Developing a meso-scale non-contact measuring method based on vision system : calibration of CCD camera

• Main Author: A.Rahman, Khairul Anuar
• Format: Thesis
• Language: English; English
• Published: 2014
• Subjects: T Technology (General); TJ Mechanical engineering and machinery
• Online Access: http://eprints.utem.edu.my/id/eprint/14905/; https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=92056

In developing a vision based measuring system, the camera’s precision has always been the bottleneck, and often being discussed. The combination of digital camera, narrow angle, relatively big distortions and focus to infinity cause some difficulties in camera calibration, as a result none of the existing camera calibration techniques is perfectly suitable for this purpose. This research compared three types of CCD camera calibration techniques namely Bouget’s Calibration Toolbox, Zhang’s Calibration Toolbox and Heikkilla’s Calibration Toolbox. The purpose is to select the most suitable camera calibration technique to fulfill the needs of users according to their desired applications. Aside from camera calibration, optimization of parameters such as effective focal length and coordinate of principle point for intrinsic parameter as well as extrinsic parameters comprises of rotation matrix and translation were performed. Experimental data for both calibration and optimization were collected to further explain the experimental results. Statistical analyses such as T-Test and ANOVA were conducted on the collected data using Minitab and EXCEL software. The results of this research indicated that the best calibration technique (toolbox) for calibrating Omron F500 CCD Camera for the purpose of measuring dimensions of meso-scale component is the Heikkilla’s Calibration Toolbox.