Spectral LiDAR analysis and terrain classification in a semi-urban environment
Loading...
Authors
McIver, Charles A.
Subjects
remote sensing
space systems operations
LiDAR
satellite laser altimetry
Optech Titan
multi-wavelength LiDAR
spectral LiDAR
terrain and building classification
space systems operations
LiDAR
satellite laser altimetry
Optech Titan
multi-wavelength LiDAR
spectral LiDAR
terrain and building classification
Advisors
Olsen, Richard C.
Stefanou, Marcus
Date of Issue
2017-03
Date
Mar-17
Publisher
Monterey, CA; Naval Postgraduate School
Language
Abstract
Remote-sensing analysis is conducted for the Naval Postgraduate School campus, containing buildings, impervious surfaces (asphalt and concrete), natural ground, and vegetation. Data is from the Optech Titan, providing three-wavelength laser data (532, 1064, and 1550 nm) at 10–15 points/m2. Analysis techniques for laser-scanner (LiDAR) data traditionally use only x, y, z coordinate information. The traditional approach is used to initialize the classification process into broad-spatial classes (unclassified, ground, vegetation, and buildings). Spectral analysis contributes a unique approach to the classification process. Tools and techniques designed for multispectral imagery are adapted to the LiDAR analysis herein. ENVI's N-Dimensional Visualizer is employed to develop training sets for supervised classification techniques, primarily Maximum Likelihood. Unsupervised classification for the combined spatial/spectral data is accomplished using a K-means classifier for comparison. The campus is classified into 10 and 16 classes, compared to the four from traditional methods. Addition of the spectral component improves the discrimination among impervious surfaces, other ground elements, and building materials. Maximum Likelihood demonstrates 75% overall classification accuracy, with grass (99.9%), turf (95%), asphalt shingles (94%), light-building concrete (89%), sand (88%), shrubs (85%), asphalt (84%), trees (80%), and clay-tile shingles (77%). Post-process filtering by number of returns increases overall accuracy to 82%.
Type
Thesis
Description
Series/Report No
Department
Information Sciences (IS)
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Approved for public release; distribution is unlimited.
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.