POLARIZATION PHASE-BASED METHOD FOR MATERIAL CLASSIFICATION IN COMPUTER VISION

A robust and accurate polarization phase-based technique for material classification is presented. The novelty of this technique is three-fo ld in (i) its theoretical development, (ii) application, and, (iii) ex perimental implementation. The concept of phase of polarization of a l ight wave is introduced to computer vision for discrimination between materials according to their intrinsic electrical conductivity, such a s distinguishing conducting metals, and poorly conducting dielectrics. Previous work has used intensity, color and polarization component ra tios. This new method is based on the physical principle that metals r etard orthogonal components of light upon reflection while dielectrics do not. This method has significant complementary advantages with res pect to existing techniques, is computationally efficient, and can be easily implemented with existing imaging technology. Experiments for r eal circuit board inspection, nonconductive and conductive glass, and, outdoor object recognition have been performed to demonstrate its acc uracy and potential capabilities.