Raspberry PI based Food Recognition for Visually Impaired using YOLO Algorithm

Abstract

A person is considered visually impaired if the person has partial vision or zero visibility at all. Visual impairment is one of the common problems that affect the quality of a person's life, we rely on our eyes on a day-to-day basis. This study presents a wearable Raspberry Pi-based food recognition using the You Only Look Once (YOLO) algorithm. A Raspberry Pi was programmed to detect 25 kinds of Filipino foods. The hardware used is composed of Raspberry Pi, a Camera Module, Earphones, and an external battery. The device was able to recognize 25 different common Filipino foods with high accuracy and it was able to calculate the food's distance from the user with a different margin of as low as 0.1 to 0.9 inches.