Abstract: MR images due to the defect of imaging devices or the patients motion including voluntary and involuntary such as breathing may be blurred. According to the role of these types of images in the diagnosis of disease and therefore persons health, their desirable quality is very important and would be deblurred. Various algorithms have been proposed to deblurring in both frequency and spatial domains but most of these algorithms take into account the limitations on algorithms such that only one type of blur can deblur or consider the blur factor in only one direction. The objective of this study is to propose an algorithm that can address deblurring without restrictions in order to the motion of objects or blured factor. In the proposed algorithm, the sparseness of image edges property is applied in iterative deblurring which uses only one uniform image with unknown blur kernel is uncertain. The experimental results show that due to the good quality of deblurring with this algorithm, the restored images is contained with roughness and smoothing algorithm can improve image quality about 20% and it is comparable with art deblurring algorithms and is a suitable approach to solve the problem.
Hamideh Farajpour Pirbasti and Asadollah Shahbahrami, 2016. Deblurring MR Images with Two-Step Algorithm. Asian Journal of Information Technology, 15: 1-6.