Landmine detection based on morphological component analysis

Ultra-wideband synthetic aperture radar (SAR) is a promising technology for landmine detection. According to the geometric shape diversity among landmines, textures and backgrounds, a novel landmine detection method based on morphological component analysis (MCA) is proposed. The key of this method is the cyclic iteration of images decomposition, coefficients updating, images reconstruction and images updating. The experimental results indicate that the proposed method is effective for landmine detection.