MOON AND EARTH IMAGE PROCESSING USING ASYMMETRIC 2-DIMENSIONAL FUNCTIONS ON IMAGE GRADIENT Conference Paper uri icon

abstract

  • Asymmetric Gaussian-type data distribution is found in the image processing at the edge of an observed illuminated body (Moon, Earth, planet) using the image gradient. The reason is, one side of this Gaussian behavior is associated with the dark background (pretty uniform) while the other side is associated with the illuminated part, which is perturbed by different reflecting areas: craters for Moon and high clouds for Earth. This variation is here implemented by modeling the standard deviation as a sigmoid function. Nonlinear least squares estimate is then applied for OpNav problem using synthetic Moon and Earth images.

author list (cited authors)

  • Mortari, D., & Borissov, S.

publication date

  • January 2016