Share Email Print

Proceedings Paper

Camouflaged target detection based on polarized spectral features
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The polarized hyperspectral images (PHSI) include polarization, spectral, spatial and radiant features, which provide more information about objects and scenes than traditional intensity or spectrum ones. And polarization can suppress the background and highlight the object, leading to the high potential to improve camouflaged target detection. So polarized hyperspectral imaging technique has aroused extensive concern in the last few years. Nowadays, the detection methods are still not very mature, most of which are rooted in the detection of hyperspectral image. And before using these algorithms, Stokes vector is used to process the original four-dimensional polarized hyperspectral data firstly. However, when the data is large and complex, the amount of calculation and error will increase. In this paper, tensor is applied to reconstruct the original four-dimensional data into new three-dimensional data, then, the constraint energy minimization (CEM) is used to process the new data, which adds the polarization information to construct the polarized spectral filter operator and takes full advantages of spectral and polarized information. This way deals with the original data without extracting the Stokes vector, so as to reduce the computation and error greatly. The experimental results also show that the proposed method in this paper is more suitable for the target detection of the PHSI.

Paper Details

Date Published: 4 May 2016
PDF: 6 pages
Proc. SPIE 9853, Polarization: Measurement, Analysis, and Remote Sensing XII, 98530Y (4 May 2016); doi: 10.1117/12.2222160
Show Author Affiliations
Jian Tan, Harbin Institute of Technology (China)
Junping Zhang, Harbin Institute of Technology (China)
Bin Zou, Harbin Institute of Technology (China)

Published in SPIE Proceedings Vol. 9853:
Polarization: Measurement, Analysis, and Remote Sensing XII
David B. Chenault; Dennis H. Goldstein, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?