Abstract: Foliage-penetration (FOPEN) radar systems operating in VHF/UHF bands enable detection of man-made targets concealed beneath dense vegetation, yet their performance is significantly constrained by strong, non-Gaussian, and temporally varying foliage clutter. Traditional signal-domain techniques such as MTI/MTD filtering, CFAR variants, adaptive LMS/RLS cancellation, entropy-weighted coherent integration, and blind-source separation methods provide only partial suppression under wind-induced decorrelation, multipath propagation, and low signal-to-clutter ratios. In parallel, image-processing strategies including diffusion filtering, Total Variation and Forward–Backward Diffusion models, anisotropic diffusion, RPCA/TV-RPCA decomposition, and background-subtraction families enhance clutter rejection in Range-Time-Intensity (RTI), Range–Doppler (RD), and spectrogram formats by exploiting spatial–temporal structure. Recent advances in machine learning and deep networks such as CNNs, autoencoders, hybrid feature-fusion architectures, and learned Doppler-feature representations further improve target localization and classification by modeling nonlinear clutter characteristics directly from radar imagery. Also, Synthetic Aperture Radar (SAR)-based FOPEN imaging provides a complementary pathway for clutter suppression. Techniques including coherent and incoherent change detection, spectral intersection filtering, circular-aperture angular correlation (ACF/FCF), physical-optics modeling, deterministic-aided STAP, and multi-channel interferometric imaging significantly improve background suppression and moving-target detection in dense forests. This paper presents a comprehensive review of signal-domain, image-domain, machine-learning, and SAR-based clutter-cancellation techniques for FOPEN radar. The review focuses on a number of key aspects, including theoretical principles, algorithmic implementation, strengths, limitations, and realistic performance in a wide range of forest environments. Such unification of classical radar processing with modern SAR imaging and deep learning presents a comprehensive reference to guide future research toward high-accuracy detection, classification, and tracking of hidden targets beneath dense vegetation.
Keywords: FOPEN, STAP, ACF/FCF, SAR, FBD, Image Processing, Clutter cancellation.
Title: A Review on Image Processing Techniques on FOPEN Radar Clutter Cancellation
Author: Gouresh Desai, Prasad Saratkar, Sohan Salunkhe, Sakshi Fulsundar, A. A. Bazil Raj
International Journal of Engineering Research and Reviews
ISSN 2348-697X (Online)
Vol. 13, Issue 4, October 2025 - December 2025
Page No: 11-37
Research Publish Journals
Website: www.researchpublish.com
Published Date: 10-December-2025
