Hyperspectral clustering and unmixing for studying the ecology of state formation and complex societies

Justin D. Kwong; David W. Messinger; William D. Middleton

doi:10.1117/12.826354

17 August 2009 Hyperspectral clustering and unmixing for studying the ecology of state formation and complex societies

Justin D. Kwong, David W. Messinger, William D. Middleton

Proceedings Volume 7457, Imaging Spectrometry XIV; 74570E (2009) https://doi.org/10.1117/12.826354
Event: SPIE Optical Engineering + Applications, 2009, San Diego, California, United States

Abstract

This project is an application of hyperspectral classification and unmixing in support of an ongoing archaeological study. The study region is the Oaxaca Valley located in the state of Oaxaca, Mexico on the southern coast. This was the birthplace of the Zapotec civilization which grew into a complex state level society. Hyperion imagery is being collected over a 30,000 km2 area. Classification maps of regions of interest are generated using K-means clustering and a novel algorithm called Gradient Flow. Gradient Flow departs from conventional stochastic or deterministic approaches, using graph theory to cluster spectral data. Spectral unmixing is conducted using the RIT developed algorithm Max-D to automatically find end members. Stepwise unmixing is performed to better model the data using the end members found be Max-D. Data are efficiently shared between imaging scientists and archaeologists using Google Earth to stream images over the internet rather than downloading them. The overall goal of the project is to provide archaeologists with useful information maps without having to interpret the raw data.

Citation Download Citation

Justin D. Kwong, David W. Messinger, and William D. Middleton "Hyperspectral clustering and unmixing for studying the ecology of state formation and complex societies", Proc. SPIE 7457, Imaging Spectrometry XIV, 74570E (17 August 2009); https://doi.org/10.1117/12.826354

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available