Threat Detection for Visually Impaired Persons
Gaurav Dahake, Sarth Khandekar, Tanushree Darwalkar, Tanaya Bure, Sakshi Rawale, Mrunal Masurkar · Zenodo (CERN European Organization for Nuclear Research) · 2023
One of the most important senses for living is vision. Many people living in this world deal with visual impairment. These people face many difficulties in navigating independently and safely, facing issues in getting information and communication. The objective of the proposed work is to change the visual world into an audio world by notifying blind people about the objects in their path. This will help visually impaired person to navigate independently without any external assistance only by using this real-time object detection system. The app uses image processing and machine learning techniques to identify objects in real time via camera and inform visually impaired people about the object and its location via audio output. The inability to differentiate between products leads to many limitations of current methods, including lower accuracy and poor results. The main objective of this proposed work is to provide effective accuracy, the best performance, and a viable option for visually impaired person to make the world a better place for them.