International Journal of Computer Applications (0975 – 8887) Volume 95– No.6, June 2014 9 A Review: Various Image Denoising Techniques Archita Singh Parihar Computer Science & Engg. Department VNS faculty of engineering, Bhopal (M.P) Megha Jain Computer Science & Engg. Department VNS Faculty of Engineering, Bhopal(M.P) ABSTRACT Removal of noise is an essential and challengeable operation in image processing. Before performing any process, images must be first restored. Images may be corrupted by noise during image transmission through electronic media. Noise effect always corrupts any recorded image which is much more harmful for future process. To overcome the problem of noise level in digital images this paper present a review of different image denoising method. In this paper various filters are used for image denoising. This proposed method adopt first and second order mean filter (FSOMF) in which for first phase we detect the impulse noise. And the second phase which is also called as filtering phase replaces the detected noise pixel. Finally able to show in our experimental result of proposed method FSOMF, is capable of filtering of impulse noise. Keywords Denoising, Mean Filter, Impulse Noise. 1. INTRODUCTION Digital images plays very important role in our day to day life applications such as medical imaging, computer vision, satellite television, computer tomography as well as in areas of research and technology such as geographical information systems etc. In image processing, impulse noise reduction is an active area of research. With the usage of multimedia material becoming more prevalent from day to day, visual data from high value digital images performances a significant role. Some with low computational complexity, a good noise filter is required to satisfy two criteria namely, to reducing the noise and suppressing the useful information in the signal. Image denoising the corrupted likeness is performed by using the preprocessing process. Preprocessing is very important because subsequent procedures (e.g., likeness segmentation, enhancement, classification, parameter estimation, etc) are mostly influenced by the value of the filtered image. The main purpose of image denoising is to reform corrupted pixels estimate as in the original image from the noisy image. Image denoising and preprocessing are the important task and in digital image processing image denoising is an important area. For impulse noise reductions there are many filters available although these methods have been improved time to time, but the quality of denoised image is still not provide satisfactory results. From past few decades a number of methods have been suggested for the removal of impulse noise, nonlinear filters can be considered as the state-of-the-art methods granted their impressive performances. For example, the mean filter is one of the choice for stifling impulse noise. It is directed in order to use some filter for identically finds all the noise and noise fee pixels. But due to these methods some fine details and image borders are blurred, the mean filter often leaves attractive minutia at best blur and at worst missing. Many difficulties has been faced to the development of various categories of filters such as the mean-based filters, adaptive filters, the performances of these filters does not produce better results. In this paper, focus on providing a robust filter that detect and correct the noisy pixels i.e. for any type of impulse noise models. propose first and second order mean filter (FSOMF), for detail preserving and restoration of information. The FSOMF filter operates at a wide range of impulse noise densities without falling in a situation in which there is a danger of loss, harm, or failure of image fine details colors and textures. The proposed filter does not require any type of training algorithm and time consuming training of parameters as well. 2. LITERATURE SURVEY In 1989 A.K. Jain [1] introduced a “Fundamentals of Digital Image Processing”. In this paper first he used median filter which is one of the most popular nonlinear filters. It is very simple to implement and much efficient as well. The median filter, especially with a larger window size destroys the fine image details due to its rank ordering process. It acts like a low pass filter which blocks all high frequency components of the image like edges and noise, thus blurs the image. As the noise density increases, the filtering window size is increased to have a sufficient number of encrypted pixels in the neighborhood. Depending upon the sliding window mask, there may be many variations of median filters. In this paper, Median filter with sliding window (3×3), (5×5) and (7×7) are reviewed. Applications of the median filter require caution because median filtering tends to remove image details such as thin lines and corners while reducing noise. In 1998 Scott E Umbaugh [2], author presented mean filter acts on an image by smoothing, it also reduces the intensity variation between adjacent pixels. The mean filter is nothing but a simple sliding window spatial filter that replaces the center value in the window with the average of all the neighboring pixel values including it. By doing this, it replaces pixels that are unrepresentative of their surroundings. It is implemented with a convolution mask, which provides a result that is a weighted sum of the values of a pixel and its neighbors. It is also called a linear filter. The mask or kernel is a square. Often a 3× 3 square kernel is used. If the coefficients of the mask sum up to one, then the average brightness of the image is not changed. If the coefficients sum to zero, the average brightness is lost, and it returns a dark image. The mean or average filter works on the shift-multiply-sum principle. In July 1993 J.N. Lin, X. Nie, and R. Unbehauen [3], introduced a LMS Adaptive Filter Incorporating a Local-Mean Estimator for Image Processing. An adaptive filter does a better job of de-noising images compared to the averaging filter. The fundamental difference between the mean filter and the adaptive filter lies in the fact that the weight matrix varies after each iteration in the adaptive filter while it remains constant throughout the iterations in the mean filter. Adaptive filters are capable of de-noising non-stationary images, that is, images