An Edge-Preserving Multilevel Method for Deblurring, Denoising, and Segmentation

We present a fast edge-preserving cascadic multilevel image restoration method for reducing blur and noise in contaminated images. The method also can be applied to segmentation. Our multilevel method blends linear algebra and partial differential equation techniques. Regularization is achieved by truncated iteration on each level. Prolongation is carried out by nonlinear edge-preserving and noise-reducing operators. A thresholding updating technique is shown to reduce “ringing” artifacts. Our algorithm combines deblurring, denoising, and segmentation within a single framework.

• Type: Conference
• Year: 2009
• Where: SCALESPACE
• Authors: Serena Morigi, Lothar Reichel, Fiorella Sgallari
• Keywords: Cascadic Multilevel Image, Computer Vision, Multilevel Method Blends, SCALESPACE 2009, Thresholding Updating Technique