A vehicle detection method under severe weather conditions for smart cars

Abstract

The invention discloses a vehicle detection method for an intelligent vehicle under severe weather conditions, which solves the problem of difficulty in vehicle detection for an intelligent vehicle in harsh environments such as night, rainy days, foggy days, and snowy days. Vehicle target detection and image capture are carried out by millimeter wave radar and thermal imaging camera respectively; the vehicle target detected by millimeter wave radar is projected to the infrared thermal image captured by thermal imaging camera through coordinate transformation to obtain the vehicle detection target in the infrared thermal image. Approximate location distribution; extract and segment the area of ​​interest of the vehicle; use the DMP target detection algorithm to verify the hypothesis of the area of ​​interest of the vehicle, and perform regression prediction on the detection frame of the vehicle; detect the millimeter-wave radar and thermal imaging camera by calculating the cost matrix The target is fused, and finally Kalman filter is used to track the fused vehicle detection target.

Description

technical field [0001] The invention relates to the field of intelligent vehicle environment perception, in particular to a bad weather vehicle detection method based on the fusion of infrared thermal image and millimeter wave radar information. Background technique [0002] Environmental perception is the basis and premise of intelligent vehicle decision-making and motion control, and the vehicle, as the main participant in traffic, is the key object of environmental perception. The real-time performance and accuracy of vehicle detection in good weather conditions have reached a high level, but there are still many challenges in vehicle detection in bad weather conditions (night, rain, snow, and fog). [0003] In their paper "Robust vehicle detection under variousenvironments to realize road traffic flow surveillance using an infraredthermal camera", Iwasaki Y et al. used infrared thermal images to detect tires, exhaust pipes, windows and other areas with obvious grayscale ...

AI Technical Summary

Problems solved by technology

[0006] The invention proposes a vehicle detection method for smart cars under bad weather conditions, which solves the problem of difficult vehicle detection for smart cars in harsh environments such as night, rainy days, foggy days, and snowy days

Images