Research Article Impact Factor: 4.226 ISSN: 2319-507X Samira Vengurlekar, IJPRET, 2015; Volume 3 (9): 361-368 IJPRET

\ Abstract: Edge detection plays an important role in object identification. Application ranges from machine vision, image processing to pattern recognition. There exist many algorithms for detecting edges such as sobel, prewitt, canny, morphological operators etc. Canny operator is more susceptible to noise. This paper analyses the advantages and disadvantages of Canny operator and Morphological operator in the aspect of image edge extraction, and proposes a new edge extraction algorithm by combining two methods. The image edges are extracted using Morphological operator and Canny operator respectively. Then the Canny edges are restraint using the morphological edge. The edge points which are present in Canny edge but not in the morphological edge are discarded so that the noise in the Canny edge is eliminated. Morphological operator finds the structuring element from the image itself using freeman chain code, which is then followed by the Morphological Gradient method to detect edges. The proposed algorithm is anti-noise, simple, efficient and fast.