In many visual problems, line variables, associated to the discontinuities present in the ideal image, are used to obtain better results. In this paper we propose to extend the use of line variables to the high-resolution reconstruction problem. Moreover, we impose that parallel discontinuities should not be close. To minimize the energy function related to this problem we use a CATILED (convex approximation technique for interacting line elements deblurring) algorithm. Experimental results confirm the goodness of this technique.
An edge-preserving high-resolution image reconstruction
2003-01-01
582538 byte
Conference paper
Electronic Resource
English
An Edge-Preserving High-Resolution Image Reconstruction
British Library Conference Proceedings | 2003
|An Edge-Preserving Image Reconstruction Using Neural Network
British Library Online Contents | 2001
|Edge-and-corner preserving regularization for image interpolation and reconstruction
British Library Online Contents | 2008
|An Iterative Method for Preserving Edges and Reducing Noise in High Resolution Image Reconstruction
British Library Conference Proceedings | 2006
|Image enhancement by edge-preserving filtering
IEEE | 1994
|