IMPULSIVE NOISE DETECTION BY SECOND ORDER DIFFERENTIAL IMAGE AND NOISE REMOVAL USING ADAPTIVE NEAREST NEIGHBOURHOOD FILTER K.Somasundaram * M.Sc., Ph.D, and P.Shanmugavadivu M.C.A. Department of Computer Science and Applications Gandhigram Rural University, Gandhigram - 624 302, Tamil Nadu, India Abstract A novel impulsive noise detection method based on the principle that the difference between the noisy pixel and the nearest good pixel will be different from the difference between two nearby good pixels. This is achieved by constructing a second order differential image. Three new noise removal methods are presented. Simulated results show that the proposed filter gives far better results than many existing filters and is comparable to the results obtained by JM filter based on Jarque-Bera test. Our filter is computationally simpler than JM and is likely to be the fastest one. Keywords: Impulsive noise, noise filters, highly corrupted image, noise detection, differential image 1. Introduction An image gets corrupted by impulse noise due to the limitations in the camera sensors during acquisition or in the transmission channels [1]. Fixed valued impulse and random valued impulse noises are the two variants of impulse noise. Salt and pepper noise is a fixed valued impulse noise which is characterized by the maximum and the minimum value in the corrupted image. To improve the quality of corrupted images, effective image enhancement techniques are frequently used to attenuate noise as well as to preserve the details of the image [2]. Though both linear and nonlinear filters are * Corresponding author. Tel:91-451-2452371; Fax: 91-451-25453071 e-mail address :somasundaramk@yahoo.com (K.Somasundaram), psvadivu@yahoo.com (P.Shanmugavadivu)