Object Detection Method Based on Variable-Scale Median Ratio

Abstract

The invention discloses a target detection method based on variable-scale median ratio, which includes the following specific steps: acquiring radar echo data; calculating non-coherent pulse accumulation; setting small-scale parameter sets and large-scale parameter sets; calculating median energy features ; calculate the detection statistics; online perception clutter parameters; look up the table to obtain the threshold; compare the size of the test statistic and the judgment threshold to determine whether the target exists; Wave power extraction and target energy extraction in a small range, so as to solve the problem that the reference unit is the target itself at high resolution. Because the present invention uses the median ratio under variable scales, the detection statistics have the ability to resist abnormal values, and can effectively solve the problem of strong fluctuations in the target amplitude.

Description

Technical field:

[0001] The invention belongs to the technical field of radar signal processing, and in particular relates to a detection method for a sea surface expansion target, which can be used for high-resolution sea search radar detection targets. Background technique:

[0002] The main mission of marine radar is to detect sea surface targets, such as ships, cargo ships, passenger ships, etc. At present, for fast detection, radars operating in fast scan mode use incoherent Constant False Alarm Rate (CFAR) detectors, typically Cell Average CFAR (CA-CFAR) detectors . In addition, a series of adaptive optimal or near-optimal coherent detectors with matching clutter models are developed, including Adaptive Matched Filter (AMF) in Gaussian clutter environment, Adaptive Matched Filter (AMF) in heavy tail clutter environment Adaptive Normalized Matched Filter (ANMF), Optimum K-distributed Detector (OKD) under K distribution, Generalized Likelihood Ratio Linear Threshold De...

Images