Level Sets Guided by SoDEF-Fitting Energy for River Channel Detection in SAR Images

To achieve river channel detection in SAR (synthetic aperture radar) images, we developed a level-set-based model (LSBM) guided by designed data-fitting energy which is called the SoDEF (sum of dual exponential functions)-fitting energy. Firstly, function computing sum functions to substitute for quadratic function, and used it construct Secondly, adaptive area-fitting centers (AFCs) were computed based on two kinds grayscale characteristics, are more accurate stable. Thirdly, Dirac gradient descent flow was displaced an edge indicator help evolving level sets stop at target edges. Moreover, some regularized terms incorporated into objective guarantee model’s stability. The experiments conducted with real images indicated that superior related state-of-the-art methods its accuracy efficiency.