Foreground segmentation using motion vector for camouflaged surveillance scenario

Detection of moving objects is the first relevant step of information extraction in many computer vision applications. It gains focus of attention in applications such as abnormal event detection, human gait characterization, congestion analysis, person identification and fall detection for elderly people. Foreground segmentation fails due to various reasons such as presence of camouflage, shadows, slow moving objects, illumination changes etc. Detecting moving objects from similarly colored background (known as camouflage problem), has been a long-standing open question in this field. Thus, camouflage is one of the major issues where foreground and background gets occluded or merged when they are of similar color. In case of terrorist attacks, terrorist disguise themselves, by dressing similar to the environment so that they can hide themselves. Foreground segmentation by eliminating camouflage in such cases play a vital role which is one of the major issues being faced nowadays. In order to overcome this issue, the proposed work extracts motion information using optical flow algorithm to detect the moving object and Motion Assisted Matrix Restoration (MAMR) for foreground segmentation in fully camouflaged surveillance scenario. Experimental results have been shown emphasizing that the proposed methodology outperforms other state-of-the-art algorithms.