Smart Electronic Assistive Device for Visually Impaired Individual through Image Processing
Ira Flores, Gary Chris Lacdang, Chessa Undangan, Jetron J. Adtoon, Noel B. Linsangan · 2021 IEEE 13th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) · 2021
According to World Health Organization, it is assessed that around 1.3 billion individuals live with some distance or near vision impairment. People with visual impairment encounter different issues in their everyday life as advanced assistive devices are frequently not meeting the consumer prerequisites in terms of price and level of assistance. The paper’s primary goal is to build a smart electronic assistive device specifically for visually impaired individuals. The study consists of two devices, which are the glasses and the smart cane. The glasses are responsible for text detection and recognition, which falls under image processing. At the same time, the smart cane is the one responsible for obstacle detection using combination sensors known as VL53L0X and Ultrasonic. This study found that the smart cane and smart glass functions well as intended, in alerting users through a voice alert about the obstacles around and navigation and reading the text captured by the camera. The smart cane’s obstacle detection accuracy is 100%, while the smart glasses’ overall accuracy is 98.13% for document text and 91.33% for natural scene text.