Polarization Optical Image Processing Used in the Target Detection and Identification

Optical image processing is an important part of photoelectric detection, but the frequently-used method is limited in the condition of complex background, so this paper introduces a thought to improve this problem that the optical image can be made polarized in order to get pictures of polarization intensities and pictures of degree of polarization. At first place, it makes an analysis of polarization of light, and points out there are obvious differences of reflected polarization characteristics between target and background which can be used to distinguish different things. Then it makes a simulation of reflection from the surface of the green paint and aluminum, and shows that the green paint has bigger degree of polarization than aluminum while smaller index of reflectance. At last, it does a series of outdoor experiments and verifies that the polarization pictures show distinct discrepancies between different materials eventually. All these works state that the polarization of an optical image is a potential way to make target detection and identification easier with higher accuracy.