and the warning distance;

[0022] The correction factor is determined based on the described road conditions;

[0023] The initial warning area is adjusted according to the correction coefficient to obtain the target warning area.

[0024] The determination of the initial warning area based on the shape, the field of view width, the field of view angle, and the warning distance includes:

[0025] A second type of the shape is determined, the second type including sector-shaped and oblique rectangular shapes;

[0026] If the second type is a fan shape, the radius of the fan shape is determined based on the field of view width and the warning distance;

[0027] The initial warning area is determined based on the field of view angle and the radius;

[0028] If the second type is the oblique rectangular shape, the horizontal dimension is determined according to the field of view width;

[0029] The longitudinal dimension is determined based on the warning distance;

[0030] The oblique angle is determined based on the stated field of view angle;

[0031] The initial warning area is determined based on the horizontal dimension, the vertical dimension, and the diagonal angle.

[0032] Determining the target risk level of the target moving object includes:

[0033] Based on the intrinsic and extrinsic parameters of the image acquisition device, the target distance between the moving target and the vehicle is determined;

[0034] Determine the second speed of the target moving object;

[0035] The target warning area is divided into multiple warning sub-areas based on the degree of danger.

[0036] Determine the warning sub-region where the target moving object is located;

[0037] The target risk level of the moving object is determined based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed.

[0038] The step of determining the target risk level of the moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed includes:

[0039] Determine the first risk coefficient of the warning sub-region where the target moving object is located;

[0040] A second risk factor for the target moving object is determined based on the second driving speed and the first driving speed of the vehicle.

[0041] A third risk factor for the moving target object is determined based on the target distance;

[0042] The target risk level is determined based on the first risk coefficient, the second risk coefficient, and the third risk coefficient.

[0043] The method further includes, before acquiring at least one original image in a predetermined acquisition direction using the image acquisition device and preprocessing the original image to obtain the target image:

[0044] Obtain the first type of the vehicle;

[0045] The current driving scenario of the vehicle is determined based on the operating status;

[0046] The at least one predetermined acquisition direction is determined based on the current driving scenario and the first type.

[0047] Secondly, embodiments of this application provide a blind spot warning device based on an electronic rearview mirror, and a control chip applied to an electronic rearview mirror system. The electronic rearview mirror system includes an image acquisition device installed outside the vehicle and a display screen installed inside the vehicle. The control chip is connected to the image acquisition device and the display screen respectively. The device includes:

[0048] The acquisition unit is used to acquire at least one original image in a predetermined acquisition direction through the image acquisition device, and to preprocess the original image to obtain a target image;

[0049] The first determining unit is used to determine the target warning area in the target image based on the vehicle's operating status, wherein the target warning area is an area in the target image that is associated with the occurrence of an accident;

[0050] A classification unit is used to classify reference moving objects in the target warning area and obtain classification results;

[0051] The second determining unit is used to determine the ranging range of the reference moving object based on the classification result;

[0052] A filtering unit is used to filter the reference moving object according to the ranging range to obtain the target moving object;

[0053] The third determining unit is used to determine the target risk level of the target moving object;

[0054] The warning unit is used to issue a warning to the vehicle based on the target risk level, and to mark the target moving object and the target risk level on the target image displayed on the screen.

[0055] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code and performs the steps of the method described in the first aspect.

[0056] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable program code, the executable program code including execution instructions for performing the steps of the method as described in the first aspect.

[0057] Fifthly, embodiments of this application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in the first aspect of embodiments of this application. The computer program product may be a software installation package.

[0058] As can be seen, in this embodiment, the image acquisition device first acquires at least one original image in a predetermined acquisition direction, and preprocesses the original image to obtain a target image; then, based on the vehicle's operating status, a target warning area in the target image is determined, where the target warning area is the area in the target image associated with the accident; next, reference moving objects in the target warning area are classified to obtain a classification result; then, the ranging range of the reference moving objects is determined based on the classification result; then, the reference moving objects are filtered based on the ranging range to obtain target moving objects; next, the target risk level of the target moving objects is determined; finally, a warning is issued to the vehicle based on the target risk level, and the target moving objects and the target risk level are marked on the target image displayed on the screen.

[0059] This application narrows the detection range and enhances scene adaptability by dynamically defining the warning area that matches the real-time vehicle status. Based on this, it matches a differentiated ranging range according to the moving object classification results and filters moving objects through this range. This not only filters out low-risk targets to reduce false alarms but also prevents missed risk detection caused by a fixed ranging range. Finally, by combining target risk level determination with visual label output, it achieves accurate response to blind spot warnings, thereby improving the accuracy and reliability of blind spot warnings. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0061] Figure 1 This application provides a system architecture diagram of an electronic rearview mirror system.

[0062] Figure 2 This is a schematic diagram of the architecture of a control chip provided in an embodiment of this application;

[0063] Figure 3 This is a schematic diagram of a chip in an installation scenario provided in an embodiment of this application;

[0064] Figure 4 This is a flowchart illustrating a blind spot warning method based on an electronic rearview mirror provided in an embodiment of this application;

[0065] Figure 5 This is a flowchart of a method for determining a target warning region in a target image, provided in an embodiment of this application.

[0066] Figure 6 This is a flowchart of a method for determining the target risk level of a moving object according to an embodiment of this application;

[0067] Figure 7 This is a schematic diagram of a target image provided in an embodiment of this application;

[0068] Figure 8 This is a functional unit block diagram of a blind spot warning device based on an electronic rearview mirror provided in an embodiment of this application;

[0069] Figure 9 This is a block diagram of the functional units of another blind spot warning device based on an electronic rearview mirror provided in this application embodiment;

[0070] Figure 10 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application. Detailed Implementation

[0071] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0072] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. 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 includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0073] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0074] Traditional optical rearview mirrors are limited by mirror size, vehicle structure, and installation angle, making it difficult to achieve full coverage of the area around the vehicle and creating blind spots. These blind spots can seriously affect the driver's judgment of the surrounding traffic conditions. For example, the driver may not be able to see a motorcycle approaching rapidly from the side or rear when changing lanes, a pedestrian may be blocked by the A-pillar when turning, or a vehicle may suddenly merge into the adjacent lane at high speed. The lack of vision in these scenarios can lead to delayed or misjudgment by the driver, greatly increasing the risk of traffic accidents.

[0075] However, while current blind spot warning systems can detect the presence of targets within blind spots, they cannot accurately determine the true threat level of the targets, leading to frequent false alarms in non-dangerous scenarios and causing driver fatigue; or they may react too slowly in truly dangerous scenarios, making it difficult to provide effective warnings.

[0076] To address the aforementioned issues, this application provides a blind spot warning method and related device based on an electronic rearview mirror. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0077] Please see Figure 1 , Figure 1 This is a system architecture diagram of an electronic rearview mirror system provided in an embodiment of this application. Figure 1 As shown, the electronic rearview mirror system 100 includes a control chip 101, an image acquisition device 102, and a display screen 103, wherein the control chip 101 is connected to the image acquisition device 102 and the display screen 103 respectively.

[0078] The image acquisition device 102 is installed outside the vehicle to cover blind spots, such as adjacent lanes to the side and rear of the vehicle, the area near the vehicle obscured by the A-pillar, and the low area below the front of the vehicle. Its installation location dynamically adapts to different vehicle types to ensure accurate blind spot monitoring. The device can be equipped with one or more high-definition wide-angle cameras to achieve multi-angle full coverage.

[0079] Specifically, the image acquisition device 102 acquires images in real time and automatically adjusts the shooting parameters according to the driving status, such as turning on night vision mode at night and adjusting the exposure in rainy weather, to adapt to all scenarios such as daytime, tunnels, snowy days, and nighttime, and eliminate visual interference in extreme environments; it continuously acquires blind spot images or video data corresponding to each installation position; and finally transmits the acquired raw image data to the control chip 101 in real time.

[0080] Please refer to Figure 2 , Figure 2 This is a schematic diagram of the architecture of a control chip provided in an embodiment of this application, as shown below. Figure 2 As shown, the control chip 101 is an electronic rearview mirror chip. The brand name of the electronic rearview mirror chip is ORITEK, which may be the Longquan 560MINI V100 chip. It is used to realize functions such as multi-side field of view acquisition, multi-scene image enhancement, blind spot target intelligent recognition, and driving risk warning.

[0081] For example, please refer to Figure 3 , Figure 3 This is a schematic diagram of a chip in an installation scenario provided in an embodiment of this application. Figure 3 This shows the integration status of the Longquan 560 chip and its hardware carrier, such as... Figure 3 As shown, in a real-world hardware installation scenario, the chip is mounted on the corresponding pad area of ​​the circuit board base and electrically connected to the circuit board via its bottom pins. The chip's surface also displays the chip identification, chip model LQ560, and production traceability code, for example, as shown below. Figure 3 As shown, the production traceability code can be 200M1120324050101500263502010933-CN.

[0082] Among them, such as Figure 2 As shown, the control chip 101 includes a microcontroller unit (MCU), a camera deserializer, a controller area network (CAN) transceiver, and a power management integrated circuit (PMIC). The MCU includes a neural processing unit (NPU) and an image signal processor (ISP). The NPU is used for blind spot warnings, and the ISP is used for image quality optimization.

[0083] The camera deserializer is used to receive the raw image serial data sent by the image acquisition device 102 through serial transmission protocols, such as Low-Voltage Differential Signaling (LVDS) and Camera Serial Interface 2 (CSI 2) under the Mobile Industry Processor Interface (MIPI), and deserialize it into parallel image data so that the ISP can process the image later. The CAN transceiver is used to establish a communication connection between the control chip 101 and the vehicle CAN bus, so as to realize bidirectional data interaction between the control chip 101 and the vehicle control system. The PMIC is used to provide a stable, accurate and rated power supply voltage that matches the working requirements of each module of the control chip 101, such as MCU, NPU, ISP, camera deserializer, and CAN transceiver, so as to ensure that each module of the control chip 101 can work continuously and stably in all weather and all scenarios.

[0084] In one possible embodiment, other integrated chips containing ISP and NPU can also be selected to implement the blind spot warning method based on electronic rearview mirrors. The specific chip is not limited here.

[0085] Specifically, the control chip 101 receives raw image data from multiple cameras; it then preprocesses the raw image data by fusing it with an ISP and an AI model. For example, the AI ​​model enables functions such as LED anti-flicker, anti-glare, dynamic range control, noise reduction, and defogging, generating images suitable for all weather conditions and scenarios.

[0086] Then, the NPU runs AI algorithms to identify the warning area, classify moving objects within the warning area, determine the distance range of the moving objects based on the classification results, screen the moving objects for risks based on the distance range, and finally combine vehicle parameters and real-time driving status to determine potential hazards and generate warning signs.

[0087] The control chip 101 is also used to overlay the optimized image data with the warning information and transmit it to the display screen 103.

[0088] The display screen 103 is used to display optimized image data and warning information in real time, and to overlay highlighted prompts on risk targets, thereby achieving a visual field and risk warning feedback, which greatly improves driving safety.

[0089] Specifically, the control chip 101, image acquisition device 102, and display screen 103 communicate with each other via their respective pin connections. For example, the control chip 101 can be packaged in a ball grid array (BGA) form.

[0090] The control chip 101 includes communication pins, functional pins, display output pins, deserializer control pins, display control pins, and vehicle communication pins. For example... Figure 1 As shown, 1 is a communication pin, 2 is a functional pin, 3 is a display output pin, 4 is a deserializer control pin, 5 is a display control pin, and 6 is a vehicle communication pin.

[0091] Among them, communication pin 1 is used to receive raw image data transmitted by image acquisition device 102; function pin 2 is used to control the data interaction and control signal transmission between the internal NPU and ISP of chip 101; display output pin 3 is used to output image data stream with superimposed warning information to display screen 103; deserializer control pin 4 is used to control the start and stop of camera deserializer; display control pin 5 is used to send start / stop, brightness adjustment and other function control commands to display screen 103; vehicle communication pin 6 is used to interact with vehicle system.

[0092] Based on this, this application provides a blind spot warning method based on an electronic rearview mirror, which will be described in detail below with reference to the accompanying drawings.

[0093] Please see Figure 4 , Figure 4 This is a flowchart illustrating a blind spot warning method based on an electronic rearview mirror provided in an embodiment of this application. Figure 4 As shown, this method is applied to the control chip of an electronic rearview mirror system. The electronic rearview mirror system also includes an image acquisition device installed outside the vehicle and a display screen installed inside the vehicle. The control chip is connected to the image acquisition device and the display screen respectively, and includes the following steps:

[0094] S410, the image acquisition device acquires at least one original image in a predetermined acquisition direction, and preprocesses the original image to obtain the target image.

[0095] The predetermined collection direction is a pre-planned shooting area based on the actual needs of vehicle blind spot monitoring, which can accurately cover various visual blind spots of the vehicle.

[0096] For example, it could be below the front grille to cover the near blind spot in front of the vehicle; it could be to the side and rear to monitor high-risk areas for lane changes and merging; or it could be to the rear to monitor high-risk areas when reversing or following another vehicle.

[0097] The preprocessing includes LED anti-flicker, which uses an AI model to detect the frequency of the LED light source in real time and dynamically adjusts the shutter speed to suppress flicker; anti-glare, which uses deep learning to identify highlight areas and intelligently compress the brightness range to avoid overexposure; dynamic range control, which merges multiple frames to expand the dynamic range and preserve details in both bright and dark areas; dehazing, which performs weather compensation and uses AI semantic segmentation to distinguish between fog, rain, and other conditions, enhancing the outlines of objects in low-visibility scenes; and local contrast optimization, which enhances edge sharpness for vehicles or pedestrians to improve their recognizability.

[0098] In one possible embodiment, before acquiring an original image in at least one predetermined acquisition direction using the image acquisition device and preprocessing the original image to obtain a target image, the method further includes: acquiring a first type of the vehicle; determining the current driving scenario of the vehicle based on the operating state; and determining the at least one predetermined acquisition direction based on the current driving scenario and the first type.

[0099] Different driving scenarios and different vehicle types correspond to different data collection directions.

[0100] Specifically, the vehicle type can be determined based on pre-stored vehicle configuration parameters or vehicle identification numbers. The first type can be passenger cars, commercial vehicles including trucks and buses, and special vehicles including engineering vehicles and sanitation vehicles.

[0101] The operating status can include information such as vehicle speed, gear position signal, steering angle, turn signal status, accelerator or brake signal, and parking status. Based on the above information, the current driving scenario is determined, which can include driving, parking, reversing, turning, and changing lanes.

[0102] In the driving scenario, the risk may come from vehicles approaching from the front and sides; in the parking scenario, the risk may come from collisions with pedestrians and non-motorized vehicles; in the reversing scenario, the risk may come from the rear and the rear side; in the turning scenario, the risk may come from the front, the side, and the rear side; and in the lane changing scenario, the risk may come from the rear on both sides.

[0103] First, the acquisition direction can be determined based on the scenario. For example, in a reversing scenario, the risk is in the rear, so image acquisition devices should be set up in the rear first. Then, the number of acquisitions can be adjusted according to the vehicle type. For example, large vehicles have a larger blind spot, so in the same scenario, the acquisition direction needs to be increased compared to small vehicles.

[0104] As can be seen, in this embodiment of the application, by combining vehicle type and driving scenario to dynamically determine the image acquisition direction, adapting to the blind spot characteristics of different vehicle models and the needs of various scenarios, the risk area is accurately covered, effectively reducing collision hazards in driving, reversing, turning and other scenarios, and improving driving safety.

[0105] S420, determine the target warning area in the target image based on the vehicle's operating status.

[0106] The target warning area is the region in the target image that is associated with the occurrence of the accident.

[0107] The target warning area is the area with a high risk of accident occurrence in the collected panoramic images.

[0108] In one possible embodiment, please refer to Figure 5 , Figure 5 This is a flowchart illustrating a method for determining a target warning region in a target image, as provided in an embodiment of this application. Figure 5 The document illustrates the specific steps for determining the target warning area in the target image based on the vehicle's operating status, including:

[0109] S510, determine the warning direction based on the operating status.

[0110] Among them, the vehicle status, such as driving status, hovering status, and parking status, is determined by the vehicle speed sensor and gear position signal.

[0111] The warning direction refers to the focus area of ​​the warning analysis, used to filter key risk areas in the target image and avoid invalid calculations. For example, the warning direction for the driving condition is the side and rear; the warning direction for the hovering condition is the entire area around the door; and the warning direction for the parking condition is the rear and side and rear.

[0112] S520, determine the shape of the target warning area according to the warning direction.

[0113] Specifically, if the warning direction is to the side and rear, the area shape can be designed as a fan shape. For example, when a small passenger car is changing lanes, it can be a 30°-60° fan shape to the side and rear; when a large commercial vehicle is reversing, it can be a 60°-90° fan shape to the side and rear.

[0114] Preferably, the blind spot warning area of ​​the blind spot detection system (BDS) can be a 45° fan-shaped area on the side and rear of the vehicle.

[0115] Furthermore, if the warning direction is to the side and rear, the shape of the area can also be designed as a diagonal rectangle. For example, a small passenger car can form a diagonal rectangular area extending diagonally backward with the line connecting the door handles to the side and rear of the vehicle as the starting edge, forming an angle of 30°-45° with the side of the vehicle, thereby monitoring the driving path of objects to the side and rear.

[0116] For warning areas that include multiple shapes, the system can be determined by referring to vehicle model and user driving habits; it can also determine the area corresponding to each shape and determine the warning area with the largest area as the target warning area; or it can be selected by the user.

[0117] The door opening warning (DOW) system can be located within a 180° area to the side of the vehicle.

[0118] Specifically, if the warning direction is the entire area around the vehicle door, the shape of the area can be designed as a 180° semicircle on the side. For example, when a small passenger car is opening its door to let passengers in or out, a 180° semicircle area with a radius of 0m-5m is formed with the outer edge of the door as the center; around the passenger doors of a bus, a 180° semicircle area with a radius of 0m-8m is formed with the central axis of each passenger door as the reference.

[0119] Specifically, if the warning direction is directly below and in front of the vehicle, the area can be designed as a downward-convex arc shape to fit the spatial structure and blind spot distribution characteristics under the vehicle's front, achieving precise coverage of obstacles in front of the vehicle. For example, in low-speed driving conditions such as in residential areas or around schools, small passenger cars can form a downward-convex 120° arc centered on the midpoint of the lower edge of the grille to identify low-lying targets such as children and pets. For large trucks and other vehicles with higher front ends, in urban road starting and slow-moving conditions at intersections, a downward-convex 150° arc centered on the midpoint of the lower edge of the grille can be formed to accommodate the wider and deeper blind spot under the vehicle's front, avoiding the risk of being run over by low-lying obstacles due to the vehicle's height.

[0120] The warning direction can also be other directions, and the corresponding area shape can be adapted to the design.

[0121] S530, obtain the first type, first driving speed and driving conditions of the vehicle.

[0122] The first type refers to the vehicle type, and the first driving speed refers to the real-time driving speed of the vehicle.

[0123] Among them, driving conditions are used to characterize the environmental features and real-time status of the road where the vehicle is traveling. They are obtained through the collaboration of onboard cameras, radar and navigation maps, and specifically include basic road types, road condition complexity and real-time environmental parameters.

[0124] Among them, basic road types include urban roads, expressways, rural roads, tunnels, bridges, etc.; road condition complexity is used to reflect the risk intensity of traffic scenarios, specifically including traffic flow, congestion status, ramp density, etc.; real-time environmental parameters are used to characterize dynamic environmental factors that affect vehicle driving safety and sensor detection capabilities, including road surface adhesion, light intensity, weather conditions, visibility index, etc.

[0125] S540, the target warning area is determined based on the shape, the first type, the first driving speed, and the driving conditions.

[0126] In one possible embodiment, determining the target warning area based on the shape, the first type, the first driving speed, and the road conditions includes: determining a warning distance based on the first driving speed; determining a field of view width and a field of view angle based on the first type; determining an initial warning area based on the shape, the field of view width, the field of view angle, and the warning distance; determining a correction coefficient based on the road conditions; and adjusting the initial warning area based on the correction coefficient to obtain the target warning area.

[0127] The first driving speed is the vehicle's real-time speed, which can be obtained through the onboard speed sensor. The warning distance is a safe warning distance calculated based on driving speed and road conditions. The faster the vehicle speed, the longer the warning distance.

[0128] In one possible embodiment, a distance calculation model can be constructed based on driving speed, combined with vehicle braking performance, road conditions, and driver reaction time, to ultimately obtain the warning distance.

[0129] In one possible implementation, the driving scenario can be adapted to divide the vehicle into multiple speed ranges, with each speed range corresponding to a warning distance.

[0130] Among them, the field of view width is the visual coverage width of the vehicle in the warning direction, and the field of view angle is the fixed angle of view of the vehicle in the warning direction.

[0131] Different vehicle types correspond to different field of vision widths and angles. For example, small passenger cars have a field of vision width of 3 meters and a field of vision angle of 45° on each side; large commercial vehicles have a field of vision width of 7 meters and a field of vision angle of 70° on each side. These can be adapted and modified according to the driving scenario to ensure that the field of vision parameters match real-time risk requirements.

[0132] In one possible embodiment, determining the initial warning area based on the shape, the field of view width, the field of view angle, and the warning distance includes: determining a second type of the shape, the second type including fan-shaped and oblique rectangular shapes; if the second type is fan-shaped, determining the radius of the fan-shaped shape based on the field of view width and the warning distance; determining the initial warning area based on the field of view angle and the radius; if the second type is oblique rectangular shape, determining the horizontal dimension based on the field of view width; determining the vertical dimension based on the warning distance; determining the oblique angle based on the field of view angle; and determining the initial warning area based on the horizontal dimension, the vertical dimension, and the oblique angle.

[0133] If the shape is fan-shaped, it is necessary to ensure that the coverage area of ​​the fan meets both the horizontal coverage requirements corresponding to the field of view width and the vertical safety requirements corresponding to the warning distance, so as to avoid missing risks due to the small radius or causing false warnings due to the large radius.

[0134] The product of the radius and sin (field of view angle) must be greater than or equal to half the field of view width. The minimum value of the radius can then be calculated and compared with the warning distance. The maximum value is taken as the radius of the sector.

[0135] Among them, the field of vision angle can be directly used as the central angle of the sector. With the vehicle's preset reference point as the center, such as the center of the grille when a forward warning is issued, the sector area is delineated by combining the determined radius and central angle to obtain the initial warning area.

[0136] Among them, the diagonal design in the diagonal rectangle type is to match the risk propagation path of the warning direction and ensure that the coverage area is in sync with the distribution of blind spots; the long axis of the rectangle, that is, the vertical dimension, corresponds to the risk coverage distance, and the short axis of the rectangle, that is, the horizontal dimension, corresponds to the horizontal coverage width. The boundaries are clear and can avoid redundancy in non-risk areas.

[0137] If the shape is a slanted rectangle, the field of view width can be directly used as the horizontal dimension of the slanted rectangle, the warning distance can be directly used as the vertical dimension, and the slanted angle can be used as the slanted angle. Starting from a preset reference point on the vehicle (e.g., the midpoint of the line connecting the door handles for side and rear warning), extend the vertical dimension along the first direction of the slanted angle and extend the horizontal dimension along the second direction of the slanted angle to form the slanted rectangle. The first and second directions together constitute the slanted angle and coordinate with the warning direction.

[0138] In one possible embodiment, the second type also includes composite shapes, which are combinations or scene-specific variations based on a single form.

[0139] If the second type is a trapezoid adapted for curved scenarios, then the vehicle's steering angle, wheelbase, and curve radius are obtained. Based on the field of view width, wheelbase, and inner wheel difference, the upper and lower base dimensions of the trapezoid are determined. Based on the warning distance and curve radius, the longitudinal dimension of the trapezoid is determined. Based on the field of view angle and steering angle, the tilt direction and side angle of the trapezoid are determined. Taking the outer side of the vehicle's rear wheel as the starting point of the lower base of the trapezoid, the longitudinal dimension is extended along the tilt direction. Based on the upper and lower base dimensions and side angle, the trapezoid is drawn in the image coordinate system to obtain the initial warning area.

[0140] If the second type is a composite shape, such as a combination of a fan shape and a diagonal rectangle, which is suitable for the multi-dimensional risks of intersections, the composite direction can be split into multiple independent sub-directions, a single shape can be assigned to each sub-direction, and sub-region parameters can be calculated. Finally, each sub-region is superimposed in the same coordinate system, the overlapping part is removed, the coverage is ensured to be without blind spots, and the initial warning area after fusion is output.

[0141] Specifically, the mapping relationship between road conditions, vehicle type, and correction coefficients is pre-stored in the control chip, and it is calibrated based on real vehicle road test data. The correction coefficient is automatically matched according to the current road conditions and vehicle type.

[0142] The correction factors include longitudinal correction factors, and / or lateral correction factors and / or oblique angle correction factors, or angle correction factors and / or radius correction factors.

[0143] After obtaining the initial warning area, if the initial warning area is fan-shaped, the angle correction coefficient and / or radius correction coefficient are obtained, and the initial warning area is dynamically adjusted to obtain the target warning area.

[0144] If the initial warning area is a diagonal rectangle, then the longitudinal correction coefficient and / or the lateral correction coefficient and / or the diagonal angle correction coefficient are obtained, and the initial warning area is dynamically adjusted to obtain the target warning area.

[0145] As can be seen, in this embodiment, the warning direction and form type are determined by combining vehicle type and operating status, the initial area is quantified based on field of vision parameters and warning distance, and finally dynamically adjusted with road condition correction coefficient. This not only allows for flexible adaptation to different vehicle types, operating conditions and road conditions, but also optimizes system computing power consumption while ensuring driving safety. It achieves full-dimensional dynamic adaptation of warning area with vehicle characteristics, driving scenarios and road condition risks, greatly improving the accuracy and robustness of system warnings.

[0146] S430, classify the reference moving objects in the target warning area to obtain the classification result.

[0147] The reference moving object could be a vehicle, pedestrian, or animal. The system detects and outputs the moving object in the target image, displaying it synchronously on the screen.

[0148] The process involves using a classification model to categorize reference moving objects within the target warning area, yielding classification results. These results include categories such as cars, buses, trucks, pedestrians, bicycles, tricycles, and animals. The classification results may also include the location information of the moving objects.

[0149] S440, determine the ranging range of the reference moving object based on the classification result.

[0150] Each classification result corresponds to a different ranging range, which can be determined based on the object's speed and the road speed limit.

[0151] For example, the distance measurement range for pedestrians is 0m-20m, for two-wheeled vehicles it is 0m-35m, for three-wheeled vehicles it is 0m-40m, and for cars it is 0m-60m, etc.

[0152] S450, the reference moving object is filtered according to the ranging range to obtain the target moving object.

[0153] The process involves calculating the actual distance between each reference moving object and vehicle. If the actual distance is within the corresponding ranging range, it is identified as a target moving object; otherwise, it is directly excluded and marked as a low-risk object outside the ranging range. This process filters out low-risk targets to reduce false alarms and prevents missed risk assessments due to fixed ranging ranges.

[0154] S460, determine the target risk level of the target moving object.

[0155] In one possible embodiment, please refer to Figure 6 , Figure 6 This is a flowchart of a method for determining the target risk level of a moving object, as provided in an embodiment of this application. Figure 6 As shown, the specific steps include:

[0156] S610, determine the target distance between the target moving object and the vehicle based on the intrinsic and extrinsic parameters of the image acquisition device.

[0157] The intrinsic parameters are the inherent parameters of the image acquisition device, including focal length, pixel size, principal point coordinates, and distortion coefficients; the extrinsic parameters are the installation parameters of the image acquisition device relative to the vehicle coordinate system, including installation position and installation attitude. The intrinsic and extrinsic parameters of the image acquisition device can be determined through calibration algorithms.

[0158] The intrinsic parameters are used to convert image pixels into 3D points in the camera coordinate system, and the extrinsic parameters are used to convert the points in the camera coordinate system into points in the world coordinate system, thereby obtaining the position of the object relative to the vehicle, and then calculating the target distance between the object and the vehicle.

[0159] S620, determine the second speed of the target moving object.

[0160] Specifically, based on two consecutive image frames with a time interval Δt, the pixel coordinates of the target are extracted using a target detection algorithm. Combined with the intrinsic and extrinsic parameters of the image acquisition device, the pixel displacement is converted into the actual physical displacement ΔS, and the real-time speed is calculated. For example, this can be calculated multiple times, and the average speed is taken as the current driving speed, i.e., the second driving speed.

[0161] S630, the target warning area is divided into multiple warning sub-areas based on the degree of danger.

[0162] Among them, target warning zones can be divided according to distance priority. The closer the target is to the vehicle, the shorter the collision reaction time and the higher the degree of danger.

[0163] In one possible embodiment, please refer to Figure 7 , Figure 7 This is a schematic diagram of a target image provided in an embodiment of this application, such as... Figure 7 As shown, this image is a target image collected in an urban road environment. The background of the image includes urban high-rise buildings, road fences, trees and other infrastructure. There are vehicles driving on the road. The vehicles in this image are low-risk moving objects identified by target detection and are marked by green boxes. They are currently near the blue warning area. At the same time, diagonal rectangular target warning areas are constructed by red, yellow and blue lines. Different colors correspond to different warning sub-areas, which follow the priority logic that the closer to the vehicle, the higher the danger level. Among them, the red area has the highest danger level, the yellow area is the second highest, and the blue area has the lowest danger level.

[0164] S640, determine the warning sub-region where the target moving object is located.

[0165] S650, determine the target risk level of the moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed.

[0166] In one possible embodiment, determining the target risk level of the target moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed includes: determining a first risk coefficient of the warning sub-area where the target moving object is located; determining a second risk coefficient of the target moving object based on the second driving speed and the first driving speed of the vehicle; determining a third risk coefficient of the target moving object based on the target distance; and determining the target risk level based on the first risk coefficient, the second risk coefficient, and the third risk coefficient.

[0167] This involves determining the warning sub-region where each moving target is located and obtaining the risk coefficient of that sub-region, i.e., the first risk coefficient. For example, the risk coefficient for a high-risk area is 1.0, for a medium-risk area it is 0.6, and for a low-risk area it is 0.4.

[0168] The distance risk coefficient is determined based on the target distance. Specifically, it can be calculated based on the warning distance and the minimum safe distance. First, the first difference between the target distance and the minimum safe distance is calculated. Then, the second difference between the warning distance and the target distance is calculated. Finally, the ratio of the first difference to the second difference is calculated, and the difference between 1 and this ratio is the second risk coefficient.

[0169] The relative speed is calculated based on the first and second driving speeds. If the object and the vehicle are traveling in opposite directions, the relative speed is the sum of the first and second driving speeds. If the object and the vehicle are traveling in the same direction, the relative speed is the difference between the first and second driving speeds. If the object is traveling laterally relative to the vehicle, the relative speed is the sum of the square of the first and second driving speeds, and then the square root of the sum.

[0170] The ratio of relative speed to the maximum safe relative speed is defined as the third risk coefficient.

[0171] The weight of each risk coefficient can be determined using real vehicle data, and then a weighted sum can be performed to obtain the final risk value, thereby determining the corresponding target risk level.

[0172] The weighting coefficients can be adaptively adjusted according to the driving scenario. For example, the speed weight can be increased in high-speed scenarios; the area weight can be increased in urban road scenarios; and the distance weight can be increased in rainy or snowy weather scenarios.

[0173] As can be seen, in this embodiment, a multi-dimensional risk assessment system is established. By dividing the warning area according to the degree of danger and dynamically adapting it to the driving scenario, a differentiated spatial risk framework is constructed. Then, distance and speed parameters are used to improve the dynamic dimensions of risk assessment. Finally, the collaborative assessment of multi-dimensional parameters is achieved by merging them to output an accurate target risk level. At the same time, the weights can be adaptively adjusted according to the driving scenario to make the risk level assessment more in line with the actual working conditions, thereby enhancing the rationality and pertinence of the assessment.

[0174] In one possible embodiment, the risk level can also be determined by combining the trajectory overlap rate. Based on the vehicle's current driving trajectory, the probability of overlap between the target moving object and the vehicle's trajectory is determined. The higher the overlap probability, the higher the degree of danger, and the higher the corresponding risk level.

[0175] S470, issue a warning to the vehicle based on the target risk level, and mark the target moving object and the target risk level on the target image displayed on the screen.

[0176] Specifically, based on the preset early warning strategy rule base, the system analyzes the early warning trigger conditions and early warning channel combinations corresponding to the risk level.

[0177] Warnings can include auditory warnings, tactile warnings, and visual warnings.

[0178] Specifically, auditory warnings include voice prompts, such as a voice announcement saying "Emergency collision risk ahead, please brake immediately" at high-risk levels. The higher the risk level, the louder the voice announcement, which can be adaptively adjusted to reflect the current ambient noise. Different risk levels correspond to different warning audio frequencies; for example, high-risk levels use a high-frequency continuous beep, medium-risk levels use a medium-frequency intermittent beep, and low-risk levels use a low-frequency warning tone.

[0179] Specifically, tactile warnings can include seat vibrations, with higher risk levels corresponding to stronger vibration intensity and longer duration.

[0180] Specifically, visual warnings can use methods such as screen flashing and pop-up prompts. For example, high-risk levels use full-screen flashing and pop-up prompts, medium-risk levels use partial flashing, and low-risk levels use text prompts.

[0181] For example, a high-risk level could be a rearview mirror warning light that is constantly on, along with a voice prompt and a red warning on the central control screen; a low-risk level could be a rearview mirror warning light that flashes and a prompt on the central control screen.

[0182] Among these measures, the system controls all early warning channels to trigger synchronously according to preset parameters, ensuring multi-sensory collaborative perception for the driver while guaranteeing the consistency of the early warning sequence.

[0183] The target image is displayed on the screen simultaneously, and the target image is marked with the target moving object and the target risk level.

[0184] The attributes and risk level of a moving object can be marked using a combination of text and icons. For example, high-risk objects are marked with a red warning box and a triangle icon with text prompts; medium-risk objects are marked with a yellow warning box and an exclamation mark icon with the text prompt "Observe carefully"; and low-risk objects are marked with a blue warning box and an arrow pointing to the direction of the object, along with the text "Safe distance".

[0185] As can be seen, in this embodiment, by dynamically defining the warning area that matches the real-time vehicle status, the detection range can be narrowed and the scene adaptability can be enhanced. On this basis, a differentiated ranging range is matched according to the moving object classification results, and the moving objects are filtered through this range. This can filter out low-risk targets to reduce false alarms and prevent risk omissions caused by fixed ranging ranges. Finally, by combining the target risk level determination and the visual marker output, a precise response to blind spot warnings can be achieved, thereby improving the accuracy and reliability of blind spot warnings.

[0186] In one possible embodiment, the display screen may include multiple display modules. For example, the left display screen is used to display images of the front and / or left side of the vehicle, and the right display screen is used to display images of the rear and / or right side of the vehicle. The display layout of these left, right, front, and rear images on the display screen corresponds to a spatial orientation mapping.

[0187] The proportion of each image on the display screen can be allocated according to the current driving scenario. For example, in a reversing scenario, the rear image is displayed first on the right-hand display screen, while the right blind spot image serves as auxiliary information and can be overlaid or displayed in a small window.

[0188] In one possible embodiment, the content displayed on the screen can be adjusted to match the driver's viewing angle.

[0189] Specifically, the system collects seat adjustment parameters in real time via built-in position sensors, including fore-and-aft adjustment distance, height adjustment, and backrest tilt angle. Based on these parameters and the driver's height, the system determines the driver's viewing position. Then, it performs viewing angle transformation, distance scaling calibration, and misalignment compensation on the target images on the left and right screens to ensure the image matches the viewing angle.

[0190] As can be seen, in this embodiment, the viewing angle of the left display screen is adapted to the driver's left-side viewing angle, so that the image of the left blind spot is presented in a proportion that is more in line with the driver's intuitive perception; the image of the right display screen will be adapted to the driver's viewing angle when looking to the right, avoiding visual misalignment caused by side viewing.

[0191] In one possible embodiment, the images displayed on the left and right displays can be adjusted to suit the driver's visual habits, taking into account the driver's cockpit position and current seating posture.

[0192] The driver's posture and position data are acquired in real time through multiple sensors in the cockpit. A driver's visual feature model is established based on the posture and position data to calculate the effective viewing distance, viewing angle range, and visual distortion compensation coefficient of the left and right screens relative to the driver. Combined with human visual habits, the adaptation priority of the left and right screens is determined, such as the left screen focusing on information readability and the right screen focusing on scene visualization.

[0193] Based on the data output by the model and the adaptation priority of the left and right screens, the image parameters of the left and right screens are adjusted. Specifically, the image scaling ratio, font size, layout offset, brightness or contrast of the left screen can be adjusted. The layout offset is used to adapt to head side movement, and the brightness or contrast is used to adapt to the viewing angle.

[0194] Among these features, it can correct image distortion on the right side screen to compensate for perspective deviation caused by seat tilt, adjust the content display area of ​​the right side screen to match the driver's line of sight, and optimize the color saturation of the right side screen to adapt to differences in visual sensitivity caused by changes in viewing distance.

[0195] In one possible embodiment, different information with different emphases are superimposed on different displays. For example, the left display focuses on showing the distance markers to the left edge of the vehicle, as the driver is closer to the left and is more sensitive to distance. The right display focuses on showing the predicted trajectory of moving objects within the warning area, as the viewing angle on the right is limited and requires clearer dynamic prompts.

[0196] In one possible embodiment, the content displayed on the screen can also be adapted and adjusted according to the driver's driving operation. For example, if the driving operation is detected as turning on the left turn signal, the left side screen will automatically enlarge the left rear area; if the driving operation is detected as turning the steering wheel, the corresponding side screen will display tire trajectory auxiliary lines to enhance spatial perception.

[0197] For examples consistent with the above embodiments, please refer to... Figure 8 , Figure 8 This is a functional unit block diagram of a blind spot warning device based on an electronic rearview mirror provided in an embodiment of this application, such as... Figure 8As shown, the blind spot warning device 80 based on the electronic rearview mirror includes: an acquisition unit 81, used to acquire at least one original image in a predetermined acquisition direction through the image acquisition device, and preprocess the original image to obtain a target image; a first determination unit 82, used to determine a target warning area in the target image according to the vehicle's operating status, wherein the target warning area is an area in the target image that is associated with the occurrence of an accident; a classification unit 83, used to classify reference moving objects in the target warning area to obtain a classification result; a second determination unit 84, used to determine the distance range of the reference moving object according to the classification result; a filtering unit 85, used to filter the reference moving object according to the distance range to obtain a target moving object; a third determination unit 86, used to determine the target risk level of the target moving object; and a warning unit 87, used to issue a warning to the vehicle according to the target risk level, and mark the target moving object and the target risk level on the target image displayed on the screen.

[0198] In one possible embodiment, in determining the target warning area in the target image based on the vehicle's operating status, the first determining unit 82 is specifically configured to: determine the warning direction based on the operating status; determine the shape of the target warning area based on the warning direction; acquire a first type, a first driving speed, and driving conditions of the vehicle; and determine the target warning area based on the shape, the first type, the first driving speed, and the driving conditions.

[0199] In one possible embodiment, in determining the target warning area based on the shape, the first type, the first driving speed, and the driving conditions, the first determining unit 82 is further configured to: determine a warning distance based on the first driving speed; determine a field of view width and a field of view angle based on the first type; and determine an initial warning area based on the warning distance; determine a correction coefficient based on the driving conditions; and adjust the initial warning area based on the correction coefficient to obtain the target warning area.

[0200] In one possible embodiment, in determining the initial warning area based on the shape, the field of view width, the field of view angle, and the warning distance, the first determining unit 82 is further configured to: determine a second type of the shape, the second type including fan-shaped and oblique rectangular shapes; if the second type is fan-shaped, determine the radius of the fan-shaped shape based on the field of view width and the warning distance; determine the initial warning area based on the field of view angle and the radius; if the second type is oblique rectangular shape, determine the horizontal dimension based on the field of view width; determine the vertical dimension based on the warning distance; determine the oblique angle based on the field of view angle; and determine the initial warning area based on the horizontal dimension, the vertical dimension, and the oblique angle.

[0201] In one possible embodiment, in determining the target risk level of the target moving object, the third determining unit 86 is specifically configured to: determine the target distance between the target moving object and the vehicle based on the intrinsic and extrinsic parameters of the image acquisition device; determine the second driving speed of the target moving object; divide the target warning area into multiple warning sub-areas based on the degree of danger; determine the warning sub-area where the target moving object is located; and determine the target risk level of the target moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed.

[0202] In one possible embodiment, in determining the target risk level of the target moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed, the third determining unit 86 is further configured to: determine a first risk coefficient of the warning sub-area where the target moving object is located; determine a second risk coefficient of the target moving object based on the second driving speed and the first driving speed of the vehicle; determine a third risk coefficient of the target moving object based on the target distance; and determine the target risk level based on the first risk coefficient, the second risk coefficient, and the third risk coefficient.

[0203] In one possible embodiment, before acquiring at least one original image in a predetermined acquisition direction through the image acquisition device and preprocessing the original image to obtain the target image, the blind spot warning device 80 based on the electronic rearview mirror is further configured to: acquire a first type of the vehicle; determine the current driving scenario of the vehicle based on the operating state; and determine the at least one predetermined acquisition direction based on the current driving scenario and the first type.

[0204] It is understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the method embodiment section in this application should be adapted to the device embodiment section in a synchronous manner, and will not be repeated here.

[0205] In the case of using integrated units, please refer to Figure 9 , Figure 9 This is a functional unit block diagram of another blind spot warning device based on an electronic rearview mirror provided in this application embodiment, such as... Figure 9 As shown, the blind spot warning device 80 based on the electronic rearview mirror includes a processing module 802 and a communication module 801. The processing module 802 controls and manages the operation of the blind spot warning device 80, for example, executing the steps of the acquisition unit 81, the first determination unit 82, the classification unit 83, the second determination unit 84, the filtering unit 85, the third determination unit 86, and the warning unit 87, and / or performing other processes of the technology described herein. The communication module 801 is used for interaction between the blind spot warning device 80 and other devices.

[0206] Among them, such as Figure 9 As shown, the blind spot warning device 80 based on the electronic rearview mirror may also include a storage module 803, which is used to store the program code and data of the blind spot warning device 80 based on the electronic rearview mirror.

[0207] The processing module 802 can be a processor or controller, such as a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0208] The communication module 801 can be a transceiver, RF circuit, or communication interface, etc. The storage module 803 can be a memory.

[0209] All relevant content for each scenario involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here. The above-mentioned blind spot warning device 80 based on electronic rearview mirror can perform the above... Figure 4 The blind spot warning method based on electronic rearview mirror is shown.

[0210] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application, as shown below. Figure 10 As shown, the electronic device 1000 includes a processor 1010, a memory 1020, a communication interface 1030, and one or more programs 1021. The one or more programs 1021 are stored in the memory and configured to be executed by the processor. When the program is executed, it includes some or all of the steps of any blind spot warning method based on electronic rearview mirror described in the above method embodiments. The processor, memory, and communication interface are interconnected and complete communication between them.

[0211] The memory can be volatile memory such as Dynamic Random Access Memory (DRAM) or non-volatile memory such as a hard disk drive. The memory stores a set of executable program code, and the processor calls the executable program code stored in the memory to execute some or all of the steps of any of the blind spot warning methods based on electronic rearview mirrors described in the above embodiments of the blind spot warning method based on electronic rearview mirrors.

[0212] As can be seen, the electronic device 1000 described in this application embodiment first acquires at least one original image in a predetermined acquisition direction through the image acquisition device, and preprocesses the original image to obtain a target image; then, it determines a target warning area in the target image based on the vehicle's operating status, the target warning area being an area in the target image associated with the occurrence of an accident; then, it classifies reference moving objects in the target warning area to obtain a classification result; next, it determines the ranging range of the reference moving objects based on the classification result; then, it filters the reference moving objects based on the ranging range to obtain target moving objects; next, it determines the target risk level of the target moving objects; finally, it issues a warning to the vehicle based on the target risk level, and marks the target moving objects and the target risk level on the target image displayed on the screen.

[0213] This application narrows the detection range and enhances scene adaptability by dynamically defining the warning area that matches the real-time vehicle status. Based on this, it matches a differentiated ranging range according to the moving object classification results and filters moving objects through this range. This not only filters out low-risk targets to reduce false alarms but also prevents missed risk detection caused by a fixed ranging range. Finally, by combining target risk level determination with visual label output, it achieves accurate response to blind spot warnings, thereby improving the accuracy and reliability of blind spot warnings.

[0214] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0215] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0216] It should be noted that, for the sake of simplicity, the aforementioned methods are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are optional, and the actions and modules involved are not necessarily essential to this application.

[0217] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0218] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units 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 devices or units may be electrical or other forms.

[0219] The units described 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, depending on actual needs.

[0220] 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 program module.

[0221] If the integrated unit is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). 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 memory and includes several 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 described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0222] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk, etc.

[0223] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The above description of the embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A blind spot warning method based on an electronic rearview mirror, characterized in that, A control chip is used in an electronic rearview mirror system, the electronic rearview mirror system further including an image acquisition device installed outside the vehicle and a display screen installed inside the vehicle, the control chip being connected to the image acquisition device and the display screen respectively, the method comprising: The image acquisition device acquires at least one original image in a predetermined acquisition direction, and preprocesses the original image to obtain the target image. The target warning area in the target image is determined based on the vehicle's operating status. The target warning area is the region in the target image associated with the accident. The warning direction is determined based on the operating status. The shape of the target warning area is determined based on the warning direction. The shape includes a fan shape, a diagonal rectangle, a downwardly convex arc, and a composite shape, where the composite shape is a combination or scene-specific deformation based on a single shape. The vehicle's first type, first driving speed, and road conditions are obtained. The target warning area is determined based on the shape, the first type, the first driving speed, and the road conditions. The warning distance is determined based on the first driving speed. The field of view width and field of view angle are determined based on the first type. Referring to the shape, an initial warning area is determined based on the field of view width, the field of view angle, and the warning distance. A correction coefficient is determined based on the road conditions. The initial warning area is adjusted based on the correction coefficient to obtain the target warning area. If the target warning area has multiple shapes, the area of ​​a candidate warning area corresponding to each shape is determined, and the candidate warning area with the largest area is determined as the target warning area. The reference moving objects in the target warning area are classified to obtain the classification results; The ranging range of the reference moving object is determined based on the movement speed and road speed limit corresponding to each category in the classification results; The reference moving objects are filtered according to the ranging range to obtain the target moving object; Determine the target risk level of the moving object; The vehicle is given a warning based on the target risk level, and the target moving object and the target risk level are marked on the target image displayed on the screen.

2. The method according to claim 1, characterized in that, The method of determining the initial warning area based on the shape, the field of view width, the field of view angle, and the warning distance includes: A second type of the shape is determined, the second type including sector-shaped and oblique rectangular shapes; If the second type is a fan shape, the radius of the fan shape is determined based on the field of view width and the warning distance; The initial warning area is determined based on the field of view angle and the radius; If the second type is the oblique rectangular shape, the horizontal dimension is determined according to the field of view width; The longitudinal dimension is determined based on the warning distance; The oblique angle is determined based on the stated field of view angle; The initial warning area is determined based on the horizontal dimension, the vertical dimension, and the diagonal angle.

3. The method according to claim 1, characterized in that, Determining the target risk level of the moving object includes: Based on the intrinsic and extrinsic parameters of the image acquisition device, the target distance between the moving target and the vehicle is determined; Determine the second speed of the target moving object; The target warning area is divided into multiple warning sub-areas based on the degree of danger. Determine the warning sub-region where the target moving object is located; The target risk level of the moving object is determined based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed.

4. The method according to claim 3, characterized in that, The step of determining the target risk level of the moving object based on the warning sub-area where the target moving object is located, the target distance, and the second driving speed includes: Determine the first risk coefficient of the warning sub-region where the target moving object is located; A second risk factor for the target moving object is determined based on the second driving speed and the first driving speed of the vehicle. A third risk factor for the moving target object is determined based on the target distance; The target risk level is determined based on the first risk coefficient, the second risk coefficient, and the third risk coefficient.

5. The method according to any one of claims 1-4, characterized in that, Before acquiring at least one raw image in a predetermined acquisition direction using the image acquisition device, and preprocessing the raw image to obtain the target image, the method further includes: Obtain the first type of the vehicle; The current driving scenario of the vehicle is determined based on the operating status; The at least one predetermined acquisition direction is determined based on the current driving scenario and the first type.

6. A blind spot warning device based on an electronic rearview mirror, characterized in that, A control chip for an electronic rearview mirror system, the electronic rearview mirror system including an image acquisition device installed outside the vehicle and a display screen installed inside the vehicle, the control chip being connected to the image acquisition device and the display screen respectively, the device comprising: The acquisition unit is used to acquire at least one original image in a predetermined acquisition direction through the image acquisition device, and to preprocess the original image to obtain a target image; A first determining unit is configured to determine a target warning region in the target image based on the vehicle's operating status, wherein the target warning region is an area in the target image associated with an accident; specifically, it is configured to determine a warning direction based on the operating status; determine the shape of the target warning region based on the warning direction, wherein the shape includes a fan shape, a diagonal rectangle-like shape, and a composite shape, wherein the composite shape is a combination or scene-specific deformation based on a single shape; acquire a first type, a first driving speed, and driving conditions of the vehicle; determine the target warning region based on the shape, the first type, the first driving speed, and the driving conditions; wherein the first determining unit is further configured to determine a warning distance based on the first driving speed; determine a field of view width and a field of view angle based on the first type; determine an initial warning region based on the shape, the field of view width, the field of view angle, and the warning distance; determine a correction coefficient based on the driving conditions; adjust the initial warning region based on the correction coefficient to obtain the target warning region; wherein, if the target warning region has multiple shapes, determine the area of ​​a candidate warning region corresponding to each shape, and determine the candidate warning region with the largest area as the target warning region; A classification unit is used to classify reference moving objects in the target warning area and obtain classification results; The second determining unit is used to determine the ranging range of the reference moving object based on the movement speed and road speed limit corresponding to each category in the classification results; A filtering unit is used to filter the reference moving object according to the ranging range to obtain the target moving object; The third determining unit is used to determine the target risk level of the target moving object; The warning unit is used to issue a warning to the vehicle based on the target risk level, and to mark the target moving object and the target risk level on the target image displayed on the screen.

7. An electronic device, characterized in that, The device includes: The system includes a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code to perform the steps of the blind spot warning method based on an electronic rearview mirror as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores executable program code, which includes execution instructions for performing the steps of the blind spot warning method based on an electronic rearview mirror as described in any one of claims 1-5.

Images