Feature level fusion of night vision images based on K-means clustering algorithm

Abstract

A region based visual and thermal image fusion technique based on k-means clustering algorithm is presented in this paper. This novel region fusion method segments regions of interest from thermal image using k-means clustering algorithm. Later on, these regions of interests are fused with visible image in DWFT domain. A prominent feature of our proposed technique is its near-real-time computation. Objective comparison of the scheme proposed in this paper has been done with other well known techniques. Experimental results and conclusion outlined in this paper will explain how well the proposed algorithm performs. © 2007 Springer.