Showing 1–1 of 1 results for author: Alshemi, M

Hardware Acceleration of Lane Detection Algorithm: A GPU Versus FPGA Comparison

Authors: Mohamed Alshemi, Sherif Saif, Mohamed Taher

Abstract: A Complete Computer vision system can be divided into two main categories: detection and classification. The Lane detection algorithm is a part of the computer vision detection category and has been applied in autonomous driving and smart vehicle systems. The lane detection system is responsible for lane marking in a complex road environment. At the same time, lane detection plays a crucial role i… ▽ More A Complete Computer vision system can be divided into two main categories: detection and classification. The Lane detection algorithm is a part of the computer vision detection category and has been applied in autonomous driving and smart vehicle systems. The lane detection system is responsible for lane marking in a complex road environment. At the same time, lane detection plays a crucial role in the warning system for a car when departs the lane. The implemented lane detection algorithm is mainly divided into two steps: edge detection and line detection. In this paper, we will compare the state-of-the-art implementation performance obtained with both FPGA and GPU to evaluate the trade-off for latency, power consumption, and utilization. Our comparison emphasises the advantages and disadvantages of the two systems. △ Less

Submitted 19 December, 2022; originally announced December 2022.