Edge Filtering through Recursive Application using Canny Edge Detector algorithm on small sub-blocks in an Image

Edges are important features of an object in an image. With the help of edge detection, the amount of data present in the image can be reduced significantly while maintaining the structure of the object. However, factors like varying illumination, random noise present in nature and defect of sensor material all contribute to noise in an image, which leads to the detection of false edges. The performance of many computer vision algorithms relies on the performance of the feature detection algorithm. In this paper, an algorithm is proposed to apply the Canny algorithm of Edge detection on small sub-blocks of images rather than the entire image. The performance analysis of the proposed algorithm is carried out and their output is compared with a Cannyedge detection algorithm.