Characterization of a polarimetric infrared imager based on the orthogonality breaking technique

Abstract

We report an in-depth experimental characterization and analysis of an infrared active polarimetric imaging system based on the orthogonality breaking polarization-sensing approach. We first recall the principle of this laser scanning polarimetric imaging technique, based on the illumination of a scene by means of a dual-frequency dual-polarization light source. The experimental design is then described, along with measurements on test scenes with known polarimetric properties used to validate/calibrate the imaging system and to characterize its optical properties (sensitivity and resolution). The noise sources that temporally and spatially affect the quality of the orthogonality breaking data are then investigated. Our results show that the raw temporal signals detected at a given location of the scene are perturbed by Gaussian fluctuations, and the spatial information contained in the images acquired through raster scan of the scene are dominated by speckle noise, which is a common characteristic of active polarimetric imaging systems. Finally, the influence of the source temporal coherence on the images is analyzed experimentally, showing that orthogonality breaking acquisitions can still be performed efficiently with a low-coherence source.