Object Detection Method for Ground Surveillance Radar

Abstract

The invention discloses a target detection method for a ground surveillance radar, and the method comprises the following steps of performing linear interpolation processing on spectral data information on a unit to be detected corresponding to a low-speed channel; constructing picture units on a unit to be detected corresponding to a non-low-speed channel, respectively constructing quick pictureunits and slow picture units in different division ways for different distance ranges, respectively setting thresholds of corresponding number of points for different picture units, and when the amplitude value threshold of a single picture unit is cumulatively calculated in multiple frames, not only considering the amplitude value threshold of the single frame unit in the previous frame, but alsocomprehensively considering the number of points occupied by the spectrum data of the single picture unit in a certain frame number range, so as to calculate and adjust the amplitude value thresholdof the single picture unit. The method provided by the invention can effectively reduce the influence of fixed strong clutter on the zero frequency channel and the low-speed channel near the zero channel on the low-speed target, improve the discovery probability of the low-speed target, effectively suppress the clutter and improve the detection capability of the target.

Description

technical field [0001] The invention relates to the technical field of radar target detection, in particular to a target detection method for ground surveillance radar. Background technique [0002] Target detection refers to detecting the spectral data information belonging to the target from the spectral data information received by the radar, and filtering the background clutter and ground clutter. And in the subsequent process of point track processing, accurate target detection can improve the accuracy of point track aggregation and track tracking quality. Therefore, choosing the optimal object detection method is crucial to accurately estimating the object parameters. [0003] The traditional target detection method is: direct constant false alarm rate processing, without considering the fixed strong clutter on the zero channel and the nearby channels of the zero channel, that is, the influence of strong ground clutter on the low-speed target, reducing the zero channe...

AI Technical Summary

Problems solved by technology

[0005] However, this patent cannot be used to detect targets on the ground. The targets in the ground scene are complex, including the detection of people, cars, and aircraft models, and the frame-to-frame changes in the ground scene are large. For example, some frames are cars, and the next few frames It is a person. Since the car is a rigid body, the detection range of the car is larger than that of the person. Therefore, using the accumulated amplitude value for target detection will lead to false detection of the human target.

Images