Object Pose Estimation using Least Non Coplanar Feature Points
Rohit Prasad1, Tejaswini Kar2
1Rohit Prasad, School of Electronics Engineering, Kalinga Institute of Industrial Technology, KIIT University, Bhubaneswar, India.
2Tejaswini Kar, School of Electronics Engineering, Kalinga Institute of Industrial Technology, KIIT University, Bhubaneswar, India.
Manuscript received on November 09, 2014. | Revised Manuscript Received on November 18, 2014. | Manuscript published on November 20, 2014. | PP: 13-16 | Volume-2, Issue-12, November 2014. | Retrieval Number: : L05421121214/2014©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper describes the pose estimation of an object using a calibrated camera. The idea is to first calibrate the camera and then implement the algorithm to find the estimated matrices that describes the three dimensional pose of the object. The camera calibration process includes capturing the images and then processing them to find the intrinsic and extrinsic parameters, which are used to estimate the object pose. And object pose estimation is carried out by first finding the corners with the help of Harris feature extraction and then comparing the image and object matrices in the POSIT algorithm and finally eliminating the errors with the help of iterations. The algorithm estimates the pose with a minimum of four non-coplanar points from the acquired image. Both camera calibration and pose estimation processes were implemented using MATLAB® Ver.18.104.22.1685 (R2011a).
Keywords: POSIT, pose estimation, camera calibration, intrinsic parameters, non-coplanar feature points.