Research on polarization imaging information parsing method

Hongwu Yuan; Pucheng Zhou; Xiaolong Wang

doi:10.1117/12.2255713

Translator Disclaimer

7 November 2016 Research on polarization imaging information parsing method

Hongwu Yuan, Pucheng Zhou, Xiaolong Wang

Author Affiliations +

Proceedings Volume 10141, Selected Papers of the Chinese Society for Optical Engineering Conferences held July 2016; 101411E (2016) https://doi.org/10.1117/12.2255713
Event: Selected Proceedings of the Chinese Society for Optical Engineering Conferences held July 2016, 2016, Changchun, China

Abstract

Polarization information parsing plays an important role in polarization imaging detection. This paper focus on the polarization information parsing method: Firstly, the general process of polarization information parsing is given, mainly including polarization image preprocessing, multiple polarization parameters calculation, polarization image fusion and polarization image tracking, etc.; And then the research achievements of the polarization information parsing method are presented, in terms of polarization image preprocessing, the polarization image registration method based on the maximum mutual information is designed. The experiment shows that this method can improve the precision of registration and be satisfied the need of polarization information parsing; In terms of multiple polarization parameters calculation, based on the omnidirectional polarization inversion model is built, a variety of polarization parameter images are obtained and the precision of inversion is to be improve obviously; In terms of polarization image fusion , using fuzzy integral and sparse representation, the multiple polarization parameters adaptive optimal fusion method is given, and the targets detection in complex scene is completed by using the clustering image segmentation algorithm based on fractal characters; In polarization image tracking, the average displacement polarization image characteristics of auxiliary particle filtering fusion tracking algorithm is put forward to achieve the smooth tracking of moving targets. Finally, the polarization information parsing method is applied to the polarization imaging detection of typical targets such as the camouflage target, the fog and latent fingerprints.

Citation Download Citation

Hongwu Yuan, Pucheng Zhou, and Xiaolong Wang "Research on polarization imaging information parsing method", Proc. SPIE 10141, Selected Papers of the Chinese Society for Optical Engineering Conferences held July 2016, 101411E (7 November 2016); https://doi.org/10.1117/12.2255713

ACCESS THE FULL ARTICLE