Quality assurance of multi-sensor systems
Date
2011
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The demand for fast and cost-effective geo-spatial data collection along with technological advances in the last few decades have triggered considerable changes in the mapping survey practices. Currently, modern mapping systems consist of multi-sensor systems, typically encompassing navigation sensors and two types of imaging sensors: digital camera (one or multiple) and a laser scanning system. In order to fully attain the potential accuracy of the system sensors and guarantee accurate multi-sensor integration, careful system Quality Assurance (QA) should be carried out. Although several activities are involved in the QA of a multi-sensor system, the system calibration is the crucial activity to ensure the attainment of the expected accuracy and the most complex activity as well. System calibration involves individual sensor calibration and the mounting parameters calibration (i.e., lever arm offset and boresight angles) relating the system components such as the Global Positioning System (GPS), the Inertial Navigation System (INS), and the imaging sensors. In this research work, new calibration procedures for photogramrnetric and Light Detection and Ranging (LiDAR) systems are introduced. The proposed procedures are based on similar point positioning equations and the system parameters are estimated while minimizing flight and control requirements. More specifically, a rigorous analysis, based on the mathematical analysis of the GPS/INSĀassisted photogrammetric/LiDAR point positioning equation, is carried out to devise the minimum control and flight configuration requirements for the calibration of airborne single-camera photogrammetric and LiDAR systems. The proposed photogrammetric system calibration is a single-step procedure, which can be used for airborne/terrestrial directly geo-referenced single or multi-camera systems. The proposed procedure also has the ability to estimate the Relative Orientation Parameters (ROP) among the cameras in the absence of GPS/INS data. Furthermore, a general model, which allows for the incorporation of prior information about the ROP among the cameras in the calibration process, is devised. From such general model, the previous models ( calibration without prior ROP information and the estimation of the ROP among the cameras) can be derived as special cases. The proposed LiDAR system calibration is an automated procedure and doesn't require specific features (e.g., planes or lines) in the covered area. Suitable primitives, which do not involve pre-processing of the data, are implemented. The correspondence between conjugate primitives is determined using a robust matching procedure. A modification to the Gauss Markov model is introduced to keep the implementation of the calibration procedure simple while utilizing higher order primitives. Experimental results using simulated and real datasets have demonstrated the feasibility/effectiveness of the proposed methodologies for the calibration of photogrammetric and LiDAR systems.
Description
Bibliography: p. 253-262
Some pages are in colour.
Some pages are in colour.
Keywords
Citation
Kersting, A. P. (2011). Quality assurance of multi-sensor systems (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/4531