INTEGRATION OF KNOWLEDGE INTO THE DETECTION OF OBJECTS IN POINT CLOUDS

The reconstruction of 3D objects from point clouds and images is a major task in many application fields. The processing of such spatial data, especially 3D point clouds from terrestrial laser scanners, generally consumes time and requires extensive interaction between a human and the machine to yield a promising result. Presently, algorithms for an automatic processing are usually datadriven and concentrate on geometric feature extraction. Robust and quick methods for complete object extraction or identification are still an ongoing research topic and suffer from the complex structure of the data, which cannot be sufficiently modelled by purely numerical strategies. Therefore, the intention of our approach is to take human cognitive strategy as an example, and to simulate these processes based on available knowledge for the objects of interest. Such processes will first, introduce a semantic structure for the objects and second, guide the algorithms used to detect and recognize objects, which will yield a higher effectiveness. Hence, our research proposes an approach using knowledge to guide the algorithms in 3D point cloud and image processing.