Epurate-Net: Efficient Progressive Uncertainty Refinement Analysis for Traffic Environment Urban Road Detection
- 31 January 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (7) , 6617-6632
Abstract
High-performance and real-time road detection plays an essential role in Advanced Driver Assistance Systems (ADAS) of intelligent transportation. However, existing approaches still suffer from ambiguous road contour in traffic environment because deep learning methods lack explicit constraints on road boundaries with similar textures and structures. To address the unsatisfactory boundaries, we propose an efficient architecture for urban road detection to refine road edges adaptively. First, we design a lightweight symmetrical data-fusion network to merge spatial responses into visual features. Second, we construct cross-layer attention transformation to aggregate non-local contextual information. Moreover, a progressive uncertainty analysis module eliminates indistinct road and obstacle edges. Finally, we introduce upgrade uncertainty loss and improved deep supervision to constrain margin error for multi-scale predictions. Results of experiments using three famous datasets confirm the superiority of our method (F1-measure of 96.91% in KITTI, 98.86% in Cityscapes, and 95.18% in R2D, processing speed of 0.02s) over previous approaches. We demonstrate that, to ensure the safety of autonomous driving, the Epurate-Net adaptively refines road contour to reach exquisite road margins. The source code will be available soon.Keywords
Funding Information
- National Natural Science Foundation of China (42371457, 41971424, 42301468, 61902330)
- Key Project of Natural Science Foundation of Fujian Province (2022J02045)
- Natural Science Foundation of Fujian Province (2020J01701, 2022J01337, 2022J01819)
This publication has 72 references indexed in Scilit:
- HEAT: Incorporating hierarchical enhanced attention transformation into urban road detectionIET Intelligent Transport Systems, 2023
- k-means Mask Transformer2022
- InverseForm: A Loss Function for Structured Boundary-Aware Segmentation2021
- LIDAR–camera fusion for road detection using fully convolutional neural networksRobotics and Autonomous Systems, 2019
- Reverse Attention for Salient Object Detection2018
- Hybrid conditional random field based camera-LIDAR fusion for road detectionInformation Sciences, 2018
- RBNet: A Deep Neural Network for Unified Road and Road Boundary Detection2017
- Feature Pyramid Networks for Object Detection2017
- Deep Residual Learning for Image Recognition2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016