ECSE Departmental Research Seminar, 15th February 2006

Abstract:

Automated generation of 3D building models from directly observed point clouds generated by LIDAR sensors is presented. The equipment used in the research is a Riegl LMS-Z420i Terrestrial Laser Scanner equipped with a calibrated Nikon D100 6 Mega Pixel digital camera with 14mm lens mounted on top of the scanner. Using RANSAC segmentation, the LIDAR points on the terrain are separated from points on buildings and other object classes, and a digital terrain model (DTM) can be computed. Points on buildings have to be separated from other points classified as off-terrain points, which is accomplished by an analysis of the height differences of a digital surface model passing through the original LIDAR points and a digital terrain model. Thus, a building mask is derived, and planar patches are created in these candidate regions by applying RANSAC and other segmentation methods. The vegetations are to be further simplified and included in the final urban models. Occlusion problems which occurred in the calibrated images can be solved using minimum number of images.

About the speaker:

Ee Hui has completed a BE (Hons) in Electrical and Computer System Engineering at Monash University in 2004 and has been working on a MEngSc by Research degree at Monash University from March 2005. Working under the supervision of A/Prof David Suter and Prof Ray Jarvis, she is in the group of Monash University’s Intelligent Robotics Research Centre (IRRC) and Institute for Vision System Engineering. Her research interest is in 3D model fitting, image processing and region growing techniques.