Learning Light Fields for Improved Lane Detection
Robust lane detection is imperative for the realization of intelligent transportation. Recently, vision-based systems that employ deep convolution neural networks (CNNs) for lane detection have made considerable progress. However, for better generalization under various road conditions learning-base...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9999224/ |
_version_ | 1797956347765981184 |
---|---|
author | Muhamad Zeshan Alam Sousso Kelouwani Jonathan Boisclair Ali Akrem Amamou |
author_facet | Muhamad Zeshan Alam Sousso Kelouwani Jonathan Boisclair Ali Akrem Amamou |
author_sort | Muhamad Zeshan Alam |
collection | DOAJ |
description | Robust lane detection is imperative for the realization of intelligent transportation. Recently, vision-based systems that employ deep convolution neural networks (CNNs) for lane detection have made considerable progress. However, for better generalization under various road conditions learning-based methods require excessive training data, which becomes non-trivial in challenging conditions such as illumination variation, shadows, false lane lines, and worn lane markings, etc. In this paper, we propose a light field (LF) based lane detection method that utilizes the additional angular information for improved prediction and increased robustness. Two different LF representations are investigated to study the possibility of maximum performance improvement and minimal additional computation cost and data labeling efforts. Experimental results successfully demonstrate that the proposed approach improves the prediction of the lane line point coordinates and is significantly robust against the aforementioned adverse conditions. |
first_indexed | 2024-04-10T23:48:40Z |
format | Article |
id | doaj.art-f3e3c0e933d14da7aabab1a1ff85de8a |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-10T23:48:40Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-f3e3c0e933d14da7aabab1a1ff85de8a2023-01-11T00:00:41ZengIEEEIEEE Access2169-35362023-01-011127128310.1109/ACCESS.2022.32321279999224Learning Light Fields for Improved Lane DetectionMuhamad Zeshan Alam0https://orcid.org/0000-0002-0114-8248Sousso Kelouwani1Jonathan Boisclair2Ali Akrem Amamou3Department of Computer Science and Mathematics, University of Brandon, Brandon, CanadaDepartment of Mechanical Engineering, University of Quebec at Trois Rivieres, Trois Rivières, CanadaDepartment of Mechanical Engineering, University of Quebec at Trois Rivieres, Trois Rivières, CanadaDepartment of Mechanical Engineering, University of Quebec at Trois Rivieres, Trois Rivières, CanadaRobust lane detection is imperative for the realization of intelligent transportation. Recently, vision-based systems that employ deep convolution neural networks (CNNs) for lane detection have made considerable progress. However, for better generalization under various road conditions learning-based methods require excessive training data, which becomes non-trivial in challenging conditions such as illumination variation, shadows, false lane lines, and worn lane markings, etc. In this paper, we propose a light field (LF) based lane detection method that utilizes the additional angular information for improved prediction and increased robustness. Two different LF representations are investigated to study the possibility of maximum performance improvement and minimal additional computation cost and data labeling efforts. Experimental results successfully demonstrate that the proposed approach improves the prediction of the lane line point coordinates and is significantly robust against the aforementioned adverse conditions.https://ieeexplore.ieee.org/document/9999224/Lane detectionlight field imagingconvolutional neural networksintelligent transportation |
spellingShingle | Muhamad Zeshan Alam Sousso Kelouwani Jonathan Boisclair Ali Akrem Amamou Learning Light Fields for Improved Lane Detection IEEE Access Lane detection light field imaging convolutional neural networks intelligent transportation |
title | Learning Light Fields for Improved Lane Detection |
title_full | Learning Light Fields for Improved Lane Detection |
title_fullStr | Learning Light Fields for Improved Lane Detection |
title_full_unstemmed | Learning Light Fields for Improved Lane Detection |
title_short | Learning Light Fields for Improved Lane Detection |
title_sort | learning light fields for improved lane detection |
topic | Lane detection light field imaging convolutional neural networks intelligent transportation |
url | https://ieeexplore.ieee.org/document/9999224/ |
work_keys_str_mv | AT muhamadzeshanalam learninglightfieldsforimprovedlanedetection AT soussokelouwani learninglightfieldsforimprovedlanedetection AT jonathanboisclair learninglightfieldsforimprovedlanedetection AT aliakremamamou learninglightfieldsforimprovedlanedetection |