Abstract—Camera calibration in an outdoor environment often relies on natural landmarks, and algorithms that are computational expensive. This paper presents a fast camera calibration method that uses robots themselves as landmarks to each other. The active landmark made of LEDs, mounted on the robots, can be turned on/off through available wireless link. Experimental measurements and validation were conducted using a low-resolution, low-computation power smart camera, the CMUcam3 camera. The experimental results show that the estimates of the intrinsic camera parameters converge to the correct values as the robot moves and more measurements are used.
Index Terms—Computer vision, robot sensing, data fusion, sensory networks.
The authors are with the Department of Engineering, Indiana University – Purdue University, Fort Wayne, IN 46809 USA (e-mail: liu@engr.ipfw.edu, raez@ipfw.edu).
[PDF]
Cite: Yanfei Liu and Carlos Pomalaza-Ráez, "Application of Active Self-landmarking to Camera Calibration,"
International Journal of Computer Theory and Engineering vol. 5, no. 1, pp. 60-64, 2013.
