Language:

Scene recognition under special traffic conditions based on deep multi-task learning

Journal of engineering (Stevenage, England), 2020-07, Vol.2020 (13), p.471-475 [Peer Reviewed Journal]

2021 The Institution of Engineering and Technology ;ISSN: 2051-3305 ;EISSN: 2051-3305 ;DOI: 10.1049/joe.2019.1191

Full text available

Citations Cited by

Actions
1. Add to My Research
2. Remove from My Research
3. E-mail
4. Print
5. Permalink
6. Citation
7. EasyBib
8. EndNote
9. RefWorks
10. Delicious
11. Export RIS
12. Export BibTeX

Title:
Scene recognition under special traffic conditions based on deep multi-task learning
Author: Hu, Xiaochang ; Xu, Xin ; Xiao, Yongqian ; Chen, Hongjun ; Zhang, Hongjia
Subjects: auxiliary tasks ; chief task ; deep multitask classification framework ; deep multitask learning ; image recognition ; learning (artificial intelligence) ; learning tasks ; neural nets ; pattern classification ; promising yet challenging tasks ; recognition model ; recognition performance ; special conditions ; special traffic scenes ; task‐specific sub‐networks ; The 3rd Asian Conference on Artificial Intelligence Technology (ACAIT 2019) ; traffic scene recognition ; typical special traffic conditions
Is Part Of: Journal of engineering (Stevenage, England), 2020-07, Vol.2020 (13), p.471-475
Description: Traffic scene recognition under special conditions is one of the most promising yet challenging tasks for autonomous driving systems. This study presents a deep multi-task classification framework for scene recognition involving special traffic conditions. The framework incorporates four learning tasks where the recognition of special traffic scenes is the chief task and the time of occurrence (daytime or night-time), the weather type and the road attribute are the three auxiliary tasks for improving the recognition performance. The four tasks share the feature map generated by a convolutional neural network followed by task-specific sub-networks which are merged in the end via a joint loss function. Moreover, a small dataset of typical special traffic conditions was built for training and testing the recognition model. Experimental results demonstrate that the proposed framework significantly improves the accuracy of scene recognition under special traffic conditions.
Publisher: The Institution of Engineering and Technology
Language: English
Identifier: ISSN: 2051-3305
EISSN: 2051-3305
DOI: 10.1049/joe.2019.1191
Source: Directory of Open Access Journals

Back to results list


INSPIRE LIBRARY - TON DUC THANG UNIVERSITY	(84-028) 37 755 057	Feedback
19 Nguyen Huu Tho St. Dist.7, HCM	thuvien@tdtu.edu.vn	Feedback

Scene recognition under special traffic conditions based on deep multi-task learning

2021 The Institution of Engineering and Technology ;ISSN: 2051-3305 ;EISSN: 2051-3305 ;DOI: 10.1049/joe.2019.1191

Searching Remote Databases, Please Wait