A Survey on Deep Learning-based Smart Assistive Aids for Visually Impaired Individuals

作者：Aetesam Ali Khan Ashar, Ajwad Abrar, Jiangjiang Liu · 年份：2023 · DOI：10.1145/3603765.3603775 · 被引用次数：5 · 研究领域：Tactile and Sensory Interactions、Smart Parking Systems Research、Video Surveillance and Tracking Methods

Visually impaired individuals face significant challenges in interacting with society. However, smart canes equipped with sensors and cameras provide a promising solution for safe navigation. AI-based assistive devices, including smart canes, utilize machine learning algorithms like YOLO and SSD for object detection and recognition through in-built cameras. These devices often employ a standard controller and switch for navigation strategy, while sensors such as Ultrasonic sensors, I/R Lasers, Rain sensors aid in detecting a variety of obstacles such as physical objects or water. With continued research, such technology could greatly enhance the visually impaired person's ability to interact with their surroundings and potentially even travel independently.