An adaptive-size median filter for impulse noise removal using neural network-based detector
R. Rashidha, Philomina Simon · International Journal of Signal and Imaging Systems Engineering · 2016
In this paper, a neural network-based Adaptive-Size Median (ASMED) filter is proposed for impulse noise removal. This method consists of two stages: noise detection and noise filtering. Noise detection is performed by a neural network-based detector, and filtering is applied only to corrupt pixels in the noisy image. Extensive experimental analysis shows that the proposed technique can be used for images with different impulse noise models. Both quantitative and qualitative analyses show the superiority of the proposed filter over other existing filters. The proposed method is applied to both colour images and greyscale images, and better results are obtained.