An Integrated Approach of Noise Reduction in Color Image Using Soft Computing

Abstract

During image pre-processing, noise reduction is considered as one of the essential step hence considered as an active area of research in image processing. Various filtering techniques are available for removing the noise from the image with their own set of merits and demerits. But most of the previously developed algorithms are either limited to reduce only a specific type of noise or fails in preserving the fine structures of the images like edge, boundaries etc. because of which loss of information occurs within the images. So, this does not lead to an effective noise reduction mechanism. In this paper, an integrated approach is proposed that will make the system intelligent enough so that it will not only limited to reduce the noise effectively by preserving the fine structures but later on will also be able to classify the images by checking them for noise present in them. Keywords: Noise, Image, Salt-and-Pepper, Speckle, DWT, Neural Network, De-noising, Noise Reduction, Wavelet, Pixel