Automated restoration of semi-transparent degradation via Lie groups and visibility laws

This paper presents a novel approach for the removal of semi-transparent defects from images of historical or artistic importance. It combines Lie group transformations with human perception rules in order to make restoration more flexible and adaptable to defects having different physical or mechanical causes. In particular, the restoration process consists of an iterative procedure that gradually reduces the visual perception of the defect. It takes advantage from Lie groups that allow to define a redundant set of transformations from which it is possible to automatically select the ones that better invert the physical formation of the defect. Experimental results on movies and photographs, affected by line-scratches and semi-transparent blotches, have shown the potential of the proposed approach in giving new guidelines and trends for human perception-based restoration.