An adaptive image enhancement algorithm

摘要

Image enhancement is a common procedure intended to process an image so that the resulting processed image is more suitable than the original one for a given application. Spatial filtering is a well-known procedure to achieve this goal. Low-pass filtering smooths the image and is often used as a preprocessing step to image analysis, removing small details and reducing noise. This filtering process, however, blurs the image. High-pass filtering sharpens the borders between the different image regions by highlighting edges. This process also highlights noise and spurious details. This study proposes a different method that first detects borders between radiometrically distinct image regions and then applies a weighting function to smooth each region internally. The proposed algorithm was implemented on a SUN Sparcstation for testing purposes. Results of the procedure are presented.