Visual Assistance Model: To Enhance Accessibility and Mobility for Visually Impaired People

Visual impairment is a major challenge for millions of people worldwide, limiting their ability to perceive and interact with their surroundings. To address this issue, a Visual Assistance Model has been developed. This model leverages cutting-edge technology to provide real-time visual support, enhancing the mobility of visually impaired individuals. Our contribution to this is a model based on speech recognition. The primary control chip collaborates with multiple functional modules to perform traffic signal recognition, obstacle avoidance, navigation capabilities, and sending alert messages to caregivers. For traffic signal recognition, we used YOLO version 8 which achieved an accuracy of 90%, compared to other research that only used simple YOLO and CNN models. We also reduced reliance on smart phones by using WiFi and GSM (Global System for Mobile Communication) mod-ules for data transmission, instead of BLE.