Dissertation PDF Print E-mail
Written by Nicholas Shorter   
Thursday, 13 August 2009 00:38

On Friday, May 22, 2009 Nicholas Shorter successfully defended his PhD dissertation entitled "Unsupervised Building Detection from Irregularly Spaced LiDAR and Aerial Imagery."  Nicholas proceeded to graduate August 8, 2009 with his doctorate of philosophy electrical engineering (PhD EE).  The individuals asking questions in the video  are Nicholas' dissertation committee Dr. Takis Kasparis (advisor), Dr. Georgios Anagnastopoulos, Dr. Michael Georgiopoulos, Dr. Lindwood Jones, Dr. Andy Lee, and Dr. Wasfy Mikhael.

 

Document Name Document Description File Size
Dissertation Manuscript Final Submitted Manuscript 8 MB (.pdf)
Dissertation Presentation Slides Power Point Presentation Slides 10 MB (.pdf)
Dissertation Defense - Part I Lecture Youtube
Dissertation Defense - Part II Lecture Youtube
Dissertation Defense - Part III Lecture Youtube
Dissertation Defense - Part IV Lecture Youtube
Dissertation Defense - Part V Committee Questions Youtube
Dissertation Defense - Part VI Committee Questions Youtube
Dissertation Defense - Part VII Committee Questions Youtube
Dissertation Defense - Part VIII Committee Questions Youtube

 

Nicholas Shorter

 

Dissertation Abstract:

As more data sources containing 3-D information are becoming available, an increased interest in 3-D imaging has emerged. Among these is the 3-D reconstruction of buildings and other man-made structures. A necessary preprocessing step is the detection and isolation of individual buildings that subsequently can be reconstructed in 3-D using various methodologies. Applications for both building detection and reconstruction have commercial use for urban planning, network planning for mobile communication (cell phone tower placement), spatial analysis of air pollution and noise nuisances, microclimate investigations, geographical information systems, security services and change detection from areas affected by natural disasters.  Building detection and reconstruction are also used in the military for automatic target recognition and in entertainment for virtual tourism.

 

Previously proposed building detection and reconstruction algorithms solely utilized aerial imagery.  With the advent of Light Detection and Ranging (LiDAR) systems providing elevation data, current algorithms explore using captured LiDAR data as an additional feasible source of information.  Additional sources of information can lead to automating techniques (alleviating their need for manual user intervention) as well as increasing their capabilities and accuracy.  Several building detection approaches surveyed in the open literature have fundamental weaknesses that hinder their use; such as requiring multiple data sets from different sensors, mandating certain operations to be carried out manually, and limited functionality to only being able to detect certain types of buildings.

 

In this work, a building detection system is proposed and implemented which strives to overcome the limitations seen in existing techniques. The developed framework is flexible in that it can perform building detection from just LiDAR data (first or last return), or just nadir, color aerial imagery.  If data from both LiDAR and aerial imagery are available, then the algorithm will use them both for improved accuracy. Additionally, the proposed approach does not employ severely limiting assumptions thus enabling the end user to apply the approach to a wider variety of different building types.  The proposed approach is extensively tested using real data sets and it is also compared with other existing techniques. Experimental results are presented.

 

Acknowledgements:

I would like thank my PhD Committee Dr. Takis Kasparis, Dr. Georgios Anagnostopoulos, Dr. Michael Georgiopoulos, Dr. Andy Lee, Dr. Linwood Jones and Dr. Wasfy Mikhael.  The valuable feedback the committee provided me with during my PhD Candidacy Exam, my dissertation defense and at several other critical stages in my doctoral research greatly enhanced the quality of my dissertation.  I thank Dr. Takis Kasparis for mentoring me for years on end during my graduate studies.  His experience and knowledge in Digital Image and Signal Processing was a great asset to me for my research.  I thank Dr. Georgiopoulos and Dr. Anagnostopoulos for their tutelage in unsupervised learning.  I thank Dr. Mikhael for his instruction on a plethora of Digital Signal Processing related concepts, and for also sharing with me practices for modeling and accounting for various phenomena when developing a proposed system.  I am thankful to Dr. Lee and Dr. Mahalanobis for providing me with insight on industrial perspectives on LiDAR and 3D Reconstruction.  I also thank Dr. Lee for informing me about various LiDAR programs and data sets.

 

I thank my immediate family, Nichole, Kathy and Sven Shorter for all of the motivation and encouragement they provided me with on a daily basis.  Their compassion was of great help to me during difficult times of struggle.  I thank my girlfriend Amanda McNally for her unconditional love she has blessed me with and her unending support.


I am very grateful to all of the researchers and companies whom have donated data for me to use for my dissertation, data valued at thousands of dollars.  I thank Dr. Simone Clode and Dr. Franz Rottensteiner for giving me access to the LIDAR data set known as 'the Fairfield data set’, which was collected and donated by AAMHatch.  I thank Mr. John Ellis with AeroMap for providing me with the LiDAR data set known as 'the Anchorage data set'.

 

I thank Dr. Andy Lee, Mr. Jay Hackett, and Mr. Jack Needham and Harris Corporation for their funding and for their feedback provided at numerous technical meetings at Harris.

Last Updated on Sunday, 02 October 2011 16:26