Image smoothing based on global sparsity decomposition and a variable parameter

Abstract Smoothing images, especially with rich texture, is an important problem in computer vision. Obtaining ideal result difficult due to complexity, irregularity, and anisotropicity of the texture. Besides, some properties are shared by texture structure image. It a hard compromise retain simultaneously remove To create algorithm for image smoothing, we face three problems. For images textures, smoothing effect should be enhanced. We overcome inconsistency results different parts necessary method evaluate effect. apply pre-removal based on global sparse decomposition variable parameter solve first two A parametric surface constructed improved Bessel used determine parameter. Three evaluation measures: edge integrity rate, removal gradient value distribution proposed cope third problem. use alternating direction multipliers complete whole obtain results. Experiments show that our better than existing algorithms both visually quantitatively. also demonstrate method’s ability other applications such as clip-art compression artifact content-aware manipulation.