Ultra-Low Velocity MTD Based on Ku-band Circular VideoSAR System

Video synthetic aperture radar (VideoSAR) is extensively utilized profiting from the characterization of ground object via sequential imageries. In this paper, we investigate the machine learning-based moving target detection (MTD) with defocusing component in 360-degree Ku-band VideoSAR proposal, notably for ultra-low velocity circumstances. The low-rank plus sparse decomposition scheme, i.e., three-term decomposition with proximal exchange-based alternating directions method of multipliers, is additionally exploited for acquiring the defocusing component in single-channel single-pass circumstance. Experiments are carried on real airborne Ku-band data to confirm the effectiveness.