Machine Learning-based Smart Assistance System for the Visually Impaired

Individuals with visual impairments face significant challenges in their daily lives, including mobility issues and difficulties recognizing people and differentiating between places. As a result, there is a pressing need for an intelligent device that can assist these individuals and help alleviate some of the difficulties they encounter. The proposed solution aims to address three of the primary challenges faced by visually impaired individuals: recognizing people, familiarizing themselves with indoor spaces, and receiving notifications in case of danger. This research work seeks to assist individuals who are blind in identifying others using a camera-on-glasses that can perform facial recognition and detect and identify faces. Additionally, the system can provide audio feedback through headphones to inform the user of the identity of individuals in their vicinity. Moreover, the system features a unique indoor navigation feature that utilizes radio frequency and machine learning technology to inform the user of their current location within a building. The system can also learn from its environment to detect emergency situations and alert a loved one in case the visually impaired individual is in danger.