Robotic System for Tracking Moving Objects Based On Their Color

Abstract

Tracking moving objects is one of a very important applications within the computer vision. The goal of object tracking is segmenting a region of interest from a video scene and keeping track of its motion and positioning. Track moving objects used in many applications such as video surveillance, robot vision, and traffic monitoring, and animation. In this paper a four-wheeled robotic system have been designed and implemented by using Arduino-Uno microcontroller. Also a useful algorithms have been developed to detect and track objects in real time. The main proposed algorithm based on the kernel based color algorithm and some geometric properties to tracking color object. Robotic system is a compromise of two principal parts which are the hardware and software parts. Hardware includes Bluetooth model to connect the phone with Arduino-Uno. Robot body consists the L298 dual H-bridge motor driver to drive four geared motor, two battery as a power supply and two servomotors to move the camera in both horizontal and vertical axis. Software is responsible for making the right decision based on the analysis of data that receives from the digital camera. Color-based tracking algorithm and border following algorithm used to detect the location of the target object in the images have been showed in the paper. All computations are accomplished within android device. Through applying the object tracking method, several parameters have been considered like frame rate, motor period time and speed of target object. All experiments were in the real environment. The proposed robotic system succeeded to track the target object with a success rate up to 97% in indoor environment.