Fast and accurate implementation of Canny edge detector on embedded many-core platform

Image processing and computer vision applications are used intensively in several domains in particular multimedia and medicine. The main challenge in developing such applications is how to guarantee both high accuracy and low execution time. Accordingly, we observe two research directions: the first focuses on improving the algorithms and the second focuses on designing fast hardware platforms. In this paper, we propose an efficient parallel implementation of an accurate extended Canny edge detection algorithm suitable for medical applications on embedded many-core platform. The proposed implementation is running at a frame rate of 10 frames/s for image size of 512×512 with high accurate and smooth line edges.