Abstract
This paper presents a novel approach for clutter removal in airborne radars using a genetic algorithm and mathematical morphology. The clutter returns are detected when constant alarm rate processing is applied on range-Doppler images. In the proposed method, mathematical morphological operations are performed on range-Doppler images to obtain clutter images. The clutter image is then applied as a mask to remove false detections due to clutter. Also, the targets embedded in clutter are detected using gray-scale morphological operations. The morphological filter and the sequence of operations are designed by a genetic algorithm. The advantage of the proposed method is that it does not require the computation of statistical measures from clutter data and filters are optimally designed using a genetic algorithm. The proposed method has shown an increase in clutter leak reduction when compared to that of a deep morphological network.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Data sharing is not applicable to this article as no new data were created or analyzed in this study.