A Comprehensive Review of Smart Glove Technologies for Enhancing Visual Impairment Assistance

The present work has proved to be the design, development, and evaluation activities of a single model of smart glove for the people with problems ranging from severe to complete vision loss. The glove usually integrates a lightweight camera, an inertial measurement unit (IMU), and a microcontroller that communicates with networks for real-time object recognition and auditory feedback. This is achieved through advanced Convolutional Neural Networks (CNNs) sometimes referred to as YOLO (You Only Look Once). As per the testing, which was carried out in both controlled and real-world environments, the glove was able to achieve an average object recognition accuracy of 95% under optimum condition and 90% under practical situations with an average response time of 1.2 seconds. Most user comments highlighted how the glove provided more autonomy and navigation while the criticisms rested mainly on areas that could be improved such as ergonomic design and battery lifetime. It presents a great opportunity for iterative improvement and user-centered design as this is very important and marks quite a significant advance in making assistive devices efficient and user-friendly. In so doing, it accords with SDG goals that relate to health and reduce inequality, as the inputs indicate the effectiveness of the gloves in improving the quality of life of visually impaired individuals.