Vehicle control method, electronic device, and vehicle

Abstract

The application provides a vehicle control method, an electronic device and a vehicle, and applies to the technical field of intelligent vehicle control, and the method comprises the following steps: acquiring distortion information of a monitoring image, determining a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and performing target detection on the monitoring image; in response to detecting a target, determining at least one target region where the target is located; based on the target region where the target is located, determining the relative position of the target and the vehicle to determine whether the target invades the warning area. The application reduces the error caused by the distortion of the monitoring image, and improves the accuracy and reliability of target invasion detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle intelligent control, and in particular to a vehicle control method, an electronic device and a vehicle. BACKGROUND

[0002] In recent years, more and more automobile brands begin to be equipped with intrusion detection functions, which monitor the abnormal situation around the vehicle in real time after the vehicle is parked and turned off through the vehicle-mounted camera and other sensors, and record the abnormal video. However, the traditional intrusion detection function mainly relies on the image captured by the fisheye camera for monitoring and judgment, but the image captured by the fisheye camera has large distortion, and the target intrusion judgment has large error. SUMMARY

[0003] Therefore, the present application aims to provide a vehicle control method, an electronic device and a vehicle to avoid large error in target intrusion judgment caused by image distortion.

[0004] To achieve the above purpose, the present application provides a vehicle control method, which comprises:

[0005] obtaining distortion information of the monitoring image, determining a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and performing target detection on the monitoring image;

[0006] in response to detecting a target, determining at least one target region where the target is located;

[0007] based on the target region where the target is located, determining the relative position of the target and the vehicle to judge whether the target intrudes into the warning region.

[0008] Further, the method further comprises:

[0009] in response to detecting a target, determining the area of the target bounding box;

[0010] in response to determining that the area is less than or equal to a preset area, determining the relative distance of the target and the vehicle, and in response to determining that the relative distance is less than or equal to a preset distance, determining that the target intrudes into the warning region.

[0011] By calculating the area of the target bounding box, the distance of the target can be preliminarily judged. A larger bounding box area usually indicates that the target is closer to the camera, while a smaller bounding box area indicates that the target is farther away. Using the area of the target bounding box as an auxiliary judgment basis improves the accuracy and reliability of target position judgment. According to the size of the area, a suitable detection method is selected to quickly judge whether the target intrudes. The accuracy and reliability of the detection are significantly improved, the false positives and false negatives are reduced, and the real-time response capability and user experience are enhanced.

[0012] Based on the same inventive concept, the application further provides a vehicle control device, comprising:

[0013] an acquisition module configured to acquire distortion information of the monitoring image, determine a plurality of target regions with different degrees of distortion in the monitoring image based on the distortion information, and perform target detection on the monitoring image;

[0014] a judgment module configured to determine at least one target region where the target is located in response to detecting the target;

[0015] a detection module configured to determine the relative position of the target and the vehicle based on the target region where the target is located, and determine whether the target intrudes into the alert region.

[0016] Based on the same inventive concept, the application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method described above when executing the computer program.

[0017] Based on the same inventive concept, the application further provides a vehicle, comprising the electronic device described above.

[0018] As can be seen from the above, the vehicle control method, electronic device, and vehicle provided by the application, wherein the method comprises: acquiring distortion information of the monitoring image, determining a plurality of target regions with different degrees of distortion in the monitoring image based on the distortion information, and performing target detection on the monitoring image; determining at least one target region where the target is located in response to detecting the target; determining the relative position of the target and the vehicle based on the target region where the target is located, and determining whether the target intrudes into the alert region. Due to the characteristics of the fisheye camera, the image distortion is more serious at the edge part of the monitoring image, resulting in a larger error in position determination. Based on the distortion information of the monitoring image, a plurality of target regions with different degrees of distortion are determined in the monitoring image, and different methods of determining whether the target intrudes into the alert region are selected based on the target region where the target is located, thereby reducing the error caused by the image distortion of the fisheye camera, accurately detecting whether the target intrudes, preventing common security threats such as vehicle scratching, door opening accidents, and theft of property, and detecting other potential abnormal situations such as suspicious persons loitering around the vehicle, thereby improving the accuracy and reliability of target intrusion detection. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the application or related art, the following will briefly introduce the drawings needed to be used in the embodiments or related art description. Obviously, the drawings in the following description are only embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 Flow chart of the vehicle control method of the embodiment of the present application;

[0021] Figure 2 Monitoring image one in the vehicle control method of the embodiment of the present application;

[0022] Figure 3 Monitoring image two in the vehicle control method of the embodiment of the present application;

[0023] Figure 4 Monitoring image three in the vehicle control method of the embodiment of the present application;

[0024] Figure 5 Monitoring image four in the vehicle control method of the embodiment of the present application;

[0025] Figure 6 Monitoring image five in the vehicle control method of the embodiment of the present application;

[0026] Figure 7 Monitoring image six in the vehicle control method of the embodiment of the present application;

[0027] Figure 8 Monitoring image seven in the vehicle control method of the embodiment of the present application;

[0028] Figure 9 Monitoring image eight in the vehicle control method of the embodiment of the present application;

[0029] Figure 10 Monitoring image nine in the vehicle control method of the embodiment of the present application;

[0030] Figure 11 Schematic diagram of the vehicle control device of the embodiment of the present application;

[0031] Figure 12 An electronic device hardware structure schematic diagram provided by the embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to specific embodiments and the accompanying drawings.

[0033] It should be noted that, unless otherwise defined, technical terms or scientific terms used in the embodiments of the present application shall have the common meaning understood by one of ordinary skill in the art to which the embodiments of the present application belong. The terms "first", "second", and similar terms used in the embodiments of the present application do not represent any order, number, or importance, but are only used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects before the terms encompass the elements or objects listed after the terms and their equivalents, without excluding other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like are only used to represent relative positional relationships, which can change accordingly when the absolute positions of the described objects change.

[0034] In the related art, with the rapid development of intelligent automobile technology, the vehicle safety monitoring system has become an important part of improving driving safety and user experience. In recent years, more and more automobile brands have begun to install intrusion detection functions, through the 360-degree fisheye camera and other sensors on the vehicle, to monitor abnormal situations around the vehicle in real time after the vehicle is parked and turned off, and to record abnormal videos. This function not only effectively prevents vehicle scratching, door opening accidents, theft of property and other events, but also provides important evidence support. However, the traditional intrusion detection function mainly relies on the image of the 360-degree fisheye camera for monitoring and judgment, but due to the large image distortion of the fisheye camera, simply relying on the image for target detection and intrusion judgment in the warning area has a large error. This error can lead to missed identification and misidentification of abnormal situations, affecting the accuracy and reliability of the intrusion detection function.

[0035] Based on the above problems, the applicant found that: obtaining distortion information of the monitoring image, determining a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and performing target detection on the monitoring image; in response to detecting a target, determining at least one target region where the target is located; based on the target region where the target is located, determining the relative position of the target and the vehicle to judge whether the target intrudes into the warning area. Due to the characteristics of the fisheye camera, the image distortion is more serious at the edge of the monitoring image, resulting in a large error in position judgment. Based on the distortion information of the monitoring image, a plurality of target regions with different distortion degrees are determined in the monitoring image, and different methods for judging whether the target intrudes into the warning area are selected based on the target region where the target is located, which reduces the error caused by the image distortion of the fisheye camera, to accurately detect whether the target intrudes, prevent common security threats such as vehicle scratching, door opening accidents, theft of property, and detect other potential abnormal situations such as suspicious persons loitering around the vehicle, thereby improving the accuracy and reliability of target intrusion detection.

[0036] Embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0037] The present application provides a vehicle control method, such as Figure 1 As shown in some embodiments, the method is executed by a vehicle controller or a data processor set independently of the vehicle controller, and subsequent embodiments are exemplarily illustrated with the vehicle controller; a camera is arranged on the rearview mirror of the vehicle to collect a monitoring image outside the vehicle; the monitoring image includes a warning area; the method comprises:

[0038] S101, obtaining distortion information of the monitoring image, determining a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and performing target detection on the monitoring image;

[0039] In specific implementation, a camera is installed on the rearview mirror of the vehicle to collect a monitoring image around the vehicle in real time. The camera type provided by the vehicle is usually a 360-degree fisheye camera, which can provide a full range of view. According to the distortion information, the monitoring image can be divided into a plurality of target regions with different distortion degrees, for example, the central region of the image may have less distortion, while the edge region may have more distortion. The division can be based on a quantitative indicator of the distortion degree, such as the numerical range of the distortion parameter, or by monitoring the area ratio of the distorted image in the monitoring image, for example, when a region is a region with an area ratio of the distorted image in the monitoring image less than or equal to a preset ratio (exemplarily, the preset ratio can be set to 0.6), the region is determined as a target region. The monitoring image may be affected by factors such as light changes, weather conditions (such as rainy and snowy weather), low-light environments at night, etc. Therefore, before target detection, the image may need to be preprocessed. The preprocessing steps may include image denoising, brightness adjustment, contrast enhancement, etc., to improve the image quality and facilitate subsequent target detection. Target detection refers to identifying and locating specific objects (such as pedestrians, vehicles, bicycles, etc.) in an image; usually, a target detection algorithm based on deep learning is used, such as R-CNN (Region-based Convolutional Neural Networks), YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), etc. These algorithms analyze the input image through a trained model, identify the targets in the image, and output the target bounding boxes and their categories (such as pedestrians, cars, bicycles, etc.).

[0040] S102, in response to detecting a target, determining at least one target region where the target is located;

[0041] In implementation, after obtaining and preprocessing the monitoring image, a target detection algorithm is used to identify the target in the image. The target detection algorithm outputs a target bounding box of each target (such as Figure 7 In the embodiment, the detected vehicle rear passing target bounding box frames the vehicle), which is used to describe the position and size of the target in the monitoring image. The position of the target can be determined by determining the position of the center point of the target bounding box, which is the geometric center point of the target bounding box. The position of the center point provides a simplified reference point, making it more intuitive and efficient to determine the position in a complex image, so as to accurately determine at least one target region where the target is located.

[0042] S103, based on the target region where the target is located, determining the relative position of the target and the vehicle to determine whether the target invades the warning area.

[0043] In implementation, the monitoring image is usually divided into multiple target regions to facilitate more accurate determination of the target position. Specifically, due to the characteristics of the fisheye camera, the image distortion is more serious at the edge of the monitoring image, resulting in a larger error in position determination. If the target region is determined to be a region in the monitoring image where the area of distorted image accounts for less than or equal to a preset proportion (for example, the target region is the part of S2 as shown in Figure 2 If the center point of the target detection box is determined to be in the target region of the monitoring image, the relative position of the target bounding box and the warning area of the vehicle is detected to determine whether the target invades the warning area; because the area of distorted image in the target region accounts for a small proportion, the accuracy of detecting whether the target invades the target region through the position of the target bounding box is higher. If the target region is determined to be a region in the monitoring image where the area of distorted image accounts for more than a preset proportion (for example, the target region is the part of the left edge 1 / 5 and the right edge 1 / 5 of the monitoring image as shown in Figure 2S1A and S1B). When it is determined that the target detection frame center point is in the target region of the monitoring image, since the distortion image area ratio of the target region is large, directly relying on the monitoring image for target position judgment may have a large error, therefore the relative distance between the target and the vehicle is determined, and whether the target invades is detected according to the relative distance; high-precision distance measurement data can be provided by the radar sensor of the vehicle, thereby improving the accuracy of target position judgment and more accurately judging whether the target invades. When it is determined that the target indeed invades the warning area, it indicates that there is a potential security threat around the vehicle, such as vehicle scratching, property theft, etc. At this time, the target invasion alarm is issued, which can issue an alarm sound through the loudspeaker inside or outside the vehicle to remind the owner or surrounding personnel; the alarm information is displayed through the instrument panel or the central control screen in the vehicle to prompt the owner to pay attention; the vehicle system is connected with the owner's mobile phone, and the alarm notification is sent to the owner's mobile phone to remind the owner to check the abnormal situation; the camera recording function can also be automatically started to record the whole process of target invasion and saved as a video file for subsequent viewing and evidence preservation. The main purpose of issuing the alarm is to remind the owner or surrounding personnel to pay attention to the potential security threat around the vehicle and take necessary precautions in time. The owner can understand the abnormal situation around the vehicle at the first time to prevent possible losses and safety accidents and ensure the safety of the vehicle and the property inside the vehicle.

[0044] In the embodiment, by dividing the target region according to the image distortion degree of the region and selecting different detection methods, the error caused by the image distortion of the fisheye camera is reduced, the target invasion is accurately detected, not only the common security threats such as vehicle scratching, door opening accidents, property theft, etc. can be prevented, but also other potential abnormal situations such as suspicious persons lingering around the vehicle can be detected. The detection accuracy and reliability are improved, and the user experience and vehicle safety protection are enhanced.

[0045] The following embodiments are used to explain in detail whether the target invades the warning area to ensure that the vehicle surroundings can be effectively monitored in different states. In response to determining that the target region is a region with a distortion image area ratio less than or equal to a preset ratio in the monitoring image, the relative position between the target and the vehicle is determined based on the target region where the target is located to judge whether the target invades the warning area, including:

[0046] The relative position between the target and the vehicle is determined based on the position of the left / bottom corner of the target bounding box of the target relative to the warning region of the vehicle to determine whether the target invades the warning region.

[0047] In specific implementation, whether the target has intruded is determined based on the position of the lower left / right corner of the target bounding box or the position of the lower boundary of the target bounding box; first, the position of the center point of the target bounding box is determined; when the center point of the target bounding box is located in the preset area on the left side of the monitoring image (when the rearview mirror is folded, the preset area on the left side of the monitoring image can be set to the left half of the monitoring image, such as...), Figure 5 (Part S3) determines whether the lower right corner of the target bounding box is within the vehicle's warning area; if it is determined that the lower right corner of the target bounding box is within the vehicle's warning area, it indicates that the target is too close to the vehicle, and the target intrusion is confirmed; if it is determined that the center point of the target bounding box is within a preset area on the right side of the monitoring image (when the rearview mirror is folded, the preset area on the right side of the monitoring image can be set to the right half of the monitoring image, such as... Figure 6 In part S4, the algorithm determines whether the lower left corner of the target bounding box is within the vehicle's detection zone. If it is determined that the lower left corner of the target bounding box is within the vehicle's detection zone (i.e., the target detection algorithm detects the target vehicle and selects it using a target detection box, and the lower left corner of the target bounding box is within the vehicle's detection zone), it indicates that the target is too close to the vehicle, and the target is confirmed to have intruded. Conversely, if it is determined that the lower left / right corner of the target bounding box is not within the vehicle's detection zone...

[0048] Whether the target has intruded into the warning area is determined based on the position of the lower left / right corner of the target's bounding box and the lower boundary of the target's bounding box relative to the vehicle's warning area;

[0049] In specific implementation, firstly, the position of the lower left / right corner of the target bounding box is used to determine whether the target has intruded. If it is determined that the lower left / right corner of the target bounding box is not within the vehicle's warning area, the position of the lower boundary of the target bounding box is used to determine whether the target has intruded. It is then determined whether the lower boundary of the target bounding box intersects simultaneously with the upper and lower boundaries of the warning area. If it is determined that the lower boundary of the target bounding box intersects simultaneously with the upper and lower boundaries of the warning area (e.g....), Figure 7 As shown in the image, this indicates that the target is too close to the vehicle, confirming that the target has intruded. This is to ensure the security monitoring of the vehicle's surrounding environment.

[0050] Whether the target has intruded into the warning area is determined based on the position of the lower boundary of the target's bounding box relative to the vehicle's warning area.

[0051] In implementation, whether the target invades can be determined directly according to the position of the lower boundary of the target bounding box, that is, whether the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time is determined; if it is determined that the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time (as shown in Figure 7 , it indicates that the target is too close to the vehicle, and it is determined that the target invades.

[0052] In this embodiment, the left and right lower corner positions or the lower boundary position of the target bounding box are used for judgment, which improves the adaptability and accuracy of detection. When the left and right lower corner positions of the target bounding box cannot be used to determine whether the target invades, the lower boundary position of the target bounding box is used for judgment, which ensures the utilization rate and detection accuracy of the monitoring image, improves the accuracy and comprehensiveness of detection, and enhances the user experience and vehicle safety guarantee.

[0053] The following embodiments describe in detail how to determine whether the target invades the alert area of the vehicle according to the positions of the left and right lower corners of the target bounding box of the target and the lower boundary of the target bounding box relative to the alert area of the vehicle, so that all potential invasion threats can be detected in time. The determination whether the target invades the alert area of the vehicle according to the positions of the left and right lower corners of the target bounding box of the target and the lower boundary of the target bounding box relative to the alert area of the vehicle includes:

[0054] In response to determining that the left and right lower corners of the target bounding box do not invade the alert area of the vehicle, whether the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time is determined.

[0055] In implementation, when it is determined that the target does not invade according to the positions of the left and right lower corners of the target bounding box, that is, it is determined that the left lower corner or the right lower corner of the target bounding box does not enter the alert area, it is further determined whether the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time. Even if the corner point of the target does not enter the alert area, its lower boundary may still intersect the alert area, thereby constituting a potential invasion threat. The lower boundary of the target bounding box refers to the bottom edge of the target bounding box. If the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time, it is determined that the target invades. If the lower boundary does not intersect the upper boundary and the lower boundary of the alert area at the same time, the target does not invade.

[0056] In response to determining that the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert area at the same time, it is determined that the target invades the alert area.

[0057] In specific implementation, when it is determined that the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert region at the same time, it indicates that the target is too close to the vehicle, and the target intrusion is determined, and an alarm is triggered in time to remind the vehicle owner to pay attention to the potential safety threat. The accuracy and reliability of the target position determination are ensured, the accuracy of the intrusion detection is improved, and the safety monitoring of the environment around the vehicle is ensured.

[0058] In the embodiment, by judging whether the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the alert region at the same time, targets whose corner points do not enter the alert region but still constitute a threat can be detected. It is ensured that even if the corner points of the target bounding box do not enter the alert region, the lower boundary thereof can still intersect the alert region, thereby constituting a potential intrusion threat, and it is ensured that all potential intrusion threats can be detected in time.

[0059] The following embodiment is based on how to determine whether the target intrudes the alert region of the vehicle according to the position of the lower boundary of the target bounding box of the target relative to the alert region of the vehicle, which improves the comprehensiveness and accuracy of the detection and reduces false positives and false negatives. The monitoring image includes a middle preset region; and the determination of whether the target intrudes the alert region of the vehicle according to the position of the lower boundary of the target bounding box of the target relative to the alert region of the vehicle includes:

[0060] In response to determining that the center point of the target bounding box is in the middle preset region of the monitoring image, the folding state of the rearview mirror of the vehicle when the monitoring image outside the vehicle is collected is determined;

[0061] In specific implementation, the middle preset region is usually arranged in the middle one-third region of the monitoring image. When it is determined that the center point of the target bounding box is in the middle preset region of the monitoring image, the folding state of the rearview mirror of the vehicle when the monitoring image outside the vehicle is collected is determined. The rearview mirror is usually installed on both sides of the vehicle and is equipped with a fisheye camera for collecting the monitoring image outside the vehicle. The folding state of the rearview mirror directly affects the collection angle of the fisheye camera. When the rearview mirror is in the unfolded state, the collected monitoring image is as shown in FIG. 1, in which the upper boundary of the alert region is basically parallel to the lower edge of the monitoring image, and the upper boundary of the alert region is on the side away from the vehicle, and the lower boundary of the alert region is on the side close to the vehicle. When the rearview mirror is in the folded state, the collected monitoring image is as shown in FIG. 2. Figure 3 Figure 4 ​As shown in the figure, the upper boundary of the warning area is at an angle with the lower edge of the monitoring image, and the upper boundary of the warning area is on the side away from the vehicle, and the lower boundary of the warning area is on the side close to the vehicle. When the center point of the target bounding box is in the middle preset area of the monitoring image, it is usually not possible to accurately determine whether the target invades according to the positions of the left and right lower corners of the target bounding box. It is necessary to determine the folding state of the vehicle rearview mirror when collecting the monitoring image outside the vehicle, and determine the strategy for judging whether the target invades according to the folding state of the vehicle rearview mirror.

[0062] In response to determining that the rearview mirror is in a non-folded state, it is determined whether the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle;

[0063] In specific implementation, when the rearview mirror is in a non-folded state, it can be more accurately determined whether the target invades by determining whether the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle. First, the coordinate values of all points of the center segment of the lower boundary of the target bounding box are extracted. If the coordinate value of any point of the center segment of the lower boundary is within the range of the warning area, it is determined that the center segment of the lower boundary enters the warning area.

[0064] In response to determining that the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle, it is determined that the target invades.

[0065] In specific implementation, it is determined whether the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle. When it is determined that the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle, it indicates that the target is too close to the vehicle, and it is determined that the target invades. The comprehensiveness and accuracy of the intrusion detection are improved, and the safety monitoring of the environment around the vehicle is ensured.

[0066] In the embodiment, by setting a middle preset area in the monitoring image, when the center point of the target bounding box is in the middle preset area of the monitoring image, the folding state of the vehicle rearview mirror when collecting the monitoring image outside the vehicle is determined. If the rearview mirror is in a non-folded state, it is determined whether the target invades by determining whether the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle. It is ensured that even if the target is located in the middle area of the monitoring image, it can be accurately determined whether it constitutes an invasion threat, and it is ensured that all potential invasion threats can be detected in time. The comprehensiveness and accuracy of the detection are significantly improved, the false positives and false negatives are reduced, and the real-time response capability and user experience are enhanced.

[0067] The target intrusion is determined based on the positions of the left and right lower corners of the target bounding box, to more accurately determine the position relationship of the target in the monitoring image, and to improve the accuracy and reliability of the target position determination. The monitoring image includes a left preset region and a right preset region. The target intrusion into the alert region of the vehicle is determined based on the positions of the left and right lower corners of the target bounding box, including:

[0068] In response to determining that the center point of the target bounding box is in the left preset region of the monitoring image, it is determined whether the right lower corner of the target bounding box is in the alert region of the vehicle.

[0069] In specific implementation, the alert region is used to determine whether the target poses a threat to the vehicle, and the size and position of the alert region can be configured according to actual needs. When it is determined that the center point of the target bounding box is in the left preset region of the monitoring image, the coordinate values (xmax, ymin) of the right lower corner of the target bounding box can be extracted. The coordinate values of the right lower corner are compared with the boundary coordinate values of the alert region. If the xmax and ymax coordinate values of the right lower corner are both within the range of the alert region, it is determined that the right lower corner enters the alert region.

[0070] In response to determining that the right lower corner of the target bounding box is in the alert region of the vehicle, it is determined that the target intrudes.

[0071] In specific implementation, if it is determined that the right lower corner of the target bounding box is in the alert region of the vehicle, it indicates that the target is too close to the vehicle, and it is determined that the target intrudes. A target intrusion alarm is issued to remind the vehicle owner to pay attention to potential safety threats. The accuracy and reliability of the target position determination are ensured, the accuracy of the intrusion detection is improved, and the safety monitoring of the environment around the vehicle is ensured.

[0072] In response to determining that the center point of the target bounding box is in the right preset region of the monitoring image, it is determined whether the left lower corner of the target bounding box is in the alert region of the vehicle.

[0073] In specific implementation, when it is determined that the center point of the target bounding box is in the right preset region of the monitoring image, the coordinate values (xmin, ymin) of the left lower corner of the target bounding box can be extracted. The coordinate values of the left lower corner are compared with the boundary coordinate values of the alert region. If the xmin and ymin coordinate values of the left lower corner are both within the range of the alert region, it is determined that the left lower corner enters the alert region.

[0074] In response to determining that the left lower corner of the target bounding box is in the alert region of the vehicle, it is determined that the target intrudes.

[0075] In practice, if the lower left corner of the target's bounding box is determined to be within the vehicle's warning zone, it indicates that the target is too close to the vehicle, confirming the target intrusion. An intrusion alarm is then issued to alert the vehicle owner to the potential security threat. This ensures the accuracy and reliability of target location determination, improves the accuracy of intrusion detection, and ensures secure monitoring of the vehicle's surrounding environment.

[0076] In this embodiment, by determining the position of the center point of the target bounding box, the positional relationship of the target in the monitoring image can be more accurately determined, which helps to improve the accuracy and reliability of target position determination. Within different target areas, determination is made based on the lower left / right corner position of the target bounding box, significantly improving the accuracy and reliability of detection, reducing false alarms and false negatives, and enhancing user experience and vehicle safety.

[0077] In some embodiments, in response to determining that the target area is a region in the monitoring image where the proportion of distorted image area is greater than a preset proportion, determining the relative position of the target and the vehicle based on the target area where the target is located, in order to determine whether the target has intruded into the warning area, includes:

[0078] Determine the relative distance between the target and the vehicle;

[0079] In response to determining that the relative distance is less than or equal to a preset distance, it is determined that the target has intruded into the warning area.

[0080] In specific implementation, if the target area is determined to be an area in the monitored image where the proportion of distorted image area is greater than a preset proportion (e.g., Figure 8 As shown, the center point of the target bounding box is located in the target area of ​​the monitored image, determining the relative distance between the target and the vehicle. High-precision distance measurement data can be provided by the vehicle's radar sensors, thereby improving the accuracy of target location determination and more accurately determining whether the target has intruded. When the relative distance is determined to be less than or equal to a preset distance (for example, the preset distance can be set to 20cm), it indicates that the target is too close to the vehicle, and the target intrusion warning zone is determined.

[0081] In response to determining that the relative distance is greater than a preset distance, it is determined that the target has not been intruded.

[0082] In practice, if the relative distance between the target and the vehicle is greater than a preset distance, it indicates that the target is far enough away from the vehicle, and therefore the target is deemed not to have intruded. This ensures the accuracy and reliability of intrusion detection by providing a quantitative standard through distance judgment to confirm whether a target poses a threat to the vehicle, thus ensuring the security monitoring of the vehicle's surrounding environment.

[0083] In this embodiment, the relative distance judgment is used to determine whether the target invades, reducing the false positives and false negatives caused by the image distortion judgment error. The relative distance judgment provides a quantitative standard to determine whether the target poses a threat to the vehicle, ensuring the objectivity and consistency of the judgment, and improving the reliability of the intrusion detection; ensuring that any potential invasion threat can be discovered in time, and providing necessary evidence support.

[0084] In some embodiments, the method further comprises:

[0085] In response to detecting the target, determining an area of a target bounding box of the target;

[0086] In implementation, the target detection algorithm is used to identify the target in the monitoring image. The target detection algorithm generates a target bounding box of the target, which describes the position and range of the target in the image. After detecting the target and generating the target bounding box, the area of the target bounding box is calculated. The size of the target bounding box area can reflect the size of the target in the image. A larger target bounding box area generally indicates that the target is closer to the fisheye camera, while a smaller target bounding box area indicates that the target is farther away from the fisheye camera.

[0087] In response to determining that the area is less than or equal to a preset area, determining a relative distance of the target from the vehicle, and in response to determining that the relative distance is less than or equal to a preset distance, determining that the target invades the alert area.

[0088] In implementation, when it is determined that the area is less than or equal to a preset area (the preset area is a threshold value set according to the specific application scenario and vehicle safety requirements, which is used to distinguish between close and distant targets, and the preset area can be set to one-third of the area of the monitoring image, for example), it indicates that the target is far away from the fisheye camera, and it is possible that the target is at the rear end of the vehicle. At this time, the accuracy of detecting whether the target invades based on the position of the target bounding box is poor, so it is necessary to determine the relative distance of the target from the vehicle, and to detect whether the target invades according to the relative distance. When it is determined that the relative distance is less than or equal to a preset distance (the preset distance can be set to 20 cm, for example), it indicates that the target is too close to the vehicle, and it is determined that the target invades the alert area.

[0089] In this embodiment, by calculating the area of the target bounding box, the distance of the target can be preliminarily judged. A larger bounding box area generally indicates that the target is closer to the camera, while a smaller bounding box area indicates that the target is farther away. Using the area of the target bounding box as an auxiliary judgment basis improves the accuracy and reliability of target position judgment. According to the size of the area, a suitable detection method is selected to quickly determine whether the target invades. This significantly improves the accuracy and reliability of the detection, reduces false positives and false negatives, and enhances real-time response capability and user experience.

[0090] The vehicle also includes a fisheye camera mounted at the front and a fisheye camera mounted at the rear; the method further includes:

[0091] Obtain distortion information of the monitoring image, determine multiple target regions with different distortion degrees in the monitoring image based on the distortion information, and perform target detection on the monitoring image;

[0092] In response to the detection of a target, at least one target region is determined where the target is located;

[0093] Based on the target area where the target is located, the relative position between the target and the vehicle is determined to determine whether the target has intruded into the warning area.

[0094] In practice, the monitoring images (such as those captured by the fisheye camera at the front and rear of the vehicle) are used in the implementation. Figure 9 The monitoring image captured by the fisheye camera on the front of the vehicle is shown; such as Figure 10 The monitoring image captured by the fisheye camera at the rear of the vehicle (shown in the image) has a similar viewing angle to the monitoring image captured by the fisheye camera on the rearview mirror when the rearview mirror is not folded. The positions of the warning areas in the monitoring images are also similar. Therefore, when it is determined that the center point of the target detection frame of the target is located in the target area of ​​the monitoring image (the target area is the area in the monitoring image where the area of ​​the distorted image is less than or equal to a preset percentage), the position of the lower left / right corner of the target boundary frame or the center segment of the lower boundary of the target boundary frame relative to the warning area is used to determine whether the target has intruded. When the center point of the target boundary frame is located in the preset area on the left side of the monitoring image, it is determined whether the lower right corner of the target boundary frame is located in the vehicle's warning area. When it is determined that the lower right corner of the target boundary frame is located in the vehicle's warning area, it indicates that the target is too close to the vehicle, and the target has intruded. When it is determined that the center point of the target boundary frame is located in the preset area on the right side of the monitoring image, it is determined whether the lower left corner of the target boundary frame is located in the vehicle's warning area. When it is determined that the lower left corner of the target boundary frame is located in the vehicle's warning area, it indicates that the target is too close to the vehicle, and the target has intruded. The monitoring image also includes a preset central region. When the center point of the target bounding box is determined to be within the preset central region of the monitoring image, it is determined whether the center segment of the lower boundary of the target bounding box is within the vehicle's warning zone. If the center segment of the lower boundary of the target bounding box is determined to be within the vehicle's warning zone, it indicates that the target is too close to the vehicle, and the target intrusion is confirmed. This improves the comprehensiveness and accuracy of intrusion detection, ensuring the security monitoring of the vehicle's surrounding environment.

[0095] In this embodiment, by setting fisheye cameras at the front and rear of the vehicle, a full range of view angles can be provided, covering all key areas in front and rear of the vehicle. This full-range monitoring ensures that there is no blind area around the vehicle, can timely discover any potential intrusion threat, significantly improves the comprehensiveness and accuracy of detection, reduces false positives and false negatives, enhances real-time response capability and user experience. Ensure that any potential intrusion threat can be discovered in time.

[0096] It should be noted that the method of the embodiments of the present application can be executed by a single device, such as a computer or a server, etc. The method of the embodiments of the present application can also be applied to a distributed scenario, and be completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only execute one or more steps in the method of the embodiments of the present application, and the multiple devices can interact with each other to complete the method.

[0097] It should be noted that some embodiments of the present application have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than that described above and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0098] Based on the same inventive concept, the present application also provides a vehicle control device corresponding to any of the above-mentioned embodiment methods.

[0099] Reference Figure 11 , the vehicle control device comprises:

[0100] The acquisition module 701 is configured to acquire distortion information of the monitoring image, determine a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and perform target detection on the monitoring image.

[0101] The determination module 702 is configured to determine at least one target region where the target is located in response to detecting the target.

[0102] The detection module 703 is configured to determine the relative position of the target and the vehicle based on the target region where the target is located, to determine whether the target intrudes into the warning area.

[0103] Further, the detection module 703 is specifically configured to:

[0104] determine whether the target intrudes into the warning area according to the position of the left / bottom corner of the target bounding box of the target relative to the warning area of the vehicle; or,

[0105] determining whether the target intrudes into the alert area of the vehicle according to positions of the left / right lower corner of the target bounding box of the target and the lower boundary of the target bounding box relative to the alert area of the vehicle; or,

[0106] determining whether the target intrudes into the alert area of the vehicle according to a position of the lower boundary of the target bounding box of the target relative to the alert area of the vehicle.

[0107] Further, the judging module 702 is specifically configured to:

[0108] in response to determining that the left / right lower corner of the target bounding box does not intrude into the alert area of the vehicle, judging whether the lower boundary of the target bounding box intersects with the upper boundary and the lower boundary of the alert area at the same time;

[0109] in response to determining that the lower boundary of the target bounding box intersects with the upper boundary and the lower boundary of the alert area at the same time, determining that the target intrudes into the alert area.

[0110] Further, the detecting module 703 is specifically configured to:

[0111] in response to determining that the center point of the target bounding box is in the middle preset area of the monitoring image, determining the folding state of the rearview mirror of the vehicle when the monitoring image outside the vehicle is collected;

[0112] in response to determining that the rearview mirror is in the unfolded state, judging whether the center segment of the lower boundary of the target bounding box is in the alert area of the vehicle;

[0113] in response to determining that the center segment of the lower boundary of the target bounding box is in the alert area of the vehicle, determining that the target intrudes into the alert area.

[0114] Further, the detecting module 703 is specifically configured to:

[0115] in response to determining that the center point of the target bounding box is in the left preset area of the monitoring image, judging whether the right lower corner of the target bounding box is in the alert area of the vehicle;

[0116] in response to determining that the right lower corner of the target bounding box is in the alert area of the vehicle, determining that the target intrudes into the alert area.

[0117] Further, the detecting module 703 is specifically configured to:

[0118] in response to determining that the center point of the target bounding box is in the right preset area of the monitoring image, judging whether the left lower corner of the target bounding box is in the alert area of the vehicle;

[0119] In response to determining that the lower left corner of the target bounding box is in the alert area of the vehicle, it is determined that the target invades the alert area.

[0120] Further, the detection module 703 is specifically further configured to:

[0121] determine the relative distance between the target and the vehicle;

[0122] In response to determining that the relative distance is less than or equal to a preset distance, it is determined that the target invades the alert area.

[0123] Further, the detection module 703 is specifically further configured to:

[0124] In response to detecting the target, determine the area of the target bounding box;

[0125] In response to determining that the area is less than or equal to a preset area, determine the relative distance between the target and the vehicle, and in response to determining that the relative distance is less than or equal to a preset distance, determine that the target invades the alert area.

[0126] For the convenience of description, the above apparatus is described in various modules in terms of functions. Of course, in the implementation of the present application, the functions of each module can be implemented in one or more software and / or hardware.

[0127] The apparatus of the above embodiments is used to implement the corresponding vehicle control method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which are not described here.

[0128] Based on the same inventive concept, corresponding to any of the above method embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle control method of any one of the above embodiments.

[0129] Figure 12 A more specific hardware structure of an electronic device provided in the present embodiment is shown, which can include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040 and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030 and the communication interface 1040 are connected to each other through the bus 1050 for communication within the device.

[0130] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc., for executing relevant programs to implement the technical solutions provided by the embodiments of the present specification.

[0131] The memory 1020 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1020 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the relevant program codes are saved in the memory 1020 and called and executed by the processor 1010.

[0132] The input / output interface 1030 is configured to connect input / output modules to implement information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input devices can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices can include a display, a speaker, a vibrator, an indicator light, etc.

[0133] The communication interface 1040 is configured to connect a communication module (not shown in the figure) to implement the communication interaction between the device and other devices. The communication module can realize communication through a wired manner (such as USB, network cable, etc.) or through a wireless manner (such as mobile network, WIFI, Bluetooth, etc.).

[0134] The bus 1050 includes a channel for transmitting information between various components (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040) of the device.

[0135] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only include the components necessary to implement the solutions of the embodiments of the present specification, and does not have to include all the components shown in the figure.

[0136] The electronic device of the above-mentioned embodiments is used to implement the corresponding vehicle control method in any of the preceding embodiments, and has the beneficial effects of the corresponding method embodiments, which are not repeated here.

[0137] Based on the same inventive concept, corresponding to the method of any of the above-mentioned embodiments, the present application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the vehicle control method of any of the above-mentioned embodiments.

[0138] The computer-readable medium of the present embodiment includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0139] The storage medium of the above-mentioned embodiments stores computer instructions for causing the computer to perform the vehicle control method of any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiments, which are not repeated here.

[0140] Based on the same concept, corresponding to the method of any of the above-mentioned embodiments, the present application also provides a computer program product comprising computer program instructions which, when executed on a computer, cause the computer to perform the method of any of the above-mentioned embodiments, have the beneficial effects of the corresponding method embodiments, which are not repeated here.

[0141] It can be understood that before using the technical solutions of various embodiments in the present disclosure, the user will be informed of the type, use range, use scenario, etc. of the personal information involved by appropriate means, and the authorization of the user will be obtained.

[0142] For example, in response to receiving the user's active request, send prompt information to the user to explicitly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can voluntarily choose whether to provide personal information to the electronic device, application program, server or storage medium, etc. software or hardware that performs the technical solutions of the present disclosure.

[0143] As an optional but non-limiting implementation, in response to accepting the active request of the user, the manner of sending the prompt information to the user may be, for example, a pop-up window manner, in which the prompt information may be presented in the form of text. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0144] It can be understood that the above notification and user authorization obtaining process is only illustrative, and does not limit the implementation of the present disclosure, and other manners meeting the relevant laws and regulations can also be applied to the implementation of the present disclosure.

[0145] It should be understood by those skilled in the art that the above discussion of any embodiment is only exemplary and is not intended to limit the scope of the present application to these examples; under the idea of the present application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes of different aspects of the embodiments of the present application as described above. In order to be brief, they are not provided in detail.

[0146] In addition, in order to simplify the description and discussion, and so as not to make the embodiments of the present application difficult to understand, the known power / ground connections of integrated circuit (IC) chips and other components can or can not be shown in the provided drawings. In addition, the devices can be shown in the form of block diagrams in order to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present application are to be implemented (i.e. these details should be entirely within the understanding of those skilled in the art). Where specific details (e.g. circuits) are set forth in order to describe an exemplary embodiment of the present application, it will be apparent to those skilled in the art that the present application can be practiced without these specific details or with variations of these specific details. Therefore, these descriptions should be considered as illustrative rather than limiting.

[0147] Although the present application has been described in conjunction with specific embodiments thereof, many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g. dynamic RAM (DRAM)) can use the embodiments discussed.

[0148] The embodiments of the present application are intended to cover all such alternatives, modifications and variations as falling within the broad scope of the application claimed. Accordingly, any omission, modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present application should be included in the protection scope of the present application.

Claims

1. A vehicle control method characterized by, The rearview mirror of the vehicle is provided with a camera for collecting a monitoring image outside the vehicle, the monitoring image comprising a warning area; the method comprises: obtaining distortion information of the monitoring image, determining a plurality of target regions with different distortion degrees in the monitoring image based on the distortion information, and performing target detection on the monitoring image; in response to detecting a target, determining at least one target region where the target is located; based on the target region where the target is located, determining the relative position of the target and the vehicle to determine whether the target intrudes into the warning area; in response to detecting a target, determining the area of the target bounding box; in response to determining that the area is less than or equal to a preset area, determining the relative distance between the target and the vehicle, and in response to determining that the relative distance is less than or equal to a preset distance, determining that the target intrudes into the warning area.

2. The vehicle control method according to claim 1, characterized by, in response to determining that the target region is a region with a distortion image area proportion less than or equal to a preset proportion in the monitoring image, the method comprises: determining whether the target intrudes into the warning area according to the position of the left / right lower corner of the target bounding box of the target relative to the warning area of the vehicle; or, determining whether the target intrudes into the warning area according to the position of the left / right lower corner of the target bounding box of the target and the lower boundary of the target bounding box relative to the warning area of the vehicle; or, determining whether the target intrudes into the warning area according to the position of the lower boundary of the target bounding box of the target relative to the warning area of the vehicle.

3. The vehicle control method according to claim 2, characterized by, the method comprises: in response to determining that the left / right lower corner of the target bounding box does not intrude into the warning area of the vehicle, determining whether the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the warning area at the same time; in response to determining that the lower boundary of the target bounding box intersects the upper boundary and the lower boundary of the warning area at the same time, determining that the target intrudes into the warning area.

4. The vehicle control method according to claim 2, characterized by the monitoring image comprises a middle preset region; the method comprises: in response to determining that the center point of the target bounding box is in the middle preset region of the monitoring image, determining the folding state of the rearview mirror of the vehicle when the monitoring image outside the vehicle is collected; in response to determining that the rearview mirror is in a non-folded state, determining whether the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle; in response to determining that the center segment of the lower boundary of the target bounding box is in the warning area of the vehicle, determining that the target intrudes into the warning area.

5. The vehicle control method according to claim 2, characterized by the monitoring image comprises a left preset region and a right preset region; the method comprises: in response to determining that the center point of the target bounding box is in the left preset area of the monitoring image, determining whether the lower right corner of the target bounding box is in the alert area of the vehicle; in response to determining that the lower right corner of the target bounding box is in the alert area of the vehicle, determining that the target invades the alert area.

6. The vehicle control method according to claim 2, characterized by The monitoring image comprises a left preset area and a right preset area; and the method for determining whether the target invades the alert area according to the position of the left / lower right corner of the target bounding box of the target relative to the alert area of the vehicle comprises: in response to determining that the center point of the target bounding box is in the right preset area of the monitoring image, determining whether the lower left corner of the target bounding box is in the alert area of the vehicle; in response to determining that the lower left corner of the target bounding box is in the alert area of the vehicle, determining that the target invades the alert area.

7. The vehicle control method according to claim 2, characterized by in response to determining that the target region is a region with a distortion image area ratio greater than a preset ratio in the monitoring image, the method for determining the relative position of the target and the vehicle based on the target region where the target is located to determine whether the target invades the alert area comprises: determining the relative distance between the target and the vehicle; in response to determining that the relative distance is less than or equal to a preset distance, determining that the target invades the alert area.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The processor implements the method according to any one of claims 1 to 7 when executing the program.

9. A vehicle characterized by comprising: The vehicle comprises the electronic device according to claim 8. The vehicle comprises the electronic device according to claim 8.

Images