Information detection method, device and electronic equipment

Abstract

The present invention provides an information detection method, device and electronic device. The method comprises: obtaining a target video stream collected for a target area, wherein the target area comprises M lanes, where M is an integer greater than 1; tracking vehicles in the target area based on the target video stream to obtain target information, wherein the target information comprises lane information of each lane in the M lanes and vehicle information of vehicles in the target area, wherein the lane information comprises a lane change rate of a vehicle from one lane to another lane; and detecting vehicle traffic information of the target lane based on the vehicle information of the vehicle in the target lane when the lane information of the target lane in the M lanes satisfies a preset condition, wherein the preset condition comprises: the lane change rate of the lane is greater than a first preset threshold. The embodiment of the present invention can improve the detection and recognition rate of vehicle traffic accidents, thereby improving the accuracy of accident warning.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of intelligent driving technology, and in particular to an information detection method, device and electronic equipment. Background Art

[0002] With the rapid expansion of urban scale and the rapid increase in the number of traffic vehicles, vehicle traffic accidents are increasing day by day. Rapid detection and early warning of vehicle traffic accidents can reduce property losses and casualties caused by untimely rescue.

[0003] At present, vehicle traffic accidents are usually detected through on-board multi-sensor fusion perception technology. However, multi-sensor fusion requires the fusion of multiple sensor information and relies on high-performance computing resources, resulting in a relatively low detection and recognition rate of vehicle traffic accidents. Summary of the invention

[0004] The embodiments of the present invention provide an information detection method, device and electronic device to solve the problem of low detection and recognition rate of vehicle traffic accidents in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides an information detection method, the method comprising:

[0006] Obtain a target video stream collected for a target area, where the target area includes M lanes, where M is an integer greater than 1;

[0007] Tracking the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane;

[0008] When lane information of a target lane among the M lanes satisfies a preset condition, vehicle traffic information of the target lane is detected based on vehicle information of vehicles in the target lane, and the preset condition includes: a lane change rate of the lane is greater than a first preset threshold.

[0009] In a second aspect, an embodiment of the present invention provides an information detection device, the device comprising:

[0010] An acquisition module, used to acquire a target video stream collected for a target area, wherein the target area includes M lanes, where M is an integer greater than 1;

[0011] a tracking module, configured to track the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane;

[0012] A detection module is used to detect vehicle traffic information of the target lane based on vehicle information of vehicles in the target lane when lane information of the target lane among the M lanes meets a preset condition, wherein the preset condition includes: a lane change rate of the lane is greater than a first preset threshold.

[0013] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the above-mentioned information detection method when executed by the processor.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned information detection method are implemented.

[0015] In an embodiment of the present invention, a target video stream collected for a target area is obtained, and the target area includes M lanes; based on the target video stream, vehicles in the target area are tracked to obtain target information, and the target information includes lane information of each lane in the M lanes and vehicle information of vehicles in the target area, and the lane information includes a lane change rate of vehicles traveling from one lane to another lane; when there is a target lane among the M lanes, the vehicle traffic information of the target lane is detected based on the vehicle information of the vehicle in the target lane, and the preset condition includes: the lane change rate of the lane is greater than a first preset threshold. In this way, by detecting the vehicle traffic information of the target lane based on the vehicle information of the vehicle in the target lane when the lane information including the lane change rate of the target lane meets the preset condition, the lane where a vehicle traffic accident may occur can be quickly locked, thereby improving the detection and recognition rate of vehicle traffic accidents, and further improving the accuracy of accident warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0017] Figure 1 is a flow chart of an information detection method provided by an embodiment of the present invention;

[0018] Figure 2 is a schematic diagram of the calibration of lanes and parking lines in the image of the target video stream;

[0019] Figure 3 It is a schematic diagram of the timing flow of the information detection system for detecting vehicle traffic accidents;

[0020] Figure 4 is a schematic diagram of the structure of an information detection device provided by an embodiment of the present invention;

[0021] Figure 5 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In related technologies, vehicle traffic accidents are usually detected through on-board multi-sensor fusion perception technology. Multi-sensor fusion (MSF) is an information processing process that uses computer technology to automatically analyze and integrate information and data from multiple sensors or multiple sources according to certain criteria to complete the required decision-making and estimation. The current mainstream multi-sensor fusion solution is to fuse information and data from sensors such as vision, lidar, and millimeter-wave radar.

[0023] However, the warning recognition range of vehicle-mounted multi-sensor fusion is limited to the range that the vehicle's own sensors can detect, and the detection range is relatively small. At the same time, multi-sensor fusion requires the fusion of multiple sensor information, relying on high-performance computing resources, and the calculation speed is relatively slow. In addition, multi-sensor fusion cannot handle vehicle traffic accident detection under occlusion. Based on this, the embodiment of the present invention provides a new information detection solution that can well solve the problems in the prior art.

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The following first describes the information detection method provided by the embodiment of the present invention.

[0026] It should be noted that the information detection method provided by the embodiment of the present invention relates to the field of intelligent driving technology, which can be widely used in multiple scenarios such as intelligent driving and map navigation. The method can be executed by the information detection device of the embodiment of the present invention. The information detection device can be configured in any electronic device to execute the information detection method, and the electronic device can be a server or a terminal, which is not specifically limited here.

[0027] In one scenario, the electronic device may be a perception server, which may detect vehicle traffic information in the target area based on the target video stream collected in the target area to detect vehicle traffic accidents in the target area, and may send relevant information about the vehicle traffic accident to the vehicle to everything (V2X) platform when a vehicle traffic accident is detected. The V2X platform may broadcast relevant information about the vehicle traffic accident to an on-board unit (OBU) that is relatively close to the location of the vehicle traffic accident to provide accident warning. This scenario detects vehicle traffic accidents in the target area through a perception server, and broadcasts vehicle traffic accidents through the V2X platform, which can increase the scope of accident warnings.

[0028] See also Figure 1 , the figure shows a schematic flow chart of the information detection method provided by an embodiment of the present invention.

[0029] like Figure 1 As shown, the method may include the following steps:

[0030] Step 101, obtaining a target video stream captured for a target area, where the target area includes M lanes.

[0031] Wherein, M is an integer greater than 1.

[0032] Here, the target area may refer to any area that can be captured by roadside equipment such as a camera, and may be an intersection area or an area that does not include an intersection, and is not specifically limited here.

[0033] The target video stream may include at least two frames of images taken in chronological order, and the target video stream may be obtained by taking pictures by a roadside device deployed on one side, or by taking pictures by a roadside device deployed on both sides, which is not specifically limited here. In an optional implementation, the acquisition of the target video stream collected for the target area includes:

[0034] Acquire a first video stream and a second video stream collected for a target area, wherein the first video stream and the second video stream are video streams collected for both sides of the M lanes respectively;

[0035] The first video stream and the second video stream are merged to obtain the target video stream.

[0036] That is, in this embodiment, the target video stream is obtained by shooting with roadside equipment deployed on both sides, which can improve the shooting coverage of the target area and improve the detection of vehicle traffic accidents under occlusion. Usually, the roadside equipment on both sides is arranged perpendicular to the M lanes. In addition, the setting height of the roadside equipment, that is, the video shooting height, can be 6 meters to minimize the occlusion.

[0037] The target area may include at least two lanes, and lane calibration may be performed on each frame of the target video stream by manual or image recognition technology to calibrate the M lanes in the target area. Meanwhile, if the target area is an intersection area, stop line calibration may also be performed on each frame of the target video stream to calibrate the stop line in the target area. Figure 2 As shown, a certain intersection area may include multiple lanes, and lanes 201 and stop lines 202 are marked out respectively for subsequent vehicle traffic accident detection.

[0038] In addition, when the embodiment of the present invention is applied to a V2X platform, it is possible to receive a target video stream sent by a roadside device, or receive a first video stream and a second video stream sent by two roadside devices, fuse the first video stream and the second video stream, and finally obtain the target video stream.

[0039] The steps of fusing the first video stream and the second video stream are as follows:

[0040] Video image denoising: For each frame of the first video stream and the corresponding frame of the second video stream, a low-pass filtering algorithm can be used to denoise the two frames of images respectively;

[0041] Feature extraction: Use image segmentation and edge detection technology to extract two feature area blocks of the common part of the image to be stitched from the two frames of images, calculate the centroid position of the feature block, and use the centroid position to stitch the two frames of images;

[0042] Video image registration: Use the distance between the feature set of the reference image and the image to be stitched and the spatial distribution of the feature set to perform matching calculations;

[0043] Determine a suitable image fusion algorithm to fuse the two frames of video image registration;

[0044] Evaluate the results of image fusion; there are two main aspects of image fusion performance evaluation, namely subjective evaluation and objective evaluation. Since there is no ideal image source in practice, a more easily achievable evaluation standard (such as computational entropy) is generally used, combined with subjective vision to give a more reasonable evaluation;

[0045] If the evaluation result is not satisfactory, the parameters are adjusted, the image fusion is performed again, and then the image fusion result is re-evaluated; wherein, the parameter estimation method can be: determining the transformation relationship between the two images by the matching relationship between some elements between the images, transforming the image to be spliced ​​into the coordinate system of the reference image according to the transformation model, the establishment of the transformation model includes the selection of the type of transformation function and the calculation of the parameters in the model, and the type of transformation model should correspond to the geometric deformation between the reference image and the image to be spliced, the image acquisition method and the required splicing accuracy.

[0046] Finally, the image fusion result is output.

[0047] Step 102: Track the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, and the lane information includes a lane change rate of the vehicle from one lane to another lane.

[0048] In this step, the vehicle in the target area can be tracked based on the target video stream to obtain target information, wherein the target information may include lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, and the lane information may include a lane change rate of the vehicle from one lane to another lane.

[0049] Among them, for a lane, the lane change rate of the lane can be an entry rate, an exit rate, or a combination of the entry rate and the exit rate. The entry rate evaluates the probability of entering the lane from other lanes, and the exit rate evaluates the probability of entering or exiting the lane from the lane to other lanes.

[0050] If a vehicle stops on the road after an accident, the vehicles behind it will change lanes and go around the faulty vehicle. There will also be vehicles changing lanes from other lanes in front of the faulty vehicle. Therefore, the lane change rate can be used to evaluate whether to start the accident detection program, which can improve the accuracy of vehicle traffic accident detection.

[0051] Lane information may also include at least one of the following: average vehicle speed of the lane and vehicle occupancy rate of the lane. The average vehicle speed of the lane may refer to the average speed of vehicles passing through the lane, and the vehicle occupancy rate of the lane may refer to the ratio of the total length of all vehicles in the detection area selected on the lane at a certain moment to the total length of the detection area.

[0052] Vehicle information may include vehicle position, driving speed, parking time and parking duration, etc. Driving speed can be obtained by converting the pixel coordinates output by the tracking algorithm into world coordinates. Specifically, the vehicle position of two adjacent frames of images can be obtained by converting the pixel coordinates output by the tracking algorithm into world coordinates, and then the time difference between the two adjacent frames of images is recorded. The distance traveled by the vehicle is determined based on the vehicle position of the two adjacent frames of images, and the driving speed of the vehicle can be calculated by dividing the distance by the time difference. The coordinate transformation formula is shown in the following formula (1).

[0053]

[0054] In the above formula (1), M is the camera intrinsic parameter matrix, R is the rotation matrix, t is the translation matrix, x, y and z are the coordinates of the world coordinate system, z is the height of the world coordinate system, which can be set to 0, and u and v are the pixel coordinates output by the tracking algorithm.

[0055] The parking time of the vehicle can be determined based on the vehicle's driving speed. For example, when the vehicle's driving speed is tracked to be close to 0, it can be determined that the vehicle has stopped and the current parking time is recorded. At the same time, the vehicle is continuously tracked in subsequent video streams to record the driving time of the vehicle.

[0056] During a period of tracking, the average speed of the vehicle in the period can be calculated. For a lane, the average speed of the vehicles in the lane can be calculated by the average speed of all vehicles tracked in the lane during the period. The calculation formula is shown in the following formula (2).

[0057]

[0058] In the above formula (2), represents the average speed of vehicles in the lane, n represents the number of vehicles tracked in the lane during this period of time, and v i It refers to the average speed of the i-th vehicle tracked during that period of time.

[0059] The vehicle occupancy rate of the lane can be calculated by the following formula (3).

[0060]

[0061] In the above formula (3), R represents the spatial occupancy, L represents the total length of the observed road, and S i represents the length of the tracked i-th vehicle, and k represents the total number of vehicles traveling on the observed road at a certain moment.

[0062] Step 103, when there is a target lane among the M lanes whose lane information meets a preset condition, the vehicle traffic information of the target lane is detected based on the vehicle information of the vehicle in the target lane, and the preset condition includes: a lane change rate greater than a first preset threshold.

[0063] In this step, when the lane information of the target lane among the M lanes meets the preset conditions, the accident detection program can be started, and the vehicle traffic information of the target lane can be detected based on the vehicle information of the vehicle in the target lane.

[0064] Among them, the preset conditions may include that the lane change rate is greater than a first preset threshold. The first preset threshold can be set according to actual conditions, which indicates that the lane change rate is relatively large during a period of continuous tracking. For example, during a period of continuous tracking, the lane change rate of a certain lane is greater than 8 times during this period. At this time, it can be explained that the lane change rate of the lane is relatively abnormal compared to normal conditions. Accordingly, the accident detection program can be started.

[0065] The preset condition may also include at least one of the following:

[0066] The average speed of vehicles in the lane is less than a fifth preset threshold;

[0067] The vehicle occupancy rate of the lane is greater than a sixth preset threshold.

[0068] Among them, the fifth preset threshold can be set according to actual conditions. In an optional implementation, the fifth preset threshold can be determined according to the historical speed of vehicles in the lane. For example, the fifth preset threshold can be the historical speed of vehicles in the lane. When the average speed of vehicles in the lane is less than the fifth preset threshold, it means that the average speed of vehicles in the lane is less than the historical speed of vehicles in the lane. The historical speed of vehicles in the lane can be given by the tracking algorithm and can be updated regularly. The sixth preset threshold can also be set according to actual conditions, such as being set to 80%.

[0069] In other words, lane information such as lane change rate, average vehicle speed in the lane, and vehicle occupancy rate in the lane can be combined to comprehensively determine whether to start the accident detection program. This can improve the accuracy of vehicle traffic accident detection and at the same time reduce the probability of false detection of vehicle traffic accidents.

[0070] When the lane information of the target lane meets the preset conditions, it means that a vehicle traffic accident has occurred in the target lane to a large extent. Accordingly, the accident detection program can be started, and the vehicle traffic information of the target lane can be detected based on the vehicle information of the vehicle in the target lane.

[0071] The vehicle traffic information may include a vehicle traffic accident result, which may include two situations, one of which is that a vehicle traffic accident exists in the target lane, and the other is that there is no vehicle traffic accident in the target lane. In the case of a vehicle traffic accident in the target lane, the vehicle traffic information may also include fault information, which may include vehicle information of the faulty vehicle, such as the vehicle location, the length of time the vehicle has been parked, etc.

[0072] For example, when there is a target vehicle in the target lane, it can be determined that there is a vehicle traffic accident in the target lane. The target vehicle can be a vehicle whose parking time in the target lane is longer than a second preset threshold, and the second preset threshold can be set according to actual conditions, indicating that the parking time of the vehicle is relatively long.

[0073] For another example, when there is a target vehicle in the target lane, and the abnormal vehicle determined based on the target vehicle is greater than the fourth preset threshold from the stop line at the intersection, it is determined that there is a vehicle traffic accident in the target lane. In this scenario, the occurrence of misjudgment of vehicle traffic accidents caused by vehicle stops caused by traffic lights can be reduced, and the accuracy of vehicle traffic accidents can be improved. The fourth preset threshold can be set according to actual conditions, such as being set to the distance between two parking spaces.

[0074] In this embodiment, a target video stream collected for a target area is obtained, and the target area includes M lanes; based on the target video stream, vehicles in the target area are tracked to obtain target information, and the target information includes lane information of each lane in the M lanes and vehicle information of vehicles in the target area, and the lane information includes a lane change rate of vehicles traveling from one lane to another lane; when the lane information of a target lane in the M lanes satisfies a preset condition, the vehicle traffic information of the target lane is detected based on the vehicle information of the vehicle in the target lane, and the preset condition includes: the lane change rate of the lane is greater than a first preset threshold. In this way, by detecting the vehicle traffic information of the target lane based on the vehicle information of the vehicle in the target lane when the lane information including the lane change rate of the target lane satisfies the preset condition, the lane where a vehicle traffic accident may occur can be quickly locked, thereby improving the detection and recognition rate of vehicle traffic accidents, and further improving the accuracy of accident warning.

[0075] Furthermore, when the embodiment of the present invention is applied to a perception server, the perception server can detect vehicle traffic information in the target area based on the target video stream collected in the target area to detect vehicle traffic accidents in the target area, and can send relevant information about the vehicle traffic accident to the V2X platform when a vehicle traffic accident is detected. The V2X platform can broadcast relevant information about the vehicle traffic accident to an OBU that is relatively close to the location of the vehicle traffic accident to provide accident warning. This scenario uses the perception server to detect vehicle traffic accidents in the target area and broadcasts vehicle traffic accidents through the V2X platform, which can increase the scope of accident warnings.

[0076] Figure 3 It is a schematic diagram of the timing flow of the information detection system for vehicle traffic accident detection, such as Figure 3 As shown, the information detection system may include a roadside camera, a perception server, a V2X platform and an OBU.

[0077] The roadside camera can capture a video stream of the target area and push the video stream to the perception server in the format of a certain protocol, such as a private protocol.

[0078] The perception server can fuse the two video streams to obtain a target video stream, and track the vehicles in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicles in the target area, and the lane information includes the lane change rate of the vehicle from one lane to another lane; when the lane information of the target lane among the M lanes meets a preset condition, the vehicle traffic information of the target lane is detected based on the vehicle information of the vehicles in the target lane.

[0079] When a vehicle traffic accident is detected, the relevant information of the vehicle traffic accident is sent to the V2X platform in the form of a message. Specifically, after the perception server obtains the traffic accident detection result, it can generate a V2X national standard message such as a Road Side Information (RSI) message, and send the RSI message to Kafka through the network. The V2X platform consumes the Kafka data to obtain the traffic accident detection result.

[0080] The RSI message may include a field RTEData, which may include two fields, namely, a field EventType and a field eventPos. The field EventType may be filled with the vehicle traffic accident result, and the field eventPos may be filled with the latitude and longitude information of the vehicle traffic accident. After the RSI message is assembled, the RSI message is sent to the V2X platform.

[0081] After receiving the RSI message, the V2X platform determines whether the RSI message meets the forwarding conditions based on its latitude and longitude and map information. When the forwarding conditions are met, such as the distance between the OBU and the location of the vehicle traffic accident is less than a certain threshold such as 500 meters, the V2X platform can forward the RSI message to the relevant OBU to remind the OBU that there is a vehicle traffic accident here, so as to issue an accident alarm.

[0082] The OBU performs corresponding emergency processing according to the received RSI message.

[0083] Optionally, the target video stream includes N frames of images, where N is an integer greater than 1, and step 102 specifically includes:

[0084] For each frame of the N frames of images, identifying a vehicle in the target area based on the image to obtain position information of the vehicle in the target area;

[0085] Determine the lane number where the vehicle is located according to the position information of the vehicle in the target area;

[0086] For each of the M lanes, when it is determined that the lane number of the vehicle for the lane has changed based on at least two consecutive frames of the N frames of images, the lane change rate of the lane is determined based on the number of times the lane number of the vehicle for the lane in the target area has changed.

[0087] In this embodiment, the lane change rate of the lane can be determined by tracking the lane number of the vehicle. Specifically, the position information of the vehicle in each frame of the image is calculated according to the vehicle tracking algorithm, and the lane number of the vehicle can be calculated according to the position information. For each lane in the M lanes, when it is found that the lane number of the vehicle in the lane has changed based on at least two consecutive frames of the N frames of images, the lane rate of the lane is increased by 1.

[0088] Among them, there may be two situations in which the lane number of the vehicle in the lane changes. The first situation may be the lane exit number of the lane, and the lane exit number of the lane refers to the number of lanes switched from the lane to other lanes. The second situation may be the lane entry number of the lane, and the lane entry number of the lane refers to the number of lanes switched from other lanes to this lane.

[0089] The number of times the lane number of vehicles in the lane changes may refer to the sum of the number of lane exits and the number of lane entrys of the lane, that is, the sum of the number of lane exits and the number of lane entrys of the lane may be determined as the lane change rate of the lane.

[0090] In this embodiment, for each of the N frames of images, the vehicle in the target area is identified based on the image to obtain the position information of the vehicle in the target area; the lane number of the vehicle is determined based on the position information of the vehicle in the target area; for each of the M lanes, when it is determined based on at least two consecutive frames of the N frames of images that the lane number of the vehicle in the lane changes, the lane change rate of the lane is determined based on the number of times the lane number of the vehicle in the target area changes. In this way, accurate monitoring of the lane change rate of each lane can be achieved.

[0091] Optionally, the vehicle information includes the parking time of the vehicle, and step 103 specifically includes:

[0092] In a case where there is a target vehicle in the target lane, vehicle traffic information of the target lane is determined based on the target vehicle, and the target vehicle is a vehicle whose parking time in the target lane is greater than a second preset threshold.

[0093] In this embodiment, when the accident detection program is started, it is possible to detect whether there is a target vehicle in the target lane. Specifically, the parking time of the vehicle in the target lane can be determined based on the driving speed of the vehicle in the target lane. Accordingly, when it is detected that the driving speed of the vehicle is almost 0, the vehicle can be continuously monitored to determine the parking time of the vehicle. When the parking time of the vehicle is greater than the second threshold, the vehicle can be determined to be the target vehicle.

[0094] The second preset threshold can be set according to actual conditions and is not specifically limited here.

[0095] In the case where there is a target vehicle in the target lane, it can be indicated that there is a faulty vehicle in the target lane to a large extent. Accordingly, the vehicle traffic information of the target lane can be determined based on the target vehicle.

[0096] The vehicle traffic information may include a vehicle traffic accident result. In an optional implementation, when there is a target vehicle in a target lane, it can be determined that there is a vehicle traffic accident in the target lane. Accordingly, the vehicle traffic accident result is that there is a vehicle traffic accident in the target lane.

[0097] In another optional embodiment, when there is a target vehicle in the target lane, it can be determined based on the target vehicle and the stop line whether the abnormal vehicle stops due to a traffic light. When it is determined that the abnormal vehicle does not stop due to a traffic light, it can be determined that there is a vehicle traffic accident in the target lane.

[0098] In the event that there is a vehicle traffic accident in the target lane, the vehicle traffic information may also include fault information. The vehicle information of the target vehicle (including vehicle location, parking duration, etc.) may be determined as the fault information, or the vehicle information of the target vehicle and a collection of abnormal vehicles in its surroundings may be determined as the fault information.

[0099] In this embodiment, when there is a target vehicle in the target lane, the vehicle traffic information of the target lane is determined based on the target vehicle, and the target vehicle is a vehicle whose parking time in the target lane is greater than a second preset threshold. In this way, the vehicle traffic information can be detected based on the abnormal vehicle.

[0100] Optionally, the vehicle traffic information includes a vehicle traffic accident result of the target lane, and determining the vehicle traffic information of the target lane based on the target vehicle includes:

[0101] Determine a first vehicle set in the target lane based on the target vehicle, the first vehicle set including a first vehicle in the target lane in a lane line direction from the target vehicle to the target lane, the first vehicle being a vehicle whose parking time is greater than a third preset threshold;

[0102] When the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, the vehicle traffic accident result is determined to be a vehicle traffic accident occurring in the target lane, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane.

[0103] In this embodiment, the first vehicle set in the target lane can be determined based on the target vehicle. Specifically, abnormal vehicles can be searched in the lane line direction of the target lane based on the vehicle position of the target vehicle. Other abnormal vehicles can be searched within the range of the target vehicle's own identification box size, in the lane line direction of the target lane plus the vehicle body position.

[0104] Afterwards, if another abnormal vehicle is found, the abnormal vehicle can be used as a starting point to search for other abnormal vehicles in the direction of the lane line of the target lane until the last abnormal vehicle in the target area in the direction of the lane line of the target lane is found. When the target area is an intersection area, the abnormal vehicle with the closest stop line in the direction of the lane line of the target lane can be found. Among them, the first vehicle set can refer to the first vehicle searched in the direction of the lane line of the target lane from the target vehicle, and the first vehicle is a vehicle whose parking time is greater than the third preset threshold. The third preset threshold can be set according to actual conditions, and the third preset threshold can be set equal to the second preset threshold.

[0105] When the target area is an intersection area, in order to avoid false detection, the vehicle traffic accident result can be determined as a vehicle traffic accident in the target lane and the target vehicle is the accident vehicle only when the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold. Otherwise, it is considered that the target vehicle has stopped due to a traffic light and there is no vehicle traffic accident in the target lane.

[0106] When it is determined that there is a vehicle traffic accident in the target lane, the vehicle traffic information also includes fault information. The fault information may include vehicle information of the target vehicle, or may include vehicle information of the first vehicle set and vehicle information of the target vehicle, or may also include vehicle information of other abnormal vehicles, which is not specifically limited here.

[0107] In this embodiment, the first vehicle set in the target lane is determined based on the target vehicle, the first vehicle set includes the first vehicle in the target lane in the direction of the lane line of the target vehicle to the target lane, and the first vehicle is a vehicle whose parking time is greater than a third preset threshold; when the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, the vehicle traffic accident result is determined to be that there is a vehicle traffic accident in the target lane, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane. In this way, it is possible to avoid the false detection of a vehicle traffic accident caused by the target vehicle stopping due to a traffic light, thereby improving the detection accuracy of vehicle traffic accidents.

[0108] Optionally, when it is determined that a vehicle traffic accident exists in the target lane, the vehicle traffic information includes fault information, and the determining of the vehicle traffic information of the target lane based on the target vehicle further includes:

[0109] Determine a second vehicle set in the target lane based on the target vehicle, the second vehicle set including a second vehicle in the target lane in a direction opposite to the lane line direction of the target lane from the target vehicle, the second vehicle being a vehicle whose parking time is greater than the third preset threshold;

[0110] The vehicle information of the target vehicle, the first vehicle set and the second vehicle set are aggregated to obtain the fault information.

[0111] In this embodiment, the second vehicle set in the target lane can be determined based on the target vehicle. Specifically, abnormal vehicles can be searched in the opposite direction of the lane line of the target lane based on the vehicle position of the target vehicle. Other abnormal vehicles can be searched within the range of the target vehicle's own identification box size, in the opposite direction of the lane line of the target lane plus the vehicle body position.

[0112] Afterwards, if another abnormal vehicle is found, the abnormal vehicle can be used as a starting point to search for other abnormal vehicles in the opposite direction of the lane line of the target lane until the last abnormal vehicle in the target area in the opposite direction of the lane line of the target lane is found. In the case where the target area is an intersection area, the abnormal vehicle with the closest stop line in the opposite direction of the lane line of the target lane can be found. Among them, the second vehicle set can refer to the second vehicle searched in the opposite direction of the lane line of the target lane from the target vehicle, and the second vehicle is a vehicle whose parking time is greater than the third preset threshold.

[0113] The target vehicle, the first vehicle set, and the second vehicle set may all be aggregated into an abnormal vehicle set, and accordingly, the vehicle information of the abnormal vehicle set may be fault information.

[0114] In this embodiment, the second vehicle set in the target lane is determined based on the target vehicle, the second vehicle set includes the second vehicle in the target lane in the opposite direction of the lane line direction from the target vehicle to the target lane, and the second vehicle is a vehicle whose parking time is longer than the third preset threshold; the vehicle information of the target vehicle, the first vehicle set and the second vehicle set are aggregated to obtain the fault information. In this way, accurate detection of abnormal vehicle sets can be achieved.

[0115] The information detection device provided by the embodiment of the present invention is described below.

[0116] See also Figure 4 , the figure shows a schematic diagram of the structure of the information detection device provided by an embodiment of the present invention.

[0117] like Figure 4 As shown, the information detection device 400 includes:

[0118] An acquisition module 401 is used to acquire a target video stream collected for a target area, where the target area includes M lanes, where M is an integer greater than 1;

[0119] A tracking module 402, configured to track the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane;

[0120] The detection module 403 is used to detect the vehicle traffic information of the target lane based on the vehicle information of the vehicle in the target lane when the lane information of the target lane in the M lanes meets the preset conditions, and the preset conditions include: the lane change rate is greater than the first preset threshold.

[0121] Optionally, the target video stream includes N frames of images, where N is an integer greater than 1, and the tracking module 402 includes:

[0122] an identification submodule, for identifying a vehicle in the target area based on each image in the N frames of images, and obtaining position information of the vehicle in the target area;

[0123] A first determination submodule, configured to determine a lane number where the vehicle is located according to the position information of the vehicle in the target area;

[0124] The second determination submodule is used to determine, for each of the M lanes, a lane change rate of the lane based on the number of times the lane number of the vehicle in the target area changes when it is determined based on at least two consecutive frames of the N frames that the lane number of the vehicle in the lane changes.

[0125] Optionally, the vehicle information includes the parking duration of the vehicle, and the detection module 403 includes:

[0126] The third determination submodule is used to determine the vehicle traffic information of the target lane based on the target vehicle when there is a target vehicle in the target lane, and the target vehicle is a vehicle whose parking time in the target lane is greater than a second preset threshold.

[0127] Optionally, the vehicle traffic information includes a vehicle traffic accident result in the target lane, and the third determination submodule includes:

[0128] A first determining unit is used to determine a first vehicle set in the target lane based on the target vehicle, the first vehicle set including a first vehicle in the target lane in a lane line direction from the target vehicle to the target lane, the first vehicle being a vehicle whose parking time is greater than a third preset threshold;

[0129] A second determination unit is used to determine that the vehicle traffic accident result is that there is a vehicle traffic accident in the target lane when the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane.

[0130] Optionally, when it is determined that a vehicle traffic accident exists in the target lane, the vehicle traffic information includes fault information, and the third determining submodule further includes:

[0131] a third determining unit, configured to determine a second vehicle set in the target lane based on the target vehicle, the second vehicle set including a second vehicle in the target lane in a direction opposite to the lane line direction of the target lane from the target vehicle, the second vehicle being a vehicle whose parking time is greater than the third preset threshold;

[0132] The aggregation unit is used to aggregate the vehicle information of the target vehicle, the first vehicle set and the second vehicle set to obtain the fault information.

[0133] Optionally, the preset condition also includes at least one of the following:

[0134] The average speed of vehicles in the lane is less than a fifth preset threshold;

[0135] The vehicle occupancy rate of the lane is greater than a sixth preset threshold.

[0136] Optionally, the acquisition module 401 is specifically used to:

[0137] Acquire a first video stream and a second video stream collected for a target area, wherein the first video stream and the second video stream are video streams collected for both sides of the M lanes respectively;

[0138] The first video stream and the second video stream are merged to obtain the target video stream.

[0139] The information detection device 400 can implement each process implemented in the above-mentioned information detection method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described here.

[0140] The electronic device provided by the embodiment of the present invention is described below.

[0141] See also Figure 5 , the figure shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 5 As shown, the electronic device 500 includes: a processor 501 , a memory 502 , a user interface 503 and a bus interface 504 .

[0142] The processor 501 is used to read the program in the memory 502 and execute the following process:

[0143] Obtain a target video stream collected for a target area, where the target area includes M lanes, where M is an integer greater than 1;

[0144] Tracking the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane;

[0145] When lane information of a target lane among the M lanes satisfies a preset condition, vehicle traffic information of the target lane is detected based on vehicle information of vehicles in the target lane, and the preset condition includes: a lane change rate greater than a first preset threshold.

[0146] exist Figure 5 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 501 and memory represented by memory 502. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface 504 provides an interface. For different user devices, the user interface 503 may also be an interface that can connect external or internal devices, and the connected devices include but are not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0147] The processor 501 is responsible for managing the bus architecture and general processing, and the memory 502 can store data used by the processor 501 when performing operations.

[0148] Optionally, the target video stream includes N frames of images, where N is an integer greater than 1, and the processor 501 is further configured to:

[0149] For each frame of the N frames of images, identifying a vehicle in the target area based on the image to obtain position information of the vehicle in the target area;

[0150] Determine the lane number where the vehicle is located according to the position information of the vehicle in the target area;

[0151] For each of the M lanes, when it is determined that the lane number of the vehicle for the lane has changed based on at least two consecutive frames of the N frames of images, the lane change rate of the lane is determined based on the number of times the lane number of the vehicle for the lane in the target area has changed.

[0152] Optionally, the vehicle information includes the parking time of the vehicle. The processor 501 is further configured to:

[0153] In a case where there is a target vehicle in the target lane, vehicle traffic information of the target lane is determined based on the target vehicle, and the target vehicle is a vehicle whose parking time in the target lane is greater than a second preset threshold.

[0154] Optionally, the vehicle traffic information includes a vehicle traffic accident result in the target lane, and the processor 501 is further configured to:

[0155] Determine a first vehicle set in the target lane based on the target vehicle, the first vehicle set including a first vehicle in the target lane in a lane line direction from the target vehicle to the target lane, the first vehicle being a vehicle whose parking time is greater than a third preset threshold;

[0156] When the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, the vehicle traffic accident result is determined to be a vehicle traffic accident occurring in the target lane, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane.

[0157] Optionally, when it is determined that a vehicle traffic accident exists in the target lane, the vehicle traffic information includes fault information, and the processor 501 is further configured to:

[0158] Determine a second vehicle set in the target lane based on the target vehicle, the second vehicle set including a second vehicle in the target lane in a direction opposite to the lane line direction of the target lane from the target vehicle, the second vehicle being a vehicle whose parking time is greater than the third preset threshold;

[0159] The vehicle information of the target vehicle, the first vehicle set and the second vehicle set are aggregated to obtain the fault information.

[0160] Optionally, the preset condition also includes at least one of the following:

[0161] The average speed of vehicles in the lane is less than a fifth preset threshold;

[0162] The vehicle occupancy rate of the lane is greater than a sixth preset threshold.

[0163] Optionally, the processor 501 is further configured to:

[0164] Acquire a first video stream and a second video stream collected for a target area, wherein the first video stream and the second video stream are video streams collected for both sides of the M lanes respectively;

[0165] The first video stream and the second video stream are merged to obtain the target video stream.

[0166] Preferably, an embodiment of the present invention further provides an electronic device, comprising a processor 501, a memory 502, and a computer program stored in the memory 502 and executable on the processor 501. When the computer program is executed by the processor 501, each process of the above-mentioned information detection method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0167] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned information detection method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0168] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0169] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0170] In the embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0171] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0172] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0173] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0174] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An information detection method, It is characterized in that The method comprises: Obtain a target video stream collected for a target area, where the target area includes M lanes, where M is an integer greater than 1; Tracking the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane; In the case where lane information of a target lane among the M lanes satisfies a preset condition, detecting vehicle traffic information of the target lane based on vehicle information of vehicles in the target lane, wherein the preset condition includes: a lane change rate of the lane is greater than a first preset threshold; The vehicle information includes a parking time of the vehicle, and the detecting of the vehicle traffic information of the target lane based on the vehicle information of the vehicle in the target lane includes: In the case where there is a target vehicle in the target lane, determining vehicle traffic information of the target lane based on the target vehicle, the target vehicle being a vehicle whose parking time in the target lane is greater than a second preset threshold; The vehicle traffic information includes a vehicle traffic accident result of the target lane, and the vehicle traffic information of the target lane is determined based on the target vehicle, including: Determine a first vehicle set in the target lane based on the target vehicle, the first vehicle set including a first vehicle in the target lane in a lane line direction from the target vehicle to the target lane, the first vehicle being a vehicle whose parking time is greater than a third preset threshold; When the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, the vehicle traffic accident result is determined to be a vehicle traffic accident occurring in the target lane, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane.

2. The method according to claim 1, It is characterized in that The target video stream includes N frames of images, where N is an integer greater than 1. Tracking the vehicle in the target area based on the target video stream to obtain target information includes: For each frame of the N frames of images, identifying a vehicle in the target area based on the image to obtain position information of the vehicle in the target area; Determine the lane number where the vehicle is located according to the position information of the vehicle in the target area; For each of the M lanes, when it is determined that the lane number of the vehicle for the lane has changed based on at least two consecutive frames of the N frames of images, the lane change rate of the lane is determined based on the number of times the lane number of the vehicle for the lane in the target area has changed.

3. The method according to claim 1, It is characterized in that In the case where it is determined that a vehicle traffic accident exists in the target lane, the vehicle traffic information includes fault information, and the vehicle traffic information of the target lane is determined based on the target vehicle, further comprising: Determine a second vehicle set in the target lane based on the target vehicle, the second vehicle set including a second vehicle in the target lane in a direction opposite to the lane line direction of the target lane from the target vehicle, the second vehicle being a vehicle whose parking time is greater than the third preset threshold; The vehicle information of the target vehicle, the first vehicle set and the second vehicle set are aggregated to obtain the fault information.

4. The method according to claim 1, It is characterized in that The preset condition also includes at least one of the following: The average speed of vehicles in the lane is less than a fifth preset threshold; The vehicle occupancy rate of the lane is greater than a sixth preset threshold.

5. The method according to claim 1, It is characterized in that The step of obtaining a target video stream collected for a target area includes: Acquire a first video stream and a second video stream collected for a target area, wherein the first video stream and the second video stream are video streams collected for both sides of the M lanes respectively; The first video stream and the second video stream are merged to obtain the target video stream.

6. An information detection device, It is characterized in that The device comprises: An acquisition module, used to acquire a target video stream collected for a target area, wherein the target area includes M lanes, where M is an integer greater than 1; a tracking module, configured to track the vehicle in the target area based on the target video stream to obtain target information, wherein the target information includes lane information of each lane in the M lanes and vehicle information of the vehicle in the target area, wherein the lane information includes a lane change rate of the vehicle from one lane to another lane; a detection module, configured to detect vehicle traffic information of the target lane based on vehicle information of vehicles in the target lane when lane information of the target lane in the M lanes satisfies a preset condition, wherein the preset condition includes: a lane change rate of the lane is greater than a first preset threshold; The vehicle information includes the parking time of the vehicle, and the detection module includes: A third determination submodule is used to determine the vehicle traffic information of the target lane based on the target vehicle when there is a target vehicle in the target lane, the target vehicle being a vehicle whose parking time in the target lane is greater than a second preset threshold; The vehicle traffic information includes the vehicle traffic accident result of the target lane, and the third determination submodule includes: A first determining unit is used to determine a first vehicle set in the target lane based on the target vehicle, the first vehicle set including a first vehicle in the target lane in a lane line direction from the target vehicle to the target lane, the first vehicle being a vehicle whose parking time is greater than a third preset threshold; A second determination unit is used to determine that the vehicle traffic accident result is that there is a vehicle traffic accident in the target lane when the distance between the first target vehicle in the first vehicle set and the stop line in the lane line direction of the target lane is greater than a fourth preset threshold, and the first target vehicle is the first vehicle in the first vehicle set that is closest to the stop line in the lane line direction of the target lane.

7. An electronic device, It is characterized in that The electronic device comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the information detection method according to any one of claims 1 to 5 when executed by the processor.

8. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the information detection method according to any one of claims 1 to 5 are implemented.

Images