The relative potential field as a novel physics-inspired method for image analysis
Xiaodong Zhuang, Nikos E. Mastorakis · 2010
Abstract:- In this paper, the relative potential field is proposed as a novel image transform inspired by the physical electro-static field. A general form of image potential field is presented, based on which the relative potential is defined by introducing the factor of gray-scale difference into the potential field. The properties of the relative potential are investigated experimentally and analyzed, based on which an image segmentation method is proposed by region division and merging in the relative potential field. Experimental results prove the effectiveness of the proposed image segmentation method, and also indicate the promising application of the relative potential in image processing tasks. Key-Words:- Relative potential field, electro-static, image transform, image segmentation 1