Evaluation of a Bisensor System for 3D Modeling of Indoor Environments

Abstract: Three dimensional (3D) sensing, analysis and reconstruction of objects and scenesin unknown environments has become an important area of research in several disciplines.Recently, Time-of-Flight (ToF) cameras have attracted attention because oftheir advantageous ability to generate full 3D range distance information at videoframe rates. However, several challenging fluctuations like systematic or random errorscan occur and thus decrease the quality of the acquired data. Therefore, the usabilityof such systems is very much dependent on an accurate camera calibration. Inaddition, ToF cameras have still limited resolution (e.g. 176x144) and they can onlyprovide monochromatic (grayscale) information.This work concentrates on the fusion of 2D vision and 3D range data in order toobtain accurate and visually appealing representations of real indoor environments. Inparticular two major fields of research are treated. The first concerns the compensationof the common errors that the ToF cameras present. This is achieved by performing acalibration schema that additionally sets the ground towards the creation of a commonreference frame for both sensors. The second task is to incorporate these two differentsensors into one system that exhibits improved accuracy and high quality output data.The calibration process addresses both photogrammetric and depth correction techniques.Moreover, due to the fact that the real world scenes increase the level of difficultythat such a system has to confront, a validation of the ToF cameras performanceemerges as an important need. For this purpose, a novel approach to evaluate and get abetter insight on the accuracy of the ToF camera is introduced, based on a 3D NormalDistribution Transform.Finally, as sensor fusion plays an important role in this work, multi-sensor applicationsare further investigated based on the incorporation of such a biform sensorysystem. Principle goal is to communicate the color information and enhance the overallresult by the use of 3D surface reconstruction approaches.