Detection method and system for target vehicle

Abstract

The invention provides a detection method and system for a target vehicle, and relates to the field of intelligent driving of vehicles. The method comprises the following steps of: extracting a plurality of feature points according to an image of a target vehicle to obtain a first feature point set; finding out a target nearest point closest to a self-vehicle from the first feature point set; acquiring the contour of the target vehicle according to the first feature point set; then fusing a target feature point with a radar feature point, which is acquired by a self-vehicle radar and is closest to the self-vehicle, of the target vehicle to obtain a fused target feature point; and further correcting the contour of the target vehicle by using the fused target feature point to obtain a finalcontour of the target vehicle so as to track the target vehicle according to the final contour. According to the method, the feature point set is extracted through the image, so that the target feature point can be accurately found out and fused with the target feature point collected by the radar, so that the recognition precision of the target feature point is further improved.

Description

technical field [0001] The invention relates to the field of vehicle intelligent driving, in particular to a detection method and system for a target vehicle. Background technique [0002] At present, there are many methods for vehicle detection. According to the sensor composition of the perception system, it can be divided into: single sensor, multi-sensor fusion. Among them, the cost of a single sensor is low and the implementation is simple, but the recognition rate is not high, the robustness is not high, and there are many scene limitations, such as poor night camera performance and poor radar detection of stationary targets. For multi-sensor fusion, although the cost is high and the fusion technology is complicated, the perception performance, stability and scene adaptability are better. [0003] According to the type of sensor, it can be divided into: camera, millimeter-wave radar, and lidar. The recognition based on the camera image has strong classification abili...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the first aspect of the present invention is to provide a detection method for a target vehicle, which solves the problem of low recognition of various sensors in the prior art when detecting a target vehicle

Images