Enhancing single camera calibration results using artificial bee colony optimisation within a virtual environment

Khanesar, Mojtaba A.; Todhunter, Luke; Pawar, Vijay; Corcoran, Hannah; MacDonald, Lindsay; Robson, Stuart; Piano, Samanta

Abstract

Close range photogrammetry is among the top candidates for non-contact metrology in high precision applications. It is frequently used within industrial environments for high precision measurement, automation, and control tasks. When using off-the-shelf cameras for such applications it is necessary first to understand how image content influences the image measurements made and in turn what effects this has on estimating imaging geometry. A virtual environment involving camera and digital objects may be used for testing the efficacy of machine learning algorithms. In this paper, enhancement of a single pose camera calibration process utilising a virtual environment and images taken with different lighting directions is investigated. The algorithm used for enhancing the calibration process is Artificial Bee Colony (ABC), a metaheuristic optimisation method. Multiple single image camera orientations are tested in this paper resulting in different extrinsic camera parameters. From multiple tests using different camera orientations, we observe that it is possible to enhance the calibration efficacy in terms of reprojection error using artificial bee colony when compared to an established two step Levenberg Marquardt (LM).

Citation

Khanesar, M. A., Todhunter, L., Pawar, V., Corcoran, H., MacDonald, L., Robson, S., & Piano, S. (2024, June). Enhancing single camera calibration results using artificial bee colony optimisation within a virtual environment. Presented at euspen’s 24th International Conference & Exhibition, Dublin, IE, June 2024, Dublin, Ireland

• Presentation Conference Type: Edited Proceedings
• Conference Name: euspen’s 24th International Conference & Exhibition, Dublin, IE, June 2024
• Start Date: Jun 10, 2024
• End Date: Jun 14, 2024
• Acceptance Date: Mar 11, 2024
• Publication Date: Jun 14, 2024
• Deposit Date: Apr 25, 2024
• Publicly Available Date: Apr 25, 2024
• Peer Reviewed: Peer Reviewed
• Public URL: https://nottingham-repository.worktribe.com/output/34105187
• Related Public URLs: https://www.euspen.eu/events/24th-international-conference-exhibition-10th-14th-june-2024/?subid=24th-international-conference-exhibition-10th-14th-june-2024