The Open Automation and Control Systems Journal

2014, 6 : 1498-1504
Published online 2014 December 31. DOI: 10.2174/1874444301406011498
Publisher ID: TOAUTOCJ-6-1498

Estimation on Three-dimensional Motion Parameters in Binocular Vision of Robot Tracking and Control System

Jiang Jianju
Shen Zhen Polytechnic, Shenzhen Guangdong 518055, China.

ABSTRACT

Motion is an important part of visual research. There are lots of motion vision problems in industrial process control, vehicle navigation, flying target tracking and traffic monitoring. Combing three-dimensional motion of computer vision to position and track spatial position of moving objects for the achieved sequential image is a great challenge for computer vision, which is worthy of deep research. The paper analyzes the determination of motion parameter of objects from two-dimensional feature and three-dimensional feature. And three-dimensional motion and structure information of motion objects is achieved by using binocular vision to timely track motion state and image sequential process of objects. And the motion information of objects is sent to the robot, and the robot uses the achieved data to track objects, which realizes tracking.

Keywords:

Camera calibration, SIFT feature matching, Camshift, Three-dimensional motion analysis.