ACCURACY IMPROVEMENT OF A IMMS IN AN URBAN SCENARIO

Indoor Mobile Mapping Systems (IMMSs) technologies are becoming increasingly popular thanks to the possibility of acquiring a massive amount of 3D data in a fast and effective way in those areas where GNSS signal is unavailable, like urban canyons, densely vegetated areas, underground sites and buildings. They offer an efficient way to produce point clouds but with noticeably lower accuracy than the traditional Terrestrial Laser Scanning (TLS). The present paper wants to analyse two different methods to improve the accuracy of the point cloud coming from an IMMS survey in a vast urban scenario. The first approach uses points collected during an RTK GNSS survey as Ground Control Points (GCPs). The second one involves TLS static scans as Ground Control Scans (GCSs). Both these procedures allow us to introduce constraints before generating the rigid final point cloud. The tested IMMS is the Backpack Heron MS Twin Color, produced by Gexcel S.r.l. The instrument was tested during the acquisition of the historical centre of Meda (MB) in the northern part of Italy. Results show that in those areas between the constraints, the maximum error of residuals on checkpoints is some decimetres. The IMMS has allowed us to quickly survey a vast area not otherwise obtainable with traditional survey techniques. The developed procedures proved to be essential for proper reconstruction of the environment.