CNN based Smart Cane: A tool for Visually Impaired People

This paper describes a study that assists visually impaired people in walking more confidently. People who are visually impaired constitute a sizable population and can be found all over the world. Technology has recently shown its being present in all fields, and cutting-edge gadgets to lend a helping hand to people in their day to day lives. This paper develops a clever, intelligent system for visually impaired people (VIPs) to aid them in moving and ensure their safety. To provide real-time navigation, the proposed method employs an automated voice. Even when they cannot see the objects around them, VIPs can sense and comprehend their surroundings. A web-based application is also built to ensure their security. The on-demand feature of this software lets the user to compromise privacy while sharing his or her location with loved ones. VIPs and family members could use this app to track their whereabouts and take pictures of them while they were at their homes. As a result, the device lets VIPs to monitor their surroundings and ensure their safety. The available literature lacked such a comprehensive device. Mobile Net architecture is used as the program must run on minimal-power end devices. For object detection and recognition, a deep convolutional neural network (CNN) model is used.