An UM-based silhouette-crease edge enhancement for noisy images

Abstract

This paper presents an improved unsharp masking (UM) technique that enhances the quality and suppresses noises for the images acquired from a noisy environment such as taken during night time. Our approach employs noise smoothing and the idea that important edges should be enhanced more than minor edges. Edges are classified as silhouette and crease edges (major and minor edges) according to their lengths. An adaptive weighting method is used to enhance the edges. In this way, the major edges (silhouette) are sharpened more than minor edges (crease). The proposed method is examined on night images as well as noisy images. It is also compared to existing UM-based methods with satisfying results.