Data processing for real-time construction site spatial modeling

The ability to quickly model work spaces using high frequency 3D imaging sensors has great potential for improving construction site resource management. Yet, the rapid processing of tens of thousands of range points, which is a crucial component of the spatial modeling process, is still an unsolved problem requiring further investigation. This paper describes a testbed that was developed to study the performance of various algorithms for processing range point data captured using 3D imaging sensors. Results of applying different combinations of data filtering, transformation, and segmentation techniques are also presented. Some of the algorithms investigated proved to be robust to sensor noise and able to accurately and rapidly process high frequency range data.