An Intelligent Navigation System for the Visually Impaired Based on Machine Vision

In recent years, the number of visually impaired individuals has increased, and the urbanization process and lack of portable assistive tools have increased the risks associated with their mobility. This paper introduces the traditional blind cane as a carrier and integrates Internet of Things (IoT) and artificial intelligence (AI) technologies by installing sensor modules such as Bluetooth, cameras, ultrasonic obstacle avoidance, and voice broadcasting on the cane. Based on cloud service platforms and the YOLOv5 algorithm, an intelligent navigation cane has been developed that can achieve intelligent obstacle avoidance, traffic and road condition recognition, and other functions. Furthermore, an accessible navigation app has been developed that integrates various public transportation resources and enables Bluetooth connection, one-click cane retrieval, and travel navigation. This forms a complete smart navigation system for visually impaired individuals, providing them with security, independence, and long-distance travel support.