Impulse noise reduction using image fusion approach

Digital image enhancement has been a hot topic during the past decades. In this paper, we have established a new fusion based approach for impulse noise reduction from multi-sensors images. Image fusion is an important technique used in remote sensing, military and medical applications. In the proposed approach, firstly, only the uncorrupted pixels from multi-sensor images are grouped and then all those grouped pixels are filter with another state of the art approach for optimal results. The proposed approach is tested with several noisy images by finding the Mean Square Error, Peak Signal to Noise Ratio and Mutual Information values between the original and restored images. The experimental results show that the proposed algorithm outclasses all methods that individually de-noised and fused images.