Efficient blind blur identification using discrete periodic Radon transform

The problem of restoring an image from its convolution with an unknown blur function is a well-known problem in image processing. Many approaches have been proposed to solve the problem and they have shown to have good performance in identifying the blur function or retrieving the original image. However, in actual implementation, various problems incurred due to the large data size and long computational time of these approaches are undesirable, even with the current computing machines. An efficient algorithm is proposed for multichannel blind blur identification based on the discrete periodic Radon transform (DPRT). With the DPRT, the original 2-dimensional multichannel blind blur identification problem can be converted into 1-dimensional probelms, which greatly reduces the memory size and computational time required. Experimental results show that a 44% reduction in the number of operations can be achieved as compared to the traditional approach.