A human-machine interactive system for efficient image restoration

Images are produced in order to record or display useful information. Due to imperfections in the electronic or photographic medium, however, the recorded image often represents a degraded version of the original scene. The degradations may have many causes, but two types of degradations are often dominant: blurring and noise. In this paper, a practical human-machine interactive system for image restoration (HMISIR) is presented to remove high degradations on the optical images. This system includes a human machine interactive process to perform efficient noise-filtering and deblurring procedures. It is shown that this system can produce high quality restoration of blurred and noisy images, especially when the signal-to-noise ratio is low