Edge quality based iterative deconvolution algorithm for motion blurred gray scale images for geometrical measures
Silvio Holder, Ke Xie, G Linß · Journal of Physics Conference Series · 2010
The main purpose of this paper is to present an algorithm for correcting motion blurred gray scale images iteratively by determination of edge parameters like the edge width. These parameters are used to describe a motion point spread function for the usage in a closed-loop deconvolution algorithm. The image restoration works well with motion blurred images with additional noise and was tested with realistic and synthetic motion blurred images. Furthermore the variation of the edge position in the image was determined by four different edge detection methods in pixel domain to proof the functionality of the restoration algorithm.