Smart Wearable System for the Blind: Audio Vision and Localization

Abstract

With the rapid advancement of artificial intelligence and embedded systems technologies, it has become possible to develop intelligent technological solutions that enhance the independence of visually impaired individuals and improve their ability to perceive the surrounding environment. This study aims to design and develop a wearable intelligent system to assist visually impaired users based on computer vision techniques. The system converts visual information into understandable audio notifications while also providing location tracking and emergency alert services.

The proposed system relies on the advanced NVIDIA Jetson Nano processing platform as the main processing unit to execute object detection algorithms and perform real-time image analysis. Images of the surrounding environment are captured using an OV7670 camera module, where computer vision algorithms analyze the scene and identify various objects such as people, obstacles, and common daily items. The detected objects are then converted into voice messages delivered to the user through earphones, enabling visually impaired individuals to understand their surroundings through immediate auditory descriptions of detected objects.

To manage control functions and integration between different system components, an Arduino Uno microcontroller is used to regulate communication among system modules and control various operational buttons. In addition, a NEO-6M GPS module is integrated to determine the user's geographic coordinates, while a SIM800L GSM module is used to send short text messages to predefined phone numbers.

The system also includes an emergency button that allows the user to send a text message containing their geographic location to family members or caregivers. The location can be directly opened through smartphone map applications, enabling rapid identification of the user's position. Furthermore, the system includes a power control button to turn the system on or off, which simplifies operation and helps reduce energy consumption.

Preliminary results demonstrate that the proposed system can successfully recognize several common objects in the surrounding environment and convert them into clear audio alerts. In addition, the system efficiently provides location tracking and emergency notification services, contributing to improved safety and independence for visually impaired individuals during navigation in both indoor and outdoor environments.