Effects of noise on various edge detection techniques

Edges in an image is used to extract very important features like line, curves and corners. The extraction of these features can be further used for real time purposes like face recognition, computer vision algorithms etc. But it is somewhat difficult to extract out all the edges efficiently without affecting the structural properties of an image. Edges in an image represents a swift change in the intensity of an image and noise in an image also signifies the same so what happens when noise is abundant in an image. There are different techniques used for this finding edges-Roberts, Sobel, Canny, Prewitt, Laplacian of Gaussian, and Zero Cross. In this the various effects of noise on these operators is discussed.