HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection

Abstract Road detection plays a vital role in automated driving and advanced driver assistance systems. In recent years, mainstream frameworks have suffered from the restrictions of a receptive field and the limitation of modelling long‐range relations. Previous methods fail to segment precise road...

Full description

Saved in:
Bibliographic Details
Main Authors: Ting Han, Chuanmu Li, Siyu Chen, Zongyue Wang, Jinhe Su, Yundong Wu, Guorong Cai
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:IET Intelligent Transport Systems
Subjects:
Online Access:https://doi.org/10.1049/itr2.12360
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850117139768279040
author Ting Han
Chuanmu Li
Siyu Chen
Zongyue Wang
Jinhe Su
Yundong Wu
Guorong Cai
author_facet Ting Han
Chuanmu Li
Siyu Chen
Zongyue Wang
Jinhe Su
Yundong Wu
Guorong Cai
author_sort Ting Han
collection DOAJ
description Abstract Road detection plays a vital role in automated driving and advanced driver assistance systems. In recent years, mainstream frameworks have suffered from the restrictions of a receptive field and the limitation of modelling long‐range relations. Previous methods fail to segment precise road boundaries when urban roads with similar surface textures are presented. Moreover, road regions are perceived as non‐road areas due to the shadow effect, which affects the completeness of the road in the traffic environment. To this end, for urban road detection, a hierarchical enhanced attention transformation (HEAT) architecture, which holds both fine details (road edges) and global contextual information (road structure) is proposed. The symmetrical data‐fusion residual network fuses visual semantic and spatial structure information. The attention consolidation units model global contextual information at different layers to realize the feature enhancement from coarse to fine. In addition, corresponding local and global features are fused hierarchically in the progressive up‐sampling modules. Comprehensive empirical studies are conducted to compare other mainstream methods in the KITTI and the Cityscapes dataset. HEAT shows highly competitive performance in that confusable areas are correctly distinguished in the presence of obstacles, shadows, and similar road textures. HEAT outperforms state‐of‐the‐art methods in urban road detection.
format Article
id doaj-art-46641a97eccb4b24a0441751d9136389
institution OA Journals
issn 1751-956X
1751-9578
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series IET Intelligent Transport Systems
spelling doaj-art-46641a97eccb4b24a0441751d91363892025-08-20T02:36:09ZengWileyIET Intelligent Transport Systems1751-956X1751-95782024-12-0118122532255110.1049/itr2.12360HEAT: Incorporating hierarchical enhanced attention transformation into urban road detectionTing Han0Chuanmu Li1Siyu Chen2Zongyue Wang3Jinhe Su4Yundong Wu5Guorong Cai6School of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaSchool of Computer Engineering Jimei University Xiamen ChinaAbstract Road detection plays a vital role in automated driving and advanced driver assistance systems. In recent years, mainstream frameworks have suffered from the restrictions of a receptive field and the limitation of modelling long‐range relations. Previous methods fail to segment precise road boundaries when urban roads with similar surface textures are presented. Moreover, road regions are perceived as non‐road areas due to the shadow effect, which affects the completeness of the road in the traffic environment. To this end, for urban road detection, a hierarchical enhanced attention transformation (HEAT) architecture, which holds both fine details (road edges) and global contextual information (road structure) is proposed. The symmetrical data‐fusion residual network fuses visual semantic and spatial structure information. The attention consolidation units model global contextual information at different layers to realize the feature enhancement from coarse to fine. In addition, corresponding local and global features are fused hierarchically in the progressive up‐sampling modules. Comprehensive empirical studies are conducted to compare other mainstream methods in the KITTI and the Cityscapes dataset. HEAT shows highly competitive performance in that confusable areas are correctly distinguished in the presence of obstacles, shadows, and similar road textures. HEAT outperforms state‐of‐the‐art methods in urban road detection.https://doi.org/10.1049/itr2.12360advanced driver assistance systemscomputer visionconvolutional neural netsglobal feature transformationimage segmentation
spellingShingle Ting Han
Chuanmu Li
Siyu Chen
Zongyue Wang
Jinhe Su
Yundong Wu
Guorong Cai
HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
IET Intelligent Transport Systems
advanced driver assistance systems
computer vision
convolutional neural nets
global feature transformation
image segmentation
title HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
title_full HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
title_fullStr HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
title_full_unstemmed HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
title_short HEAT: Incorporating hierarchical enhanced attention transformation into urban road detection
title_sort heat incorporating hierarchical enhanced attention transformation into urban road detection
topic advanced driver assistance systems
computer vision
convolutional neural nets
global feature transformation
image segmentation
url https://doi.org/10.1049/itr2.12360
work_keys_str_mv AT tinghan heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT chuanmuli heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT siyuchen heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT zongyuewang heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT jinhesu heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT yundongwu heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection
AT guorongcai heatincorporatinghierarchicalenhancedattentiontransformationintourbanroaddetection