Processing Multitemporal TM Imagery to Extract Forest Cover Change Features in Cleveland National Forest, Southern California
Project Start Date
01/01/2000
Project End Date
01/01/2003
Project Call Name
Science Theme Name
Regional_Initiative_Name
Region
Solicitation
Team Members:
Person Name | Person role on project | Affiliation |
---|---|---|
Janet Franklin | Principal Investigator | Arizona State University, San Diego, United States |
Abstract
Multitemporal Spectral Mixture Analysis • Highly accurate in depicting changes in forest cover: Soil, NPV and GV are sensitive to inter-date change in ‘natural’ cover • Decision Tree Classifier outperformed a Conventional Maximum Likelihood Classifier by 10% overall and by an average of 9% for each cover change class • Inclusion of ancillary data sets to reduce spectral confusion and texture images to enhance accuracy
Project Documents