Accuracy analysis of a mobile mapping system for close range photogrammetric projects

Image-based mapping solutions require accurate exterior orientation parameters independently of the cameras used for a survey. This paper analyses the inclusion of up to two stereo-based geometric constraints in the form of baseline distance and convergence angle between camera axes to boost the integrated sensor orientation performance on outdoor close-range projects. A terrestrial low-cost mobile mapping GNSS/IMU multi-camera system is used to test the performance of the stereo-based geometric constraint on a weak geometric network in a stop-and-go survey. The influence of the number of control points (CPs) is analysed to confirm the performance and usability of the geometric constraints in real live terrestrial projects where far from ideal setups can exist across the survey. Improvements in image residuals up to 9 times and deviation errors better than 1 cm are expected when at least three CPs are incorporated into the adjustment.