A novel small dim infrared target detecting method based on grayscale merging and connected components analysis

Target detecting algorithm in infrared image is drawing extensive attention both at home and abroad, expecially when the infrared images own complex backgrounds and low resolution. How to make sure of the accuracy of infrared dim-small target detecting is the key point during the processing procedure. In this paper, a novel region of interest detecting algorithm is proposed for aerial small dim infrared target with clutter background. Different from traditional region of interest extraction method, we extract region of interest based on the grayscale merging algorithm and connected components analysising approach. First, the infrared image is reconstructed by separating the grayscale into several interval sections, then binarization is used to segment the processed image according to estimate the threshold value using minimum error probability; finally, analysis and select the connected components, to sift the accurate connected domain which contain the target. The validity of the proposed approach is demonstrated by experiment of infrared target region extraction.