Edge Filtering Method for Brain MRI using CED

Edge detection is an image processing technique for finding the boundaries of objects within images. Its performance greatly influences the overall output of the computer vision application. Though it is impossible to remove all effects of noise, it usually presents itself as an unstructured collection of points in an image rather than forming straight lines. The Canny algorithm is considered as an optimal edge detector. In this paper, an algorithm is proposed to measure the probability of a point being an edge point based on the distribution of points in its surroundings. Under this condition, it was marked as an edge point by the Canny algorithm. Here, the probability is used as a threshold to remove false edges. Also, analysis of various parameters like the size of the neighborhood is carried out and their effects on the output produced are measured.