Polarization Imagery

The polarization of reflected radiation can provide useful information which could be used in remote sensing applications to help distinguish different natural surfaces with similar spectral signatures. Yet the use of polarization has been almost completely neglected in remote sensing applications partially because of the lack of understanding of the information contained in the polarization field. In this paper, examples of the polarization of natural scenes will be presented and the information contained in the polarization field will be discussed. The imagery presented was collected by taking sets of four photographs of various natural scenes using a polarizing filter to detect the polarization field. The polarization field was analyzed by digitizing the photographs and processing the results at the Image Processing Laboratory on the Berkeley campus of the University of California. Introduction A beam of incoherent radiation reflected from a natural surface can be completely described at a given wavelength by the Stokes vector I Q U V where I, Q, U, and V are physical observables describing independent features of the electric vector. The quantity I represents the intensity of the radiation and the quantities Q, U, and V are associated with the polarization of the radiation. Traditionally, in remote sensing experiments only I is detected while Q, U, and V are neglected. However, there is useful physically independent information contained in the polarization parameters, as we will show presently. Detection of Polarization Parameters A simple way to detect the Stokes parameters Q and U for a natural scene is to take photographs of the scene with a standard polarizing filter in front of a camera. The intensity seen by the camera is then I'(e) = (I + Q cos2e + U sin2e) /2, (1) where I, Q, and U are the Stokes parameters before passing through the polarizer, and e is the angle of the transmission axis of the polarizer with respect to the horizontal. We took sets of four photographs of natural scenes with e set to 0, 45, 90, and 135 degrees. The intensity recorded in each of the four photographs is, from Equation (1), I'(0) _ (I + Q) /2, (2) I'(45) _ (I + U) /2, (3) I'(90) _ (I Q) /2, (4) and I'(135) _ (I U) /2. (5) From Equations (2) -(5) it follows that I = (I'(0) + I'(45) + I'(90) + I'(135))/2, (6) Q = I'(0) I'(90), (7) and U = I'(45) I'(135). (8) When discussing polarization, it is easier to think in terms of the magnitude, P, and the phase angle, 60, of the polarization. Equation (1) can be rewritten in terms of P and eo as I'(e) = I (1 + P cos2(6-60)/2. (9) 164 /SP /E Vol. 112 Optical Polarimetry (1977) I I t l t. Land, Air and Water Resources i r it f lif i , i 95