Restoring a comprehensible image from a single motion-blurred image appropriate to camera shake has extended one challenging difficulty in digital imaging. Existing blind de-blurring techniques also only can eliminate simple motion blurring, or need user interactions to effort on more complex cases. In this proposed a few common types of convolution that cause previous methods to fail, such as pixel diffusion and non-Gaussian noise. We propose a new novel blur model that explicitly takes build a robust non-blind de-convolution algorithm upon it, which can effectively reduce the visual artifacts. A new joint optimization problem, which concurrently maximizes the sparsity of the blur kernel and the sparsity of the clear image under positive suitable redundant tight frame systems . Without requiring any prior information of the blur kernel is input, our proposed approach is able to recover high-quality images from given blurred images. The effectiveness of our method is established by experimental results on both synthetic and real-world examples.
A.Vidhya (2016).
"AN EFFICIENT BASED BLIND MOTION DEBLURRING USING MINIMIZATION ALGORITHM".
International Journal of Contemporary Research in Computer Science and Technology,
2(1), pp. 381-384.
