An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation
Blind and visually impaired people face different challenges when navigating indoors and outdoors. In this context, we suggest developing an obstacle detection system based on a modified YOLO v5 neural network architecture. The suggested system is capable of recognizing and locating a set of landmar...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | Ain Shams Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2090447923002769 |
| _version_ | 1797299706541375488 |
|---|---|
| author | Ahmed Ben Atitallah Yahia Said Mohamed Amin Ben Atitallah Mohammed Albekairi Khaled Kaaniche Sahbi Boubaker |
| author_facet | Ahmed Ben Atitallah Yahia Said Mohamed Amin Ben Atitallah Mohammed Albekairi Khaled Kaaniche Sahbi Boubaker |
| author_sort | Ahmed Ben Atitallah |
| collection | DOAJ |
| description | Blind and visually impaired people face different challenges when navigating indoors and outdoors. In this context, we suggest developing an obstacle detection system based on a modified YOLO v5 neural network architecture. The suggested system is capable of recognizing and locating a set of landmark indoor and outdoor objects that are extremely useful for Blind and Visually Impaired (BVI) navigation aids. Training and evaluation experiments were conducted using two datasets: the IODR dataset for indoor object detection and the MS COCO dataset for outdoor object detection. We used several optimization strategies, such as model width scaling, quantization, and channel pruning, to guarantee that the suggested work is implemented in embedded devices in a lightweight manner. The proposed system was successful in achieving results that were extremely competitive in terms of processing time as well as the precision of obstacle detection. |
| first_indexed | 2024-03-07T22:55:35Z |
| format | Article |
| id | doaj.art-b4c4399360a543ffa538bbd29a13c7c6 |
| institution | Directory Open Access Journal |
| issn | 2090-4479 |
| language | English |
| last_indexed | 2024-03-07T22:55:35Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ain Shams Engineering Journal |
| spelling | doaj.art-b4c4399360a543ffa538bbd29a13c7c62024-02-23T04:59:31ZengElsevierAin Shams Engineering Journal2090-44792024-02-01152102387An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigationAhmed Ben Atitallah0Yahia Said1Mohamed Amin Ben Atitallah2Mohammed Albekairi3Khaled Kaaniche4Sahbi Boubaker5Department of Electrical Engineering, College of Engineering, Jouf University, Sakaka, Saudi Arabia; Corresponding author.Remote Sensing Unit, College of Engineering, Northern Border University, Arar, Saudi Arabia; Laboratory of Electronics and Microelectronics (LR99ES30), University of Monastir, TunisiaLaboratory of informatics, Gaspard-Monge, A3SI, ESIEE Paris, CNRS, Gustave Eiffel University, France; LETI, ENIS, University of Sfax, Sfax, TunisiaDepartment of Electrical Engineering, College of Engineering, Jouf University, Sakaka, Saudi ArabiaDepartment of Electrical Engineering, College of Engineering, Jouf University, Sakaka, Saudi ArabiaCollege of Computer Science and Engineering, University of Jeddah, Jeddah, Saudi ArabiaBlind and visually impaired people face different challenges when navigating indoors and outdoors. In this context, we suggest developing an obstacle detection system based on a modified YOLO v5 neural network architecture. The suggested system is capable of recognizing and locating a set of landmark indoor and outdoor objects that are extremely useful for Blind and Visually Impaired (BVI) navigation aids. Training and evaluation experiments were conducted using two datasets: the IODR dataset for indoor object detection and the MS COCO dataset for outdoor object detection. We used several optimization strategies, such as model width scaling, quantization, and channel pruning, to guarantee that the suggested work is implemented in embedded devices in a lightweight manner. The proposed system was successful in achieving results that were extremely competitive in terms of processing time as well as the precision of obstacle detection.http://www.sciencedirect.com/science/article/pii/S2090447923002769Obstacle detectionNavigation AssistanceVisually ImpairedDeep LearningLightweight implementation |
| spellingShingle | Ahmed Ben Atitallah Yahia Said Mohamed Amin Ben Atitallah Mohammed Albekairi Khaled Kaaniche Sahbi Boubaker An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation Ain Shams Engineering Journal Obstacle detection Navigation Assistance Visually Impaired Deep Learning Lightweight implementation |
| title | An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| title_full | An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| title_fullStr | An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| title_full_unstemmed | An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| title_short | An effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| title_sort | effective obstacle detection system using deep learning advantages to aid blind and visually impaired navigation |
| topic | Obstacle detection Navigation Assistance Visually Impaired Deep Learning Lightweight implementation |
| url | http://www.sciencedirect.com/science/article/pii/S2090447923002769 |
| work_keys_str_mv | AT ahmedbenatitallah aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT yahiasaid aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT mohamedaminbenatitallah aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT mohammedalbekairi aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT khaledkaaniche aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT sahbiboubaker aneffectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT ahmedbenatitallah effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT yahiasaid effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT mohamedaminbenatitallah effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT mohammedalbekairi effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT khaledkaaniche effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation AT sahbiboubaker effectiveobstacledetectionsystemusingdeeplearningadvantagestoaidblindandvisuallyimpairednavigation |