Robust ground moving target detection for airborne radar using a novel feature-based machine learning approach

• Published in: Journal of the Franklin Institute Vol. 359; no. 9; pp. 4449 - 4467
• Main Authors: Ahmed, Rafi, Golam Kibria, B. M.
• Format: Journal Article
• Language: English
• Published: Elmsford Elsevier Ltd 01-06-2022; Elsevier Science Ltd
• Subjects: Airborne radar; Clutter; Doppler effect; Euclidean geometry; Euclidean space; Feature extraction; Machine learning; Moving targets; Pixels; Radar detection; Radar targets; Sensors; Target detection
• ISSN: 0016-0032; 1879-2693; 0016-0032
• DOI: 10.1016/j.jfranklin.2022.04.031

A novel ground-moving target detection method is introduced using a distinguishing target, and clutter feature for airborne radar. The clutter proximity feature is extracted based on the Euclidean distance between a signal pixel and the expected clutter ridge in the angle-Doppler domain. Subsequently, target and clutter pixels are classified based on the extracted features for target detection without actually removing clutters or clutter estimation. The proposed technique is especially suitable for effective airborne radar target detection in the unknown ground clutter. The experimental results have validated the effectiveness of the new approach, which enables ground moving target detection in inhomogeneous clutter.