In addition to spectral information acquired by traditional multi/hyperspectral systems, passive electro optical and infrared (EO/IR) polarimetric sensors also measure the polarization response of different materials in the scene. Such an imaging modality can be useful in improving surface characterization; however, the characteristics of polarimetric systems have not been completely explored by the remote sensing community. Therefore, the main objective of this research was to advance our knowledge in polarimetric remote sensing by investigating the impact of polarization phenomenology on material discriminability. The first part of this research focuses on system validation, where the major goal was to assess the fidelity of the polarimetric images simulated using the Digital Imaging and Remote Sensing Image Generation (DIRSIG) model. A theoretical framework, based on polarization vision models used for animal vision studies and industrial defect detection applications, was developed within which the major components of the polarimetric image chain were validated. In the second part of this research, a polarization physics based approach for improved material discriminability was proposed. This approach utilizes the angular variation in the polarization response to infer the physical characteristics of the observed surface by imaging the scene in three different view directions. The usefulness of the proposed approach in improving detection performance in the absence of apriori knowledge about the target geometry was demonstrated. Sensitivity analysis of the proposed system for different scene related parameters was performed to identify the imaging conditions under which the material discriminability is maximized. Furthermore, the detection performance of the proposed polarimetric system was compared to that of the hyperspectral system to identify scenarios where polarization information can be very useful in improving the target contrast.
Library of Congress Subject Headings
Remote sensing--Data processing; Polarimetric remote sensing--Evaluation; Image processing--Digital techniques; Polarization (Light)
Department, Program, or Center
Chester F. Carlson Center for Imaging Science (COS)
Devaraj, Chabitha, "Polarimetric remote sensing system analysis: Digital Imaging and Remote Sensing Image Generation (DIRSIG) model validation and impact of polarization phenomenology on material discriminability" (2010). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus