Calibration of multi-camera photogrammetric systems
- 1Department of Geomatics Engineering, University of Calgary, 2500 University Drive NW, Calgary, Alberta, T2N 1N4, Canada
- 2Lyles School of Civil Engineering, Purdue University, 550 Stadium Mall Drive, West Lafayette, Indiana, 47907-2051, USA
Keywords: digital close range photogrammetry, geometric camera calibration, bundle adjustment, multisensor systems
Abstract. Due to the low-cost and off-the-shelf availability of consumer grade cameras, multi-camera photogrammetric systems have become a popular means for 3D reconstruction. These systems can be used in a variety of applications such as infrastructure monitoring, cultural heritage documentation, biomedicine, mobile mapping, as-built architectural surveys, etc. In order to ensure that the required precision is met, a system calibration must be performed prior to the data collection campaign. This system calibration should be performed as efficiently as possible, because it may need to be completed many times. Multi-camera system calibration involves the estimation of the interior orientation parameters of each involved camera and the estimation of the relative orientation parameters among the cameras. This paper first reviews a method for multi-camera system calibration with built-in relative orientation constraints. A system stability analysis algorithm is then presented which can be used to assess different system calibration outcomes. The paper explores the required calibration configuration for a specific system in two situations: major calibration (when both the interior orientation parameters and relative orientation parameters are estimated), and minor calibration (when the interior orientation parameters are known a-priori and only the relative orientation parameters are estimated). In both situations, system calibration results are compared using the system stability analysis methodology.