Target detection method and system based on camera and radar fusion

Abstract

The invention discloses a target detection method and system based on camera and radar fusion. The detection method comprises the steps that firstly, radar data detected by a radar and an image acquired by a camera synchronously are acquired; secondly, the radar target in the radar data is subjected to information correlation filtering and Kalman filtering, and the effective radar target is screened out; then, the camera target in the image is detected through a deep learning method and the target information of the camera target is acquired; and finally, the effective radar target and the camera target are fused by a cross-small ratio method, and the fused target result is screened and output. By adopting the detection method, the target leakage detection rate and the false detection rateare effectively reduced, the fusion target result is screened, and the accuracy of the fusion target result is ensured.

Description

technical field [0001] The invention belongs to the technical field of target detection, and in particular relates to a target detection method and system based on camera and radar fusion. Background technique [0002] Intruders at intersections such as trams will cause great harm to the safety of track operation. At present, the method or system of obstacle detection is mainly based on the vehicle, used to detect obstacles in front, and the detection is based on the motion coordinate system. There are methods that only use the region of interest (ROI) superimposition to achieve detection, and there is a high false detection rate. In the field of rail transit detection, how to realize obstacle detection based on the static coordinate system and make full use of the two-dimensional information fusion of camera and radar has become an urgent technical problem to be solved. Contents of the invention [0003] In view of the above problems, the present invention provides a tar...

AI Technical Summary

Problems solved by technology

[0002] Intruders at intersections such as trams will cause great harm to the safety of track operation. At present, the method or system of obstacle detection is mainly based on the vehicle, used to detect obstacles in front, and the detection is based on the motion coordinate system. There is a method of only using the region of interest (ROI) overlay to achieve detection, and there is a high false detection rate

Images