20 November 2014 An infared polarization image fusion method based on NSCT and fuzzy C-means clustering segmentation algorithms
Author Affiliations +
Proceedings Volume 9300, International Symposium on Optoelectronic Technology and Application 2014: Infrared Technology and Applications; 93000W (2014) https://doi.org/10.1117/12.2071158
Event: International Symposium on Optoelectronic Technology and Application 2014, 2014, Beijing, China
Abstract
The integration between polarization and intensity images possessing complementary and discriminative information has emerged as a new and important research area. On the basis of the consideration that the resulting image has different clarity and layering requirement for the target and background, we propose a novel fusion method based on non-subsampled Contourlet transform (NSCT) and fuzzy C-means (FCM) segmentation for IR polarization and light intensity images. First, the polarization characteristic image is derived from fusion of the degree of polarization (DOP) and the angle of polarization (AOP) images using local standard variation and abrupt change degree (ACD) combined criteria. Then, the polarization characteristic image is segmented with FCM algorithm. Meanwhile, the two source images are respectively decomposed by NSCT. The regional energy-weighted and similarity measure are adopted to combine the low-frequency sub-band coefficients of the object. The high-frequency sub-band coefficients of the object boundaries are integrated through the maximum selection rule. In addition, the high-frequency sub-band coefficients of internal objects are integrated by utilizing local variation, matching measure and region feature weighting. The weighted average and maximum rules are employed independently in fusing the low-frequency and high-frequency components of the background. Finally, an inverse NSCT operation is accomplished and the final fused image is obtained. The experimental results illustrate that the proposed IR polarization image fusion algorithm can yield an improved performance in terms of the contrast between artificial target and cluttered background and a more detailed representation of the depicted scene.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xuelian Yu, Qian Chen, Guohua Gu, Weixian Qian, Mengxi Xu, "An infared polarization image fusion method based on NSCT and fuzzy C-means clustering segmentation algorithms", Proc. SPIE 9300, International Symposium on Optoelectronic Technology and Application 2014: Infrared Technology and Applications, 93000W (20 November 2014); doi: 10.1117/12.2071158; https://doi.org/10.1117/12.2071158
PROCEEDINGS
9 PAGES


SHARE
Back to Top