Super-resolution target mapping from soft classified remotely sensed imagery

摘要: A simple, efficient algorithm is presented for sub-pixel target mapping from remotely-sensed images. Following an initial random allocation of “soft” pixel proportions to “hard” binary classes, the works in a series iterations, each which contains three stages. For pixel, all locations, distance-weighted function neighboring sub-pixels computed. Then, representing class with minimum value function, and background maximum are found. Third, these two swapped if swap results increase spatial correlation between sub-pixels. The new predicted accurately when applied simple simulated real It represents accessible tool that can be coded readily by remote sensing investigators.