Publications PDF Print E-mail
Written by Nicholas Shorter   
Tuesday, 25 March 2008 06:10
Nicholas Shorter, for his master’s thesis in electrical engineering, successfully demonstrated Fuzzy Simplified Adaptive Resonance Theory’s (FSART) ability to automatically detect planar segments in the triangulation of a raw LiDAR point cloud.  Some of the significant contributions of the thesis were that the FSART clustering in conjunction with a post processing planar regression technique, realized 3D reconstruction of isolated buildings from the raw (not interpolated) LiDAR point cloud.  While in comparison several other techniques existent in the literature work on the interpolated point cloud (interpolation errors) and may not be completely automated, relying on the manual input of various parameters.  The master’s thesis work resulted in a conference (ISTASC 2007) and journal publication (WSEAS).  

For Nicholas’ PhD dissertation research, he has continued working on attempting to automatically detect and reconstruct buildings with the raw LiDAR point cloud.  In March of 2008, Nicholas published another conference paper (ISCCSP 2008) showcasing his work with the automatic detection of buildings from the triangulation of the raw LiDAR point cloud using hierarchical, triangulated connected sets.  The work showcased in that conference paper is significant as there currently exist few techniques in the literature working with the raw LiDAR point cloud for automatic building detection, whereas most are using the interpolated data and several require input parameters – such as the largest building size in the data set (for morphological filtering techniques).  Then in June of 2008, Nicholas published a conference paper (CITSA 2008) comparing the ability of Gaussian ART, Fuzzy ART and Fuzzy SART’s ability to automatically detect planar segments (building roof tops) within the triangulation of the raw LiDAR point cloud.  In July of 2008 Nicholas published a conference paper (IGARSS 2008) showcasing his research on automatically registering a LiDAR range image to an aerial image.  While the LiDAR range image exists as interpolated LiDAR data, a scheme was proposed for relating that interpolated data back to the raw data so that after the registration the aerial image pixels could then be related to the raw LiDAR point cloud data points.  Nicholas has recently published a conference paper (S+SSPR 2008) regarding using Fuzzy ART for image color quantization.  Nicholas has implemented a version of the JSEG color image segmentation algorithm, which uses the Fuzzy ART color quantization algorithm as a preprocessing technique, to work with high resolution aerial images and is currently working using the JSEG algorithm along with other techniques to automatically detect buildings from a single, ortho-rectified aerial image.

 

Written Publications

No. Journals
1 Shorter, N.; Kasparis, T. Automatic Vegetation Identification and Building Detection from a Single Nadir Aerial Image. Remote Sens. 2009, 1, 731-757.
2 Shorter, N. S.; Kasparis, T. Fuzzy SART Clustering for 3D Reconstruction from Irregular LIDAR Data. WSEAS Transactions on Signal Processing2006, 2, No. 8, 1122 to 1129.

 

No. Conferences
1 Shorter, N.S; Kasparis, T. (2008) Fuzzy ART for Relatively Fast Unsupervised Image Color Quantization. Proceedings for the Joint IAPR International Workshops on Structural and Syntactic Pattern Recognition (SSPR 2008) and Statistical Techniques in Pattern Recognition (SPR 2008), Orlando, Florida, December 4-6 (147 - 156)
2 Shorter, N.S.; Kasparis, T. (2008). Autonomous Registration of LiDAR Data to Single Aerial Image. IEEE International Geoscience & Remote Sensing Symposium, Boston, Massachusetts, July 6 - 11 (published)
3 Shorter, N.S.; Kasparis, T. (2008). Clustering Irregular Spaced LiDAR TINs for 3D Reconstruction. The 5th International Conference On Cybernetics and Information Technologies, Systems and Applications, Orlando, Florida, June 29 - July 2 (published)
4 Shorter, N. S.; Kasparis, T. (2008). Triangulated, Connected Sets for Building Detection from Irregularly Spaced LiDAR. The 3rd International Symposium on Communications, Control and Signal Processing, St. Julians, Malta, March 12 – 14 (published)
5 Shorter, N. S.; Kasparis, T. (2006). 3D Reconstruction of Irregular Spaced LIDAR, Proceedings of the 6th WSEAS International Conference on Systems Theory and Scientific Computation, Elounda, Greece, August 21-23 (pp. 19-24).

 

No. PhD Dissertation & Master's Thesis
1 Shorter, N. S. (2009). Unsupervised Building Detection From Irregularly Spaced LiDAR and Aerial Imagery, PhD Dissertation, University of Central Florida, Orlando, Florida, 2009.
2 Shorter, N. S. (2006). Heuristic 3D Reconstruction of Irregular Spaced LIDAR, Masters Thesis, University of Central Florida, Orlando Florida, 2006.

 

Presentations & Lectures


No. Presentations
1 Florida Institute of Technology Lecture presented on 10/30/09:  Pseudo Homotopy Trees for Unsupervised Building Detection in Irregularly Spaced LiDAR
2 Fuzzy ART for Relatively Fast Unsupervised Image Color Quantization presented  at S+SSPR 2008 in Orlando Florida at the University of Central Florida on December 5
3 1st Cyprus Workshop on Signal Processing and Informatics in Nicosia, Cyprus on July 8
4 Autonomous Registration of LiDAR Data to Single Aerial Image presented at IGARSS 2008 in Boston, Massachusetts at the Hynes Convention Center on July 11
5 Clustering Irregular Spaced LiDAR TINs for 3D Reconstruction presented at CITSA 2008 in Orlando, Florida at the Rosen Center Hotel on June 30
6 Posterboard for Triangulated, Connected Sets for Building Detection from Irregularly Spaced LiDAR presented at The 3rd International Symposium on Communications, Control and Signal Processing
7 Harris Seminar - Heuristic 3D Reconstruction of Irregular LiDAR
8 PhD Candidacy Exam Presentation

 

Algorithm Executables


No. Algorithms
Fuzzy ART Based Color Quantization Algorithm
Last Updated on Saturday, 07 November 2009 22:50