11 November 2014 Infrared image segmentation method based on spatial coherence histogram and maximum entropy
Author Affiliations +
Abstract
In order to segment the target well and suppress background noises effectively, an infrared image segmentation method based on spatial coherence histogram and maximum entropy is proposed. First, spatial coherence histogram is presented by weighting the importance of the different position of these pixels with the same gray-level, which is obtained by computing their local density. Then, after enhancing the image by spatial coherence histogram, 1D maximum entropy method is used to segment the image. The novel method can not only get better segmentation results, but also have a faster computation time than traditional 2D histogram-based segmentation methods.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Songtao Liu, Songtao Liu, Tongsheng Shen, Tongsheng Shen, Yao Dai, Yao Dai, } "Infrared image segmentation method based on spatial coherence histogram and maximum entropy", Proc. SPIE 9275, Infrared, Millimeter-Wave, and Terahertz Technologies III, 92750U (11 November 2014); doi: 10.1117/12.2071462; https://doi.org/10.1117/12.2071462
PROCEEDINGS
8 PAGES


SHARE
Back to Top