Work

Efficient and real-time lane detection using CUDA-based implementation

Year: 2024

Type: article

Abstract: \nLane detection is an essential component of autonomous driving systems, enabling vehicles to accurately identify and follow road markings. In this paper, we look at an lane detection approach that integrates median filtering and the Hough transform. Median filtering is an essential pre-processing step for reducing noise and improving lane detection accuracy. However, given its high computational demands, optimization of this process is essential for real-time applications. To this end, we used CUDA for acceleration, taking advantage of its parallel computing capabilities to improve performance. We implemented and tested this optimised lane detection algorithm on the NVIDIA Jetson Nano and on a desktop, providing a comparative analysis of improvements in efficiency and speed. This approach highlights the potential of real-time path detection in embedded and high-performance computing environments.\n

Source: Springer Link (Chiba Institute of Technology)

DOI: 10.1051/itmconf/20246901007/pdf

Authors: Hoda El Boussaki, Rachid Latif, Amine Saddik

Language: English

Topic: Autonomous Vehicle Technology and Safety

Subfield: Automotive Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: green

Funders: Ministère de l'Education Nationale, de la Formation professionnelle, de l'Enseignement Supérieur et de la Recherche Scientifique, Centre National pour la Recherche Scientifique et Technique

Award: G4850730461

OpenAlex ID: W6884564529

FWCI:

Cites:

Cited by:

Related to: