Ann based image restoration in approach of multilayer perceptron
Prabira Kumar Sethy, Leeza Panda, Santi Kumari Behera · 2016
Image restoration is the method of fixing degraded images which cannot be taken again or the process of obtaining the image again is expensive. Image restoration is done in two fields: spatial domain and frequency domain. In spatial domain the refining action for restoring the images is done by directly operating on the pixels of the digital image. Restoration adequacy was checked by taking peak signal to noise ratio (PSNR) and mean square error (MSE) into considerations. There are several approaches for image recognition. Among those approaches, application of soft computing models on digital image has been considered to be an approach with a better result. The main objective of the present work is to provide a new approach for image identification using Artificial Neural Networks. Initially an original gray scale intensity image will been taken for transformation. The Input image will then be added with Salt and Pepper noise. Adaptive median Filter will be applied on noisy image such that the noise can be removed and the output image would be considered as Refined Image. The estimated Error and average error of the values stored in filtered image matrix will then be calculated with reference to the values stored in original data matrix for the purpose of checking of proper noise removal.