TY - GEN
T1 - An adaptive norm algorithm for image restoration
AU - BERTACCINI, Daniele
AU - CHAN, Raymond H.
AU - MORIGI, Serena
AU - SGALLARI, Fiorella
PY - 2012
Y1 - 2012
N2 - We propose an adaptive norm strategy designed for the re-storation of images contaminated by blur and noise. Standard Tikhonov regularization can give good results with Gaussian noise and smooth images, but can over-smooth the output. On the other hand, L 1-TV (Total Variation) regularization has superior performance with some non-Gaussian noise and controls both the size of jumps and the geometry of the object boundaries in the image but smooth parts of the recovered images can be blocky. According to a coherence map of the image which is obtained by a threshold structure tensor, and can detect smooth regions and edges in the image, we apply L 2-norm or L 1-norm regularization to different parts of the image. The solution of the resulting minimization problem is obtained by a fast algorithm based on the half-quadratic technique recently proposed in [2] for L 1-TV regularization. Some numerical results show the effectiveness of our adaptive norm image restoration strategy.
AB - We propose an adaptive norm strategy designed for the re-storation of images contaminated by blur and noise. Standard Tikhonov regularization can give good results with Gaussian noise and smooth images, but can over-smooth the output. On the other hand, L 1-TV (Total Variation) regularization has superior performance with some non-Gaussian noise and controls both the size of jumps and the geometry of the object boundaries in the image but smooth parts of the recovered images can be blocky. According to a coherence map of the image which is obtained by a threshold structure tensor, and can detect smooth regions and edges in the image, we apply L 2-norm or L 1-norm regularization to different parts of the image. The solution of the resulting minimization problem is obtained by a fast algorithm based on the half-quadratic technique recently proposed in [2] for L 1-TV regularization. Some numerical results show the effectiveness of our adaptive norm image restoration strategy.
UR - http://www.scopus.com/inward/record.url?scp=84855698574&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-24785-9_17
DO - 10.1007/978-3-642-24785-9_17
M3 - Conference paper (refereed)
AN - SCOPUS:84855698574
SN - 9783642247842
T3 - Lecture Notes in Computer Science
SP - 194
EP - 205
BT - Scale Space and Variational Methods in Computer Vision: Third International Conference, SSVM 2011, Revised Selected Papers
A2 - BRUCKSTEIN, Alfred M.
A2 - ROMENY, Bart M. Haar
A2 - BRONSTEIN, Alexander M.
A2 - BRONSTEIN, Michael M.
PB - Springer Berlin Heidelberg
T2 - 3rd International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2011
Y2 - 29 May 2011 through 2 June 2011
ER -