Copula-Based Robust Landmine Detection in Muti-View Forward-Looking GPR Imagery

We propose a scheme for detecting landmines using forward-looking ground-penetrating radar. The detector is applied to tomographic radar images obtained from multiple viewpoints of the investigation area and is based on a robust version of the likelihood-ratio test. The statistical dependence between multi-view images is incorporated via a copula-based function. The test is designed to maximize the worst-case performance over all feasible target and clutter distribution pairs, thereby eliminating the need for a strong assumption about the clutter distribution. Using numerical radar data of shallow buried targets, we demonstrate the superior performance of the proposed detector over existing approaches.