A traffic flow detection method and related device

Abstract

The embodiment of the present application provides a traffic flow detection method and related devices. By matching the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target, the trajectory points of the same detection target at the same time can be fused to determine the trajectory information of each detection target. According to the direction vector of each trajectory point in the trajectory information of each detection target, and the calibrated lane information, the trajectory information of each detection target is corrected, and the trajectory points with large errors can be removed to avoid affecting the traffic flow detection results. On this basis, the traffic flow detection results can be obtained by comparing the corrected trajectory information with the detection line coordinate information. The method provided by the embodiment of the present application reduces the instability and error caused by the target detection process and improves the reliability of the traffic flow detection results.

Claims

1. A traffic flow detection method, characterized in that: include: Based on the coordinate information of multiple trajectory points of at least one detection target, matching the multiple trajectory points to determine the trajectory information of each of the detection targets; According to the direction vector of each trajectory point in the trajectory information of each detected target and the calibrated lane information, the trajectory information of each detected target is corrected to obtain the corrected trajectory information corresponding to each detected target; The corrected trajectory information is compared with the detection line coordinate information to obtain a traffic flow detection result.

2. The method according to claim 1, characterized in that The calibrated lane information includes lane position information and lane direction, and the trajectory information of each of the detected targets is corrected according to the direction vector of each trajectory point in the trajectory information of each of the detected targets and the calibrated lane information to obtain the corrected trajectory information corresponding to each of the detected targets, including: Determining the lane to which the trajectory information of each detected target belongs according to the overlap between the trajectory information of each detected target and the lane position information; According to the direction vector of each trajectory point in the trajectory information of each detected target and the lane direction of the lane to which it belongs, the trajectory information of each detected target is corrected to obtain the corrected trajectory information corresponding to each detected target.

3. The method according to claim 2, characterized in that The step of correcting the trajectory information of each of the detected targets according to the direction vector of each trajectory point in the trajectory information of each of the detected targets and the lane direction of the lane to which the detected target belongs, and obtaining the corrected trajectory information corresponding to each of the detected targets, includes: Determine the angle between the direction vector of each trajectory point and the lane direction according to the direction vector of each trajectory point in the trajectory information of each detected target and the lane direction of the lane to which the target belongs; Determine the confidence of each track point according to the angle, the change value of the single-frame speed of the detected target relative to the previous and next frames, the change value of the single-frame heading angle of the detected target relative to the previous and next frames, and the distance of the single-frame position of the detected target relative to the track information; The trajectory points whose confidence level is lower than the confidence level threshold in the trajectory information of each of the detected targets are deleted to obtain the corrected trajectory information corresponding to each of the detected targets.

4. The method according to claim 1, characterized in that: The step of matching the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target to determine the trajectory information of each detection target includes: Based on the coordinate information of multiple trajectory points of at least one detection target, matching the multiple trajectory points to obtain a matching result of the trajectory points; When the matching result of the trajectory point is that there is a sub-trajectory point that matches the trajectory point, the trajectory point and the sub-trajectory point are fused to determine the trajectory information of the detection target; when the matching result of the trajectory point is that there is no sub-trajectory point that matches the trajectory point, the trajectory information of the detection target is determined based on the trajectory point.

5. The method according to claim 1, characterized in that The step of comparing the corrected trajectory information with the detection line coordinate information to obtain a traffic flow detection result includes: Comparing the corrected trajectory information with the detection line coordinate information to obtain a comparison result; When the comparison result is that the corrected trajectory information intersects with the detection line coordinate information, the count value of the traffic flow is increased to obtain a traffic flow detection result.

6. The method according to claim 1, characterized in that The step of matching the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target to determine the trajectory information of each detection target includes: Based on the coordinate information of multiple trajectory points of at least one detection target, matching the multiple trajectory points to determine sub-trajectory information of each of the detection targets; The sub-trajectory information of each of the detected targets is filtered to determine the trajectory information of each of the detected targets.

7. The method according to any one of claims 1 to 6, characterized in that: Before matching the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target to determine the trajectory information of each detection target, the method further includes: Obtaining pixel coordinates of multiple trajectory points of at least one detection target; Based on a mapping relationship between world coordinates and pixel coordinates, coordinate conversion is performed on the pixel coordinates of the multiple track points of the at least one detection target to obtain the world coordinates of the multiple track points of the at least one detection target; The step of matching the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target to determine the trajectory information of each detection target includes: Based on the world coordinates of multiple track points of at least one detection target, the multiple track points are matched to determine the track information of each of the detection targets.

8. A traffic flow detection device, characterized in that: include: A matching unit, a correction unit, and a comparison unit; The matching unit is used to match the multiple trajectory points based on the coordinate information of the multiple trajectory points of at least one detection target to determine the trajectory information of each of the detection targets; The correction unit is used to correct the trajectory information of each of the detected targets according to the direction vector of each trajectory point in the trajectory information of each of the detected targets and the calibrated lane information, so as to obtain the corrected trajectory information corresponding to each of the detected targets; The comparison unit is used to compare the corrected trajectory information with the detection line coordinate information to obtain a traffic flow detection result.

9. A computer device, characterized in that: The computer device comprises: a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the steps of a traffic flow detection method as described in any one of claims 1-7 according to the instructions in the program code.

10. A computer-readable storage medium, 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 a traffic flow detection method according to any one of claims 1 to 7 are implemented.

Images