We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Restoration of multispectral images distorted by spatially nonuniform camera motion.
- Authors
Kober, V.; Karnaukhov, V.
- Abstract
The majority of known methods for restoration of distorted multispectral images use a model of blur as the convolution of the initial image with a uniform kernel. However, the image blur in real low-light conditions is often spatially nonuniform, since the camera in an actual shoot is shaken and rotated in three dimensions while its shutter is open. A method for 'blind' locally adaptive image restoration in the domain of sliding orthogonal transform is proposed. It is assumed that the signal blur operator is spatially homogeneous in a small-sized sliding window. The known approaches to the restoration of color images blurred by spatially nouniform camera motion are compared to the proposed algorithm.
- Subjects
IMAGE reconstruction; SPATIAL analysis (Statistics); NON-uniform motion; COMPUTER algorithms; MATHEMATICAL convolutions
- Publication
Journal of Communications Technology & Electronics, 2015, Vol 60, Issue 12, p1366
- ISSN
1064-2269
- Publication type
Article
- DOI
10.1134/S1064226915120153