Open AccessImplementation of Real-Time Edge Detection Using Canny and Sobel Algorithms
Author(s) -
Norman Lynn,
A I Sourav,
A J Santoso
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1096/1/012079
Subject(s) - canny edge detector , sobel operator , deriche edge detector , edge detection , computer science , image gradient , artificial intelligence , computer vision , algorithm , blob detection , image processing , image (mathematics)
The process of performing tasks related to image processing requires the use of potential techniques such as edge detection. The technique can be used to accomplish tasks related to video and image manipulation. The process of edge detection is very virtual as it acts as the first phase of image analysis and understanding. In this paper, the edges of real-time images were detected for easy extraction of meaningful information from images. A prototype was implemented on OpenCV which is an in-built function cv2. () using python. Canny and Sobel edge detection algorithms are chosen to be used in this paper for comparison purposes to find out which method is better at edge detection. After using both algorithms to detect edges of real-time images, the result showed that the Canny algorithm produced thick edges compared to the Sobel algorithm. The canny algorithm follows the criteria of; Good detection, Good localization, and Minimal response, to output good edge detection results More so, canny uses a double threshold for edge revelation and applies the Gaussian filter which removes of any noise from an image, unlike Sobel which is not resistant to noise. The developed prototype for this study can detect edges of real-time objects without the intervention of any sensor input.