SPVINet: A Lightweight Multitask Learning Network for Assisting Visually Impaired People in Multiscene Perception

Պահպանված է:
Մատենագիտական մանրամասներ
Հրատարակված է:IEEE Internet of Things Journal vol. 11, no. 11 (2024), p. 20706
Հիմնական հեղինակ: Hong, Kaipeng
Այլ հեղինակներ: He, Weiqin, Tang, Hui, Zhang, Xing, Li, Qingquan, Zhou, Baoding
Հրապարակվել է:
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Խորագրեր:
Առցանց հասանելիություն:Citation/Abstract
Ցուցիչներ: Ավելացրեք ցուցիչ
Չկան պիտակներ, Եղեք առաջինը, ով նշում է այս գրառումը!

MARC

LEADER 00000nab a2200000uu 4500
001 3058293262
003 UK-CbPIL
022 |a 2327-4662 
024 7 |a 10.1109/JIOT.2024.3371978  |2 doi 
035 |a 3058293262 
045 2 |b d20240101  |b d20241231 
084 |a 267632  |2 nlm 
100 1 |a Hong, Kaipeng  |u School of Architecture and Urban Planning, Guangdong Key Laboratory of Urban Informatics, MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen Key Laboratory of Spatial Information Smart Sensing and Services, Shenzhen University, Shenzhen, China 
245 1 |a SPVINet: A Lightweight Multitask Learning Network for Assisting Visually Impaired People in Multiscene Perception 
260 |b The Institute of Electrical and Electronics Engineers, Inc. (IEEE)  |c 2024 
513 |a Journal Article 
520 3 |a Visual perception technology is an important means to facilitate safe navigation for visually impaired people based on Internet of Things (IoT)-enabled camera sensors. However, due to the rapid development of urban traffic systems, traveling outdoors is becoming increasingly complicated. Visually impaired individuals must implement different types of tasks simultaneously, such as finding roads, avoiding obstacles, and viewing traffic lights, which is challenging for both them and navigation assistance methods. To solve these problems, we propose a multitask visual navigation method for visually impaired individuals using an IoT-based camera. A lightweight neural network is designed, which adopts a multitask learning architecture to perform scene classification and path detection tasks simultaneously. We propose two modules, i.e., an enhanced inverted residuals (EIRs) block and a lightweight vision transformer (ViT) block (LWVIT block), to effectively combine the properties of convolutional neural networks (CNNs) and ViT networks. The two modules allow the network to better learn local features and global representations of images while remaining lightweight. The experimental results show that the proposed method can achieve these tasks simultaneously in a lightweight manner, which is important for IoT-based navigation applications. The accuracy of our method in scene classification reaches 91.7%. The path direction and endpoint detection errors are 6.59° and 0.09, respectively, for blind road and 6.81° and 0.06, respectively, for crosswalk. The number of parameters of our method is 0.993 M, which is smaller than that of the comparison methods. An ablation study further demonstrates the effectiveness of the proposed method. 
653 |a Visual impairment 
653 |a Navigation 
653 |a People with disabilities 
653 |a Internet of Things 
653 |a Classification 
653 |a Visual perception 
653 |a Traffic signals 
653 |a Artificial neural networks 
653 |a Neural networks 
653 |a Cameras 
653 |a Ablation 
653 |a Modules 
653 |a Obstacle avoidance 
700 1 |a He, Weiqin  |u School of Architecture and Urban Planning, Guangdong Key Laboratory of Urban Informatics, MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen Key Laboratory of Spatial Information Smart Sensing and Services, Shenzhen University, Shenzhen, China 
700 1 |a Tang, Hui  |u School of Architecture and Urban Planning, Guangdong Key Laboratory of Urban Informatics, MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen Key Laboratory of Spatial Information Smart Sensing and Services, Shenzhen University, Shenzhen, China 
700 1 |a Zhang, Xing  |u School of Architecture and Urban Planning, Guangdong Key Laboratory of Urban Informatics, MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen Key Laboratory of Spatial Information Smart Sensing and Services, Shenzhen University, Shenzhen, China 
700 1 |a Li, Qingquan  |u College of Civil and Transportation Engineering, Guangdong Key Laboratory of Urban Informatics, Shenzhen Key Laboratory of Spatial Smart Sensing and Services, MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area Guangdong Laboratory of Artificial Intelligence and Digital Economy, Shenzhen University, Shenzhen, China 
700 1 |a Zhou, Baoding  |u College of Civil and Transportation Engineering and the Institute of Urban Smart Transportation and Safety Maintenance, Shenzhen University, Shenzhen, China 
773 0 |t IEEE Internet of Things Journal  |g vol. 11, no. 11 (2024), p. 20706 
786 0 |d ProQuest  |t ABI/INFORM Trade & Industry 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3058293262/abstract/embedded/75I98GEZK8WCJMPQ?source=fedsrch