Electronic Device Designed for Object Recognize and Navigation Assistant by Artificial Vision for People with Visual Impairment

Abstract

In this project, a prototype of object recognition and mobility aid is developed for people with visual disabilities using artificial vision, stereo vision, and neural networks. This prototype has two operation modes, the first generates obstacle-free routes using mobile robotics navigation algorithms and the second recognizes the objects that are around the user and indicates what type of objects there are and where they are placed in the work environment, this information is delivered to the user through spanish audio messages. The system design was developed in the Python programming language in a single board computer Raspberry pi 4B. The prototype has two web cameras that were used to acquire the visual environment.