10.22376/ijpbs.2019.10.1.p1-12 Volume 6 Issue 3 2015 (July - September) The primary motivation behind image processing is to change over an image into significant information. Image Enhancement is the most important step that must be carried out in all image handling applications. A few images are inclined to noises from different sources, for example slip in camera adjustment etc. This paper imparts the identification of filter techniques strategies utilizing distinctive sorts of brain images. This paper is based on the identification analysis on the output of noisy and filtered images. This technique is acquiring the noisy images is built on the four types of images: Salt-and-pepper, Gaussian noise, Speckle Noise and Poisson Noise induced in a peripheral brain image and there are removed using four types of spatial filters: Mean Filter, Median Filter, Gaussian filter and Wiener Filter, in order to critic the efficiency of various filters over different kind of noise. An algorithm is created which execute all the sorts of filtering technique on the input image and arithmetic parameters are computed according to the correlation between output and input images. These arithmetic parameters are exhibit distinctly and they are compared for both the noisy and filtered images. For the calculation of the performance of filters the mathematical parameters are required such as Mean Square Error (MSE), Normalized Absolute Error (NAE) and Normalized correlation (NK), Peak Signal to Noise Ratio (PSNR) are used and the MATLAB codes required in calculating these parameters are developed and it has been shown that Weiner filter is an optimum filter for noisy images. R.NIVEDHA AND SP.CHOKKALINGAM Noise, Errors, De-noising, Enhancement 107-119