Edge- and Contour Detection for Correction of Low-Pass Filtered Images

摘要

Methods of extracting data from Remote Sensing images for correction of spatial low-pass filtered image are studied. Basically this is an edge detection problem, since edges in low-pass filtered images are degraded. Commonly used edge detectors are not feasible for correction purpose as will be shown. A new method for contour- and average grey level extraction of monochrome images will be introduced and the extracted data is used for correction of the low-pass filtered images. Parts of the extraction algorithm are developed for on-line implementation; therefore they can be implemented in on-board systems of satellites for data reduction purpose.