Real-Time Segmentation of Unstructured Environments by Combining Domain Generalization and Attention Mechanisms
Sensors · 28.06.2023 · 23 (13) · 6008
DOI: 10.3390/s23136008 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 28.06.2023. Verfasst von Nuanchen Lin und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Electronic Engineering (College of Artificial Intelligence), South China Agricultural University, Guangzhou 510642, China. Hauptthemen laut MEDLINE: Agriculture, Awareness. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Provincial Science and Technology Innovation Strategy Special Funds Project, The open competition program of top ten critical priorities of Agricultural Science and Technology Innovation for the 14th Five-Year Plan of Guangdong Province.
Abstract (Originalsprache)
This paper presents a focused investigation into real-time segmentation in unstructured environments, a crucial aspect for enabling autonomous navigation in off-road robots. To address this challenge, an improved variant of the DDRNet23-slim model is proposed, which includes a lightweight network architecture and reclassifies ten different categories, including drivable roads, trees, high vegetation, obstacles, and buildings, based on the RUGD dataset. The model’s design includes the integration of the semantic-aware normalization and semantic-aware whitening (SAN–SAW) module into the main network to improve generalization ability beyond the visible domain. The model’s segmentation accuracy is improved through the fusion of channel attention and spatial attention mechanisms in the low-resolution branch to enhance its ability to capture fine details in complex scenes. Additionally, to tackle the issue of category imbalance in unstructured scene datasets, a rare class sampling strategy (RCS) is employed to mitigate the negative impact of low segmentation accuracy for rare classes on the overall performance of the model. Experimental results demonstrate that the improved model achieves a significant 14% increase mIoU in the invisible domain, indicating its strong generalization ability. With a parameter count of only 5.79M, the model achieves mAcc of 85.21% and mIoU of 77.75%. The model has been successfully deployed on a a Jetson Xavier NX ROS robot and tested in both real and simulated orchard environments. Speed optimization using TensorRT increased the segmentation speed to 30.17 FPS. The proposed model strikes a desirable balance between inference speed and accuracy and has good domain migration ability, making it applicable in various domains such as forestry rescue and intelligent agricultural orchard harvesting.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Electronic Engineering (College of Artificial Intelligence), South China Agricultural University, Guangzhou 510642, China.
ORCID-Profile: Nuanchen Lin, Shenghao Liang
Förderung
- Guangdong Provincial Science and Technology Innovation Strategy Special Funds Project
- The open competition program of top ten critical priorities of Agricultural Science and Technology Innovation for the 14th Five-Year Plan of Guangdong Province
Zitieren
APA
Lin, N., Zhao, W., Liang, S., & Zhong, M. (2023). Real-Time Segmentation of Unstructured Environments by Combining Domain Generalization and Attention Mechanisms. Sensors, 23(13), 6008. https://doi.org/10.3390/s23136008
Vancouver
Lin N, Zhao W, Liang S, Zhong M. Real-Time Segmentation of Unstructured Environments by Combining Domain Generalization and Attention Mechanisms. Sensors. 2023;23(13):6008. doi:10.3390/s23136008
BibTeX
@article{Lin2023,
author = {Lin, Nuanchen and Zhao, Wenfeng and Liang, Shenghao and Zhong, Minyue},
title = {Real-Time Segmentation of Unstructured Environments by Combining Domain Generalization and Attention Mechanisms},
journal = {Sensors},
year = {2023},
volume = {23},
number = {13},
pages = {6008},
doi = {10.3390/s23136008}
}
EUVIMED