The picture noise is an irregular variation of brightness and color information in pictures. It decreases picture quality and permeability of specific elements inside the picture. The most surely understood noise that corrupts the photo with impulse noise. In this work, an effective algorithm is intended to identify and remove noise from a picture. An improved de-noising calculation in view of the median filter is exhibited for greyscale and colored images. The algorithm incorporates two cases: I) if the chose window contains all pixel values "0" to "255" at that point center preparing pixel supplanted by the mean of qualities. II) If the chosen window does not contain all components "0" and "255" then eliminate "0" and "255" and central preparing pixel is replaced by the median of remaining pixels values. The performance is checked off the purposed algorithm by comparing it with corresponding filters. The experiment checked at various noise proportion 5% to 80% for greyscale and color pictures. Results are checked as far as MSE and PSNR and even at high noise proportion; it gives better outcomes over other existing techniques.

Image division refers to the way toward dividing an advanced picture into various portions. Image division says to a parcel of an image into various divisions that are homogeneous or comparable. The objective of the division is to Simplify or potentially changes the portrayal of an image into something that is more important and simpler to dissect. Advancement of precise image division different image division strategies is utilized to take care of a particular issue. The motivation behind this survey is to give an overview of various image division methods. These methods are sorted into four sorts: an) Edge based division b) Threshold Segmentation c) Clustering-based division D) Region-based division. This survey tended to different image division methods, their correlation and presents the issues identified with those procedures.

Image de-noising is a procedure that used to upgrade the picture quality after corrupted by the noise. There are a few techniques have been proposed for picture de-noising. Noise lessening and reclamation of image is relied upon to enhance the subjective review of a picture and the execution criteria of quantitative picture examination systems Digital picture is slanted to an assortment of commotion which influences the nature of picture. The criteria of the commotion expulsion issue rely on upon the noise sort by which the picture is defiling. To diminish the image commotion a few sorts of direct and non strategies separating methods and de-noising calculation have been proposed. Straight channels are not ready to successfully take out motivation commotion as they tend to obscure the edges of a picture. Then again non straight channels are suited for managing drive commotion. Diverse methodologies for decrease of commotion and image upgrade have been viewed as, each of which has their own restriction and favorable circumstances.