An adaptive background estimation for real-time object localization on a color-coded environment

Object localization is a powerful technique to analyze certain features on particular objects for either general or specific purpose(s). To localize object within image frames, the background subtraction scheme is generally implemented as an unequivocal technique that pre-processes die corresponding input frame. Nevertheless, despite the prior developments and applications of various techniques for estimating backgrounds, eminent challenges remain in achieving real-time applications and robust detection. In this paper, an adaptive background estimation technique models specific color-coded environment to localize objects through color-based pixel-wise subtraction is proposed. Experimental evaluations indicate the grandeur of the proposed method compared with prior methods.