Improve computational complexity of sobel edge detection using parallel contract anytime algorithm

Abstract

Edge detection is a considerably important factor in image or video processing. Detecting the edges of an image play a significant role in image segmentation, data compression, well matching, and image reconstruction. There are several approaches available to detect the edges of an image. In this paper we focus on Sobel edge detection using contract-time anytime algorithm in CUDA. To reduce the computational complexity we implemented our proposed edge detection method using an NVIDIA GPU. In the experimental setup we have used NVIDIA GTX 550Ti GPU along with AMD FX8150 Processor and 8 GB RAM. Finally, we measure speedup as well as quick, moderate and final (3steps of contract) of our proposed parallel implemented model. Comparing with conventional serial CPU based edge detection we have experienced maximum 4X speedup of proposed implementation for 16 block dimension.