Obstacle Detection Method, Device, and Computer Equipment

Abstract

The present invention relates to the technical field of image processing, and particularly to an obstacle detection method, device, and computer device, including the following steps: Step 1: Obtain a road surface image, record the corresponding road section area of the road surface image as a monitored road section, divide the road surface image into multiple lane rectangles according to road segmentation markings, and divide the lane rectangles into multiple unit rectangles along the lane length direction; Step 2: After the vehicle enters the monitored road section, a set of vehicle passing images is composed of multiple road surface images collected. Based on the set of vehicle images, determine multiple relevant lanes passed by, and perform comparative analysis on the images corresponding to the relevant lanes passed by to determine abnormal road surface rectangles; Step 3: Based on the data when the passing vehicle passes through the monitored road section, mark the unit rectangle as a first abnormal rectangle or a second abnormal rectangle; Step 4: Based on the number of times the unit rectangle is marked, comprehensively calculate the abnormal confidence level of the unit rectangle; Step 5: Generate different signals based on the magnitude of the abnormal confidence level of the unit rectangle.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to an obstacle detection method, device, and computer device. Background Art

[0002] The obstacle monitoring method is a core technology that uses multiple sensors and intelligent algorithms to continuously perceive and identify obstacles in the environment in real time, providing a basis for safe navigation and decision-making for autonomous systems. The current mainstream solutions adopt multi-sensor fusion, combined with algorithms such as Kalman filtering and deep learning, to achieve all-weather and highly robust obstacle detection in the fields of autonomous driving, robotics, intelligent transportation, etc., taking into account real-time performance and adaptability to complex scenarios, and promoting the evolution of intelligent systems from collision avoidance warning to full-autonomous decision-making.

[0003] Currently, the mainstream obstacle detection methods can achieve safety warning and route avoidance during vehicle driving through various types of sensors carried by the vehicle. However, this type of detection method is limited by the vehicle's own conditions, such as the number of sensors carried by the vehicle, the sensitivity of the vehicle's obstacle monitoring system, and the vehicle's moving speed, resulting in inconsistent obstacle detection effects for different vehicles and affecting road traffic safety. Summary of the Invention

[0004] In view of the above-mentioned drawbacks of the prior art, the present invention provides an obstacle detection method, device, and computer device, which can effectively solve the problem of inconsistent obstacle detection effects for different vehicles in the prior art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0006] The present invention provides an obstacle detection method, device, and computer device, including at least the following steps:

[0007] Step 1: Obtain a road surface image, record the corresponding road section area of the road surface image as a monitored section, divide the road surface image into multiple lane rectangles according to the road segmentation markings, and divide each lane rectangle into multiple unit rectangles along the lane length direction;

[0008] Step 2: Periodically obtain a road surface image and record it as a regular road surface image, collect a road surface image when the vehicle enters the monitored section and record it as a vehicle road surface image. A plurality of regular road surface images collected after obtaining the vehicle road surface image and the vehicle road surface image together form a vehicle passing image set, and each vehicle passing image set corresponds to a triggering vehicle;

[0009] Based on the analysis of the vehicle image set, obtain a plurality of relevant lanes corresponding to the triggering vehicle, and perform image comparison and analysis on the plurality of unit rectangles corresponding to the relevant lanes. Among them:

[0010] Obtain the rectangular images of the unit rectangle before and after the triggering vehicle passes by, compare them, and determine the road surface abnormal rectangle based on the similarity between the two rectangular images;

[0011] Step 3: Analyze based on the acceleration change and steering wheel torque change when the vehicle passing by passes through the monitored section, and mark the unit rectangle as the first abnormal rectangle or the second abnormal rectangle;

[0012] Step 4: Based on the number of times the unit rectangle is marked, comprehensively calculate the abnormal confidence level of the unit rectangle;

[0013] Step 5: Compare the abnormal confidence level of the unit rectangle with the preset first confidence threshold and second confidence threshold, and generate different signals based on the size of the abnormal confidence level of the unit rectangle.

[0014] Further, the process of constructing the vehicle passing image set is as follows:

[0015] Obtain the speed limit value of the monitored section , substitute it into the formula for calculation to obtain the associated period , where is the length of the monitored section, set the road surface vehicle image acquisition time as , and denote the regular road surface images collected within the time interval as the associated images of the road surface vehicle image at moment. Combine the road surface vehicle image and its associated images to form the vehicle passing image set.

[0016] Further, the process of obtaining the relevant passing lanes is as follows:

[0017] Denote the road surface images in the vehicle passing image set as relevant images. Mark the vehicle contour of the triggering vehicle in the relevant images, construct the smallest rectangle containing the vehicle contour as the vehicle rectangle, and construct the straight line passing through the midpoints of the two short sides of the vehicle rectangle as the vehicle body direction line;

[0018] Draw the lane centerlines corresponding to each lane rectangle in the relevant images and denote them as respectively, where f is the serial number of the lane centerline;

[0019] Obtain the intersection points of the vehicle body direction line and each lane centerline in the relevant images as the passing intersection points, and denote the lane containing the passing intersection points as the passing relevant lane.

[0020] Further, the process of marking the road surface abnormal rectangle is as follows:

[0021] Respectively obtain the two road surface comparison images collected most recently before and after the time interval as the comparison images;

[0022] Obtain two rectangular images corresponding to the same unit rectangle in two comparison images, calculate the similarity between the two rectangular images, and when the similarity between the two rectangles is less than or equal to a preset similarity threshold, mark the corresponding unit rectangle as an abnormal rectangle area;

[0023] Obtain the similarities of multiple abnormal rectangle areas. When the similarities of the multiple abnormal rectangle areas are not equal, mark the abnormal rectangle areas as road surface abnormal rectangles, and mark the corresponding positions as suspected obstacle positions.

[0024] Further, the process of marking the first abnormal rectangle is as follows:

[0025] Preset a time limit period, obtain the passing vehicles passing through the monitored section within the time limit period as target vehicles, obtain the driving routes of the target vehicles within the time limit period, and any point on the driving route corresponds to a passing moment, and obtain the passing time interval corresponding to the monitored section;

[0026] Obtain the vertical acceleration change curve of the target vehicle within the passing time interval, obtain multiple peak points of the vertical acceleration change curve, preset an acceleration threshold, and mark the moment corresponding to the peak point where the vertical acceleration value is greater than the acceleration threshold as the first abnormal moment;

[0027] Obtain the vehicle position corresponding to the first abnormal moment as the first abnormal position, and mark the unit rectangle containing the first abnormal position as the first abnormal rectangle.

[0028] Further, the process of marking the second abnormal rectangle is as follows:

[0029] Obtain the steering wheel torque change curve of the target vehicle within the passing time interval, preset a torque threshold, mark the part of the steering wheel torque change curve higher than the torque threshold as the torque abnormal part, and analyze the steering angle within the time interval corresponding to the torque abnormal part to obtain a steering abnormality value;

[0030] Compare the steering abnormality value with a preset steering abnormality threshold. When the steering abnormality value is greater than the preset steering abnormality threshold, mark the start moment of the torque abnormal part as the second abnormal moment;

[0031] Obtain the vehicle position corresponding to the second abnormal moment as the second abnormal position, and mark all the unit rectangles corresponding to the lane where the second abnormal position is located as the second abnormal rectangles.

[0032] Further, the specific process of calculating the steering abnormality value is as follows:

[0033] Let the time interval corresponding to the torque abnormal part be , obtain the time interval and obtain the angle sequence of the steering wheel rotation within the time interval The angular sequence contains multiple rotation angles and each rotation angle corresponds to a one-way rotation of the steering wheel;

[0034] When the one-way rotation direction is clockwise, the rotation angle is positive, and when the one-way rotation direction is counterclockwise, the rotation angle is negative;

[0035] Substitute into the formula for calculation;

[0036] Obtain the one-way rotation amplitude and the total rotation amplitude ;

[0037] Where:

[0038] ;

[0039] means taking the maximum value between ;

[0040] Substitute the one-way rotation amplitude and the total rotation amplitude into the formula for calculation to obtain the steering anomaly value, where are all preset weight coefficients.

[0041] Furthermore, the calculation process of the anomaly confidence of the unit rectangle is as follows:

[0042] Obtain the number of times each unit rectangle is marked as a road surface anomaly rectangle and record it as the first influence value ; obtain the number of times each unit rectangle is marked as the first anomaly rectangle and record it as the second influence value ; obtain the number of times each unit rectangle is marked as the second anomaly rectangle and record it as the third influence value ;

[0043] Substitute into the formula for calculation to obtain the anomaly confidence of each unit rectangle, where:

[0044] are all preset weight coefficients;

[0045] m is the number of unit rectangles on each lane rectangle.

[0046] An obstacle detection device, including road surface monitoring devices arranged on both sides of the road, and the road surface monitoring devices face the road surface directly;

[0047] It further includes a vehicle-mounted device, which is electrically connected to a three-axis accelerometer and a torque sensor. The three-axis accelerometer is arranged at the front of the vehicle to measure the three-axis acceleration of the vehicle, and the torque sensor is arranged at the steering wheel shaft to measure the torque when the steering wheel rotates.

[0048] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above method are implemented.

[0049] The technical solution provided by the present invention has the following beneficial effects compared with the known prior art:

[0050] 1. Based on the existing road surface monitoring, the present invention can analyze and judge the lane where the triggering vehicle passes by triggering multiple images collected when the vehicle passes, and then compare and analyze the images before and after the triggering vehicle passes corresponding to each lane to determine whether the triggering vehicle drops an object to form a road obstacle. This method utilizes the characteristic that "most road obstacles come from vehicle drops", effectively improving the efficiency of road surface monitoring to identify obstacles.

[0051] 2. The present invention integrates vehicle driving data and road surface monitoring data to comprehensively judge the position information of obstacles. On the one hand, it can reduce the equipment pressure of road surface monitoring equipment to identify obstacles, and on the other hand, it can make full use of the driving data of different vehicles passing through the monitoring section to analyze the road surface conditions. The combination of the two can effectively improve the accuracy of obstacle identification. Compared with the existing single-angle obstacle identification method, it is more suitable for road safety management for the driving safety of all vehicles in the current situation of the rapid development of "Internet of Vehicles". BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0053] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0055] The present invention will be further described below with reference to the embodiments.

[0056] Currently, the mainstream obstacle detection methods can use various types of sensors carried by vehicles to achieve safety warnings and route avoidance during vehicle driving. However, the ability of this type of detection method to identify obstacles is limited by the vehicle's own conditions, such as the number of sensors carried by the vehicle, the sensitivity of the vehicle's obstacle monitoring system, and the vehicle's moving speed. As a result, different vehicles have different recognition effects on the same obstacle on the road surface, causing some vehicles to be unable to detect road surface obstacles in time, thus triggering road risks. Therefore, there is an urgent need for an obstacle detection method that can call road surface monitoring, combine the vehicle's own obstacle recognition system, and map navigation to achieve unified obstacle detection and obstacle avoidance push, ensuring the driving safety of all vehicles on the road surface.

[0057] Refer to Figure 1 , the obstacle detection method includes at least the following steps:

[0058] Step 1: Obtain road surface images through road surface monitoring, independently analyze the monitoring area of each road surface monitoring camera, and obtain the road section area corresponding to the road surface monitoring camera (each road surface monitoring camera is calibrated for the road section it monitors and covers at the initial installation to clarify the road section area corresponding to each road surface monitoring camera), denoted as the monitored road section. Divide the road surface image corresponding to the road surface monitoring camera into multiple rectangular areas along the road segmentation markings, denoted as lane rectangles. Each lane rectangle corresponds to a lane, and the lane rectangle is further divided into multiple unit rectangles along the lane length direction;

[0059] Step 2: Identify suspected obstacles on the road surface based on the road surface image, where:

[0060] There is a preset image fixed acquisition period, and road surface images are regularly acquired at intervals of the image fixed acquisition period, denoted as road surface regular images. When the vehicle passes through the monitored road section, road surface images are acquired, denoted as road surface vehicle images. Identify the vehicles in the road surface regular images and count the number. Denote the road surface regular images without vehicles as road surface comparison images (vehicle recognition and vehicle passing detection are existing image recognition technologies, which will not be elaborated here).

[0061] It should be noted that the road surface vehicle image and the regular road surface image do not conflict. The same road surface image can be either a road surface vehicle image or a regular road surface image. The difference lies in the triggering conditions for collecting road surface images. Whenever a vehicle enters the monitored section, a road surface vehicle image will be generated.

[0062] Obtain the speed limit value of the monitored section , substitute it into the formula for calculation to obtain the correlation period , where is the length of the monitored section. Let any road surface vehicle image acquisition time be , and record the regular road surface images collected within the time interval as the associated images of this road surface vehicle image. Combine the road surface vehicle image and its associated images to form a vehicle passing image set. The vehicle that triggers the image acquisition signal corresponding to the road surface vehicle image is denoted as the triggering vehicle, and each vehicle passing image set is bound to the corresponding triggering vehicle;

[0063] It should be noted that the road surface monitoring equipment in the prior art can intelligently identify the vehicle that triggers the image acquisition signal when collecting road surface vehicle images. The same function is also applied to radar speed measurement and capture equipment and lane change capture equipment. The road surface vehicle image and its associated images correspond to multiple images of a vehicle during its travel in the monitored section. Through these images, the trajectory of the vehicle passing through the monitored section can be analyzed to determine the lane the vehicle passes through. In addition, through the limitation of the correlation period, it can be ensured that the associated images must contain the triggering vehicle (when the triggering vehicle does not exceed the speed limit).

[0064] Analyze each element in the vehicle passing image set one by one to obtain multiple passing-related lanes, where:

[0065] Record the road surface images in the vehicle passing image set as related images. Mark the vehicle contour of the triggering vehicle in the related images. Construct the smallest rectangle containing the vehicle contour and denote it as the vehicle rectangle. Construct a straight line passing through the midpoints of the two short sides of the vehicle rectangle and denote it as the vehicle body direction line (the vehicle body direction line is used to represent the vehicle driving direction). Draw the lane centerlines corresponding to each lane rectangle (the lane centerline is a line segment with a length equal to ) in the related images and denote them as respectively. f is the serial number of the lane centerline (the serial number of the corresponding lane from left to right). Obtain the intersection points of the vehicle body direction line (only the line segment in the related image and not including its extension line) and each lane centerline in the related image and denote them as passing intersection points. Denote the lane corresponding to the lane rectangle containing the passing intersection points as the passing-related lane.

[0066] It should be noted that the relevant lanes passed by correspond to the lanes that the triggering vehicle may pass through, and the obstacles appearing on the lanes usually come from the passing vehicles. Therefore, by performing obstacle recognition and analysis on the road surface within the relevant lanes passed by, obstacles can be recognized more efficiently. Each vehicle passing image set corresponds to one or more relevant lanes passed by.

[0067] Obtain the time intervals (corresponding to each vehicle passing image set) respectively The two road surface comparison images collected most recently before and after are recorded as comparison images. Based on the front-back changes corresponding to multiple relevant lanes passed by in the comparison images, the suspected positions of obstacles are determined. The analysis process is as follows:

[0068] Obtain the two rectangular images corresponding to the same unit rectangle in the two comparison images, calculate the similarity of the two rectangular images. When the similarity of the two rectangles is less than or equal to the preset similarity threshold, the corresponding unit rectangle is recorded as an abnormal rectangle area. Obtain the similarity of multiple abnormal rectangle areas. When the similarities of multiple abnormal rectangle areas are not equal, mark the abnormal rectangle area as the road surface abnormal rectangle, and mark the corresponding position as the suspected position of the obstacle;

[0069] It should be noted that when a vehicle passes through a pothole, it will leave water marks and tire marks on the road surface. These water marks will mislead the image comparison and also cause changes in the similarity of the unit rectangle. However, through the comparison of the similarities of multiple abnormal rectangle areas, the influence of the water marks can be excluded because generally, the dropped obstacles will not form multiple abnormal rectangle areas with equal similarities.

[0070] Step two can, on the basis of the existing road surface monitoring, analyze and judge the lanes passed by the triggering vehicle through multiple images collected when the triggering vehicle passes, and then compare and analyze the images before and after the passing of the triggering vehicle corresponding to each lane to determine whether the triggering vehicle drops an object to form a road surface obstacle. This method utilizes the characteristic that "most road surface obstacles come from vehicle drops", effectively improving the efficiency of road surface monitoring for obstacle recognition.

[0071] Step three: Record the vehicles passing through the monitored section of the driving route as passing vehicles (the determination of passing vehicles can be known based on the driving route provided by the in-vehicle navigation software). Obtain the driving trajectory of the passing vehicles passing through the monitored section through the in-vehicle computer, and judge the suspected positions of obstacles based on the driving trajectory data of the passing vehicles, where:

[0072] Preset a time limit period. Obtain the passing vehicles passing through the monitored section within the time limit period as target vehicles. Obtain the driving routes of the target vehicles within the time limit period. Any point on the driving route corresponds to a passing time. Obtain the passing time interval corresponding to the monitored section;

[0073] Obtain the vertical acceleration change curve of the target vehicle within a time interval (obtained through a triaxial accelerometer inside the vehicle), obtain multiple peak points of the vertical acceleration change curve (i.e., the peak points of multiple peaks), preset an acceleration threshold, and record the moment corresponding to the peak point where the vertical acceleration value is greater than the acceleration threshold as the first abnormal moment;

[0074] Obtain the steering wheel torque (i.e., the force exerted by the driver to turn the steering wheel, which can be obtained through a sensor inside the steering wheel) change curve of the target vehicle within a time interval, preset a torque threshold, and record the part of the steering wheel torque change curve that is higher than the torque threshold as the torque abnormal part. Let the time interval corresponding to the torque abnormal part be and obtain the time interval the angle sequence of the steering wheel rotation within it The angle sequence contains multiple rotation angles and each rotation angle corresponds to a one-way rotation of the steering wheel (the rotation process of the steering wheel within the time interval can be split into multiple one-way rotations). Taking the clockwise rotation direction as the positive direction and the counterclockwise rotation direction as the negative direction, when the one-way rotation direction is the positive direction, the rotation angle is positive, and when the one-way rotation direction is the negative direction, the rotation angle is negative. Substitute into the formula for calculation to obtain the one-way rotation amplitude and the total rotation amplitude where:

[0075] ;

[0076] means taking the maximum value between ;

[0077] Substitute the one-way rotation amplitude and the total rotation amplitude into the formula for calculation to obtain the steering abnormality value, where are all preset weight coefficients. When the steering abnormality value is greater than the preset steering abnormality threshold, record the moment as the second abnormal moment;

[0078] It should be noted that when the driver discovers an obstacle and evades it, the driver will turn the steering wheel to make an emergency lane change, and at this time, the steering wheel will show a significantly abnormal rotation, which is different from the slight rotation during normal driving. Therefore, it is possible to judge whether the steering wheel conforms to the situation of evading an obstacle through the one-way angle and the overall angle during the steering wheel rotation.

[0079] The vehicle position corresponding to the first abnormal moment is obtained and denoted as the first abnormal position. The unit rectangle containing the first abnormal position is marked as the first abnormal rectangle. The vehicle position corresponding to the second abnormal moment is obtained and denoted as the second abnormal position. All unit rectangles corresponding to the lane where the second abnormal position is located are marked as the second abnormal rectangles.

[0080] By analyzing the driving data of passing vehicles to obtain the first abnormal rectangle and the second abnormal rectangle, it helps to locate the area where obstacles may exist. Thus, based on the recognition of road surface monitoring equipment, the driving data of each passing vehicle can be fully utilized to further verify the position of the obstacles, improving the diversity of obstacle detection methods and the accuracy of obstacle detection.

[0081] It should be noted that although road surface monitoring cameras can also capture the trajectory of vehicles passing through the monitored section, if trajectory analysis is performed on passing vehicles through road surface monitoring cameras, it means that the passing videos of multiple passing vehicles need to be saved simultaneously and trajectory extraction and further obstacle analysis need to be performed on these videos. This places extremely high requirements on the storage capacity and computing power of the monitoring equipment, and the actual implementation cost is too high. In contrast, using the trajectory data provided by the in-vehicle computer and adopting an independent analysis method of "one vehicle, one calculation" can fully share the computing power pressure while avoiding the interference of the routes of other vehicles, thereby improving the analysis efficiency and analysis accuracy.

[0082] Step 4: Combine the obstacle position analysis results in Step 2 and Step 3, and comprehensively analyze the confidence levels of obstacles at different positions, where:

[0083] The number of times each unit rectangle is marked as a road surface abnormal rectangle is obtained and denoted as the first influence value , the number of times each unit rectangle is marked as the first abnormal rectangle is obtained and denoted as the second influence value , the number of times each unit rectangle is marked as the second abnormal rectangle is obtained and denoted as the third influence value , substitute into the formula for calculation to obtain the abnormal confidence level of each unit rectangle, where are all preset weight coefficients, and m is the number of unit rectangles on each lane rectangle;

[0084] It should be noted that the abnormal confidence level of each unit rectangle is proportional to the number of times the unit rectangle is determined to be a road surface abnormal rectangle, the first abnormal rectangle, and the second abnormal rectangle. When a unit rectangle is determined to be an abnormal rectangle (regardless of which type of abnormal rectangle) more times, and is determined to be an abnormal rectangle under multiple determination methods, it means that the possibility of an obstacle existing in the area corresponding to the unit rectangle is high. Therefore, the abnormal confidence level of the unit rectangle can help the staff judge whether there is an obstacle in the area corresponding to the unit rectangle and the possibility of the existence of an obstacle.

[0085] Step Five: Preset a first confidence threshold and a second confidence threshold, where the first confidence threshold is less than the second confidence threshold. When the anomaly confidence of the unit rectangle is greater than or equal to the first confidence threshold and less than the second confidence threshold, generate a regional anomaly signal, retrieve the road surface monitoring image corresponding to the unit rectangle and mark the area corresponding to the unit rectangle, so as to facilitate the staff to timely check whether there are obstacles in this area and clean them in time if there are obstacles;

[0086] When the anomaly confidence of the unit rectangle is greater than or equal to the second confidence threshold, generate a regional avoidance signal, obtain the position of the area corresponding to the retrieved unit rectangle and record it as the obstacle area, obtain the driving routes of multiple passing vehicles, and record the vehicles whose current driving routes do not pass through the obstacle area as avoidance vehicles. When the avoidance vehicle is R meters away from the obstacle area, send an avoidance warning message to the avoidance vehicle through the navigation software. The avoidance warning message can be a voice reminder or a flashing dashboard indicator. The calculation formula for the distance R is , is the preset reaction time.

[0087] Adjust the output instruction based on the anomaly confidence according to the preset first confidence threshold and second confidence threshold, so that the method in the present invention can adjust the final output strategy according to the different confidence levels, so as to avoid the impact of misidentification of obstacles on traffic order as much as possible without affecting the obstacle handling speed (when an obstacle is misidentified, the confidence level is often not high, and at this time, a warning instruction will not be sent to all vehicles passing through this area), thereby ensuring the normal road traffic in the monitored section.

[0088] The present invention can combine the existing road surface monitoring and recognition technology with the data recorded by the vehicle body navigation software, so as to detect and locate the position of road surface obstacles from different angles, and further construct the confidence level by combining the results of different analysis methods, thereby improving the accuracy of the final output result. Compared with the single-angle detection and recognition technology in the prior art, it is less dependent on the number and types of vehicle's own sensors, is applicable to all vehicles, and is convenient to ensure the unity of the obstacle avoidance effect of all vehicles.

[0089] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps in the above method are implemented.

[0090] An obstacle detection device includes road surface monitoring devices arranged on both sides of the road, and the road surface monitoring devices face the road surface;

[0091] It further includes a vehicle-mounted device, which is electrically connected to a three-axis accelerometer and a torque sensor. The three-axis accelerometer is arranged at the front of the vehicle to measure the three-axis acceleration of the vehicle, and the torque sensor is arranged at the steering wheel shaft to measure the torque when the steering wheel rotates.

[0092] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. An obstacle detection method, characterized in that: The following steps are involved: Step 1: Obtain a road surface image, record the road section area corresponding to the road surface image as a monitoring section, divide the road surface image into a plurality of lane rectangles according to the road segmentation markings, and divide the lane rectangle into a plurality of unit rectangles along the lane length direction; Step 2: periodically acquire road images and record them as road periodic images. When a vehicle enters a monitored road section, acquire road images and record them as road vehicle images. Multiple road periodic images acquired after acquiring the road vehicle images and the road vehicle images together constitute a vehicle passing image set. Each vehicle passing image set corresponds to a triggering vehicle. Based on the vehicle image set analysis, multiple passing lanes corresponding to the triggering vehicle are obtained, and image comparison analysis is performed on multiple unit rectangles corresponding to the passing lanes, where: Obtaining rectangular images of the unit rectangle before and after the triggering vehicle passes through and comparing them, and marking the unit rectangle as a road surface abnormality rectangle based on the similarity between the two rectangular images; Step 3: Based on the analysis of the acceleration change and steering wheel torque change when the passing vehicle passes through the monitored road section, the unit rectangle is marked as the first abnormal rectangle or the second abnormal rectangle; Step 4: Based on the number of times the unit rectangle is marked, comprehensively calculate the abnormality confidence of the unit rectangle; Step 5: Compare the abnormality confidence of the unit rectangle with the preset first confidence threshold and second confidence threshold, and generate different signals based on the abnormality confidence of the unit rectangle.

2. The obstacle detection method according to claim 1, characterized in that: The process of constructing the vehicle passing image set is as follows: Get the speed limit value of the monitored section , substitute into the formula Calculate the correlation period ,in In order to monitor the length of the road section, the time of collecting road vehicle images is , will be in the time interval The road surface images collected periodically are recorded as The associated images of the road vehicle images at the moment, the road vehicle images and the corresponding associated images together form a vehicle passing image set.

3. The obstacle detection method according to claim 1, characterized in that: The process of obtaining the relevant lanes is as follows: The road surface image in the vehicle passing image set is recorded as the related image, the vehicle outline of the triggering vehicle is marked in the related image, the minimum rectangle containing the vehicle outline is constructed as the vehicle rectangle, and the straight line passing through the midpoints of the two short sides of the vehicle rectangle is constructed as the vehicle body direction straight line; In the relevant image, the lane center lines corresponding to each lane rectangle are drawn and recorded as , f is the serial number of the lane centerline; The intersection points of the vehicle body direction straight lines and the center lines of each lane in the relevant images are recorded as passing intersection points, and the lanes containing the passing intersection points are recorded as passing related lanes.

4. The obstacle detection method according to claim 2, characterized in that: The process of marking the road surface abnormality rectangle is as follows: Get the time intervals separately The two road surface comparison images collected most recently before and after are recorded as comparison images; Obtain two rectangular images corresponding to the same unit rectangle in the two comparison images, calculate the similarity of the two rectangular images, and when the similarity of the two rectangles is less than or equal to a preset similarity threshold, record the corresponding unit rectangle as an abnormal rectangular area; The similarities of multiple abnormal rectangular areas are obtained. When the similarities of multiple abnormal rectangular areas are not equal, the abnormal rectangular area is marked as a road surface abnormal rectangle, and the corresponding position is marked as a suspected obstacle position.

5. The obstacle detection method according to claim 1, characterized in that: The first abnormal rectangle marking process is as follows: A time limit period is preset, and vehicles passing through the monitored road section within the time limit period are obtained as target vehicles. The driving route of the target vehicle within the time limit period is obtained, and any point on the driving route corresponds to a passing time, and the passing time interval corresponding to the monitored road section is obtained; Obtaining a vertical acceleration change curve of the target vehicle within a time interval, obtaining multiple peak points of the vertical acceleration change curve, presetting an acceleration threshold, and recording the moment corresponding to the peak point where the vertical acceleration value is greater than the acceleration threshold as the first abnormal moment; The vehicle position corresponding to the first abnormal time is obtained and recorded as the first abnormal position, and the unit rectangle including the first abnormal position is marked as the first abnormal rectangle.

6. The obstacle detection method according to claim 5, characterized in that: The second abnormal rectangle marking process is as follows: Obtain a steering wheel torque change curve of the target vehicle within a time interval, preset a torque threshold, record the portion of the steering wheel torque change curve that is higher than the torque threshold as a torque abnormality portion, analyze the steering wheel rotation angle within the time interval corresponding to the torque abnormality portion, and obtain a steering abnormality value; The steering abnormality value is compared with a preset steering abnormality threshold value, and when the steering abnormality value is greater than the preset steering abnormality threshold value, the start time of the torque abnormal part is recorded as the second abnormal time; The vehicle position corresponding to the second abnormal moment is obtained and recorded as the second abnormal position, and all unit rectangles corresponding to the lane where the second abnormal position is located are marked as second abnormal rectangles.

7. The obstacle detection method according to claim 6, characterized in that: The steering outlier value calculation process is specifically as follows: Assume that the time interval corresponding to the abnormal torque is , get the time interval Angle series of inner steering wheel rotation , the angle sequence contains multiple rotation angles , each rotation angle corresponds to a one-way rotation of the steering wheel; When the one-way rotation direction is clockwise, the rotation angle Positive, when the unidirectional rotation direction is counterclockwise, the rotation angle is negative; Substitute into the formula Calculation is performed in Get the one-way rotation amplitude and the total rotation amplitude ; in: ； Indicates taking The maximum value between The one-way rotation range and the total rotation amplitude Substitute into the formula Calculate in and get the steering anomaly value, where All are preset weight coefficients.

8. The obstacle detection method according to claim 1, characterized in that: The calculation process of the abnormal confidence of the unit rectangle is as follows: Obtain the number of times each unit rectangle is marked as a road surface anomaly rectangle and record it as the first impact value , obtain the number of times each unit rectangle is marked as the first abnormal rectangle and record it as the second influence value , obtain the number of times each unit rectangle is marked as the second abnormal rectangle and record it as the third influence value ; Substitute into the formula Calculate the abnormal confidence of each unit rectangle, where: All are preset weight coefficients; m is the number of unit rectangles on each lane rectangle.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images