A method for automatic updating of radar clutter maps for micro-target detection on pavement

Abstract

The invention provides a method for automatically updating the radar clutter map for micro-target detection on the pavement, which can update the clutter map in real time, improve the detection probability of micro-targets whose echo power is on the edge of the detection threshold, and avoid missed and false alarms. After the road surface micro target detection radar is turned on, it performs a pre-scan to generate two clutter images and stores them, one of which is smoothed and used for CFAR detection during the detection scan, and the other image is directly stored without any processing . In each detection scan, the smoothed clutter map is called. After the scan is completed, the unsmoothed clutter map is updated according to the automatic update judgment conditions of the clutter map, and the updated clutter map is smoothed and replaced by the previous one. The smoothed clutter map is used for the next detection scan, and so on. Because of the existence of automatic update judgment conditions, it is avoided to update the target into the clutter map, resulting in missed alarms; in rainy and snowy weather, the clutter map can also be updated in real time to avoid false alarms caused by accumulated water or snow.

Description

technical field [0001] The invention belongs to the technical field of tiny target detection, and in particular relates to a method for automatically updating radar clutter images for detection of tiny targets (below 5 cm) on the road surface. Background technique [0002] At present, millimeter-wave micro-target detection radars mostly use the CFAR detection algorithm based on clutter maps. After the radar is turned on, it first conducts a pre-scan to generate a clutter map, and subsequent micro-target detection radars do not change after the pre-scanning generates a clutter map. The clutter map will be regenerated until the next boot. [0003] First, when the radar operating environment has a large temperature difference between day and night, the reflection characteristics of the road surface will change slightly, resulting in the clutter map used not completely matching the current actual scene. Although this will not seriously affect the detection results, it will For ...

AI Technical Summary

Problems solved by technology

[0003] First, when the radar operating environment has a large temperature difference between day and night, the reflection characteristics of the road surface will change slightly, resulting in the clutter map used not completely matching the current actual scene. Although this will not seriously affect the detection results, it will For some targets whose echo power is on the edge of the detection threshold, it is easy to cause missed alarms; second, when encountering rainy and snowy weather, the echo characteristics of the road surface will change greatly, and the clutter map at this time is completely consistent with the actual scene. Inconsistency, resulting in errors in the CFAR detection of the clutter map, and the occurrence of missed and false alarms at the same time

Images