Tracking Objects using Artificial Neural Networks and Wireless Connection for Robotics

摘要

Tracking objects are used in many areas, and one of them is robotics. The goal in this work focuses on a robot that can follow an object that is in front of it. This application has two links: wireless and Bluetooth. The first one connects a mobile phone mounted on a robot for image acquisition and a personal computer (PC), and the second links a PC and a mobile robot to control the motors by open source, Arduino Board. The algorithm uses several patterns for training the Artificial Neural Network (ANN) and for object identification. Then, it is complemented by the extraction feature in Hue Saturation Value (HSV) color space. This algorithm uses C ++ language with OpenCV libraries for computer vision.