International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 3 Issue: 12 6820 - 6828 ______________________________________________________________________________________ 6820 IJRITCC | December 2015, Available @ http://www.ijritcc.org _______________________________________________________________________________________ A Survey of Non-Linear Filtering Techniques for Image Noise Removal 1 Ms. Mrunali P. Mahajan (ME, Student) 1 Dept of Electronics Engineering, SSVPS’s B.S. Deore College of Engg,NMU Dhule mahajan.mrunali@gmail.com 2 Prof. Sanjeev N Jain (Professor) Dept of Electronics Engineering, SSVPS’s BS Deore College of Engg,NMU Dhule Sanjeev_n_jain@yahoo.com Abstract: Image is captured or noninheritable by any image capturing device like camera or scanner and then it is stored in the mass storage of the computer system. In many of these applications the existence of impulsive noise among the noninheritable pictures is one altogether common problem. This noise is characterized by spots on the image and is usually related to the innate image because of errors in image sensors and information transmission. Now-a-days there are numerous strategies that are offered to remove noise from digital images. Most of the novel methodology includes 2 stages: the primary stage is to find the noise within the image and the second stage is to eliminate the noise from the image. This paper explores the varied novel methods for the removal of noise from the digital images. The distinctive feature of the all the described filters is that offers well line, edge and detail preservation performance while, at the constant time, effectively removing noise from the input image. In later section, we present a short introduction for various strategies for noise reduction in digital images. Keywords: Decision based median filter, negative selection algorithm, and salt and pepper noise, MSE, PSNR, IEF. __________________________________________________*****_________________________________________________ I. INTRODUCTION Digital image processing is one of most useful technique for computer algorithms to perform image processing on digital images [1]. A digital image is consist of a finite number of elements called pixels, every of that includes an explicit location and value. Image is captured or noninheritable by any image capturing device like camera or scanner and then it is stored in the mass storage of the computer system. Then it is processed exploitation the image processing/editing tool like MATLAB, photo- shop etc. so displayed on the displaying device and conjointly transmitted to the owner needed party. The image is subjected to many kinds of distortion during the stages that it might pass through such as such as storing, processing, compressing and transmitting etc. During transfer noise may get added along the actual information. Noise is essentially unwanted information that gets added along the required information due to certain environmental variations, faulty locations in a memory or during transfer [2]. Noise may be arise at the time of transmitting or clicking or taking images. Whenever noise is introduced, the each pixel present within the image show completely different values of intensity instead of the true pixel values. Image noise may be categorised as Impulse Noise (Salt & Pepper noise), Gaussian Noise (Amplifier noise), Multiplicative Noise (Speckle Noise), and Poisson Noise (Photon Noise). 1.1 Filtering To recover the image from its noise there exits many median filtering techniques which are application oriented. Some filtering techniques have better effects than the others according to noise category. Median filtering techniques are described below