The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
ALTERNATIVE PROCEDURES FOR THE INCORPORATION OF LIDAR-DERIVED 
LINEAR AND AREAL FEATURES FOR PHOTOGRAMMETRIC GEO-REFERENCING 
A.F. Habib, M. Aldelgawy 3 
a Department of Geomatics Engineering, University of Calgary, Canada 
habib@geomatics.ucalgary.ca, mmaldelg@ucalgary.ca 
WG 1/2 WG 1/2 - SAR and LiDAR Systems 
KEY WORDS: Digital Photogrammetry, Digital Orthophoto, Aerial Photogrammetry, Adjustment, Land Use Mapping, 
LiDAR, Linear Features, Areal Features 
ABSTRACT: 
The positional and spectral information in LiDAR and photogrammetric data are optimal for providing a complete description of 3D 
environments. However, the synergistic attributes of the LiDAR and photogrammetric data can be only achieved after their proper 
registration to a common reference frame. This paper presents alternative methodologies for utilizing LiDAR-derived features for 
geo-referencing the photogrammetric data relative to the LiDAR reference frame. Since the LiDAR footprints are irregularly 
distributed, no point-to-point correspondence can be assumed between the photogrammetric and LiDAR data. In other words, it is 
almost impossible to identify distinct conjugate points in overlapping photogrammetric and LiDAR data. Consequently, LiDAR 
linear and areal features will be used as control information for the geo-referencing of the photogrammetric data. The paper will 
present three alternative methodologies to solve this task. The first approach outlines constraints that can be added to current bundle 
adjustment procedures to incorporate LiDAR linear and areal features. The second approach utilizes existing point-based bundle 
adjustment procedures for the incorporation of linear and areal features after manipulating the variance-covariance matrices 
associated with the points representing these features. Finally, the third approach will be based on weight restrictions imposed on the 
points representing the linear and areal features. After the introduction of the proposed methodologies, the paper will proceed by 
discussing experimental results using simulated datasets through a root mean square error analysis of a number of check points. 
1. INTRODUCTION 
Considering the characteristics of acquired spectral and spatial 
data from imaging and LiDAR systems, one can argue that their 
integration will be beneficial for accurate and complete 
description of the object space. It is evident that the 
disadvantages of one system can be compensated for by the 
advantages of the other system (Baltsavias, 1999; Satale and 
Kulkami, 2003). However, the synergic characteristics of both 
systems can be fully utilized only after ensuring that both 
datasets are geo-referenced relative to the same reference frame 
(Habib and Schenk, 1999). Traditionally, photogrammetric geo- 
referencing is either indirectly established with the help of 
ground control points (GCP) or directly defined using GPS/INS 
units on board the imaging platform (Cramer et al, 2000). On 
the other hand, LiDAR geo-referencing is directly established 
through the GPS/INS components of a LiDAR system. In this 
regard, this paper presents alternative methodologies for 
utilizing LiDAR features as a source of control for 
photogrammetric geo-referencing. There are various techniques 
dealing with linear and areal features in photogrammetry (Habib 
et al, 2007). Three approaches are presented in this research. 
The first one outlines a constraint that should be added to the 
current bundle adjustment procedures, while the second and 
third ones utilize the existing point-based bundle adjustment 
procedures for the incorporation of linear and areal features. 
The second approach manipulates the variance-covariance 
matrices associated with the points representing image and/or 
object space linear features, and the third one manipulates the 
weight matrices. 
The paper introduces the different approaches for incorporation 
of both linear and areal features for image geo-referencing as 
well as experimental results and analysis in the following 
sections. Section 2 discusses the incorporation of linear features. 
The coplanarity-based incorporation of linear features is 
discussed in sub-section 2.1. Sub-section 2.2 explains the point- 
based incorporation of linear features. The error ellipse 
expansion and the weight restriction methodologies are 
discussed in sub-sections 2.2.1 and 2.2.2 respectively. An 
illustration of the applications of the point-based approaches for 
linear features is shown in sub-section 2.2.3. Afterwards, the 
incorporation of areal features is illustrated in section 3. The 
coplanarity-based incorporation of planar patches is shown in 
sub-section 3.1. Sub-section 3.2 outlines the point-based 
approaches for incorporation of planar patches. The error ellipse 
expansion and the weight restriction approaches for planar 
patches are illustrated in sub-sections 3.2.1 and 3.2.2, 
respectively. In addition, experimental results and analysis for 
simulated dataset are discussed in section 4. Experiments for 
both single photo resection and bundle adjustment are shown in 
sub-sections 4.1 and 4.2, respectively. Finally, the conclusions 
and recommendations for future work are summarized in 
section 5. 
2. INCORPORATION OF LINEAR FEATURES FOR 
IMAGE GEO-REFERENCING 
This section presents the approaches used for incorporating 
linear features extracted from LiDAR data for the geo- 
referencing of photogrammetric data. The first approach is the 
coplanarity-based incorporation of linear features, while the 
second one is the point-based incorporation of linear features, 
where we can either expand the error ellipse or restrict the 
weight matrix. These approaches are provided in details in the 
219