Object Tracking by Color and Active Contour Models Segmentation

Abstract

This work presents a novel approach for an object recognition and tracking system, here known as SRCCA (Software de Reconhecimento de Cor e Contorno Ativo or Active Contour and Color Recognizing Software) by using color recognition in the HSV color space and active contour models (ACM). The main innovations proposed makes use of the association of distinct techniques employed for object tracking and detection, as well as the elaboration of an open source software through the use of C++ programming language and OpenCV functions. To enhance the tracking efficiency for targets in motion, the video capturing device was controlled by two servomotors, enabling the target to be followed, in case it moves in the x- or y-axis. The final system was tested and presented robust and efficient tracking and contour segmentation capabilities for targets in motion, regardless of shape and size.