|
|
Publication
|
| Research Title |
COMPARISON OF SPECTRAL MIXTURE ANALYSIS AND VEGETATION INDICES FOR FOREST CLASSIFICATION USING THEOS DATA |
| Date of Distribution |
25 October 2018 |
| Conference |
| Title of the Conference |
The 39th Asian Conference on Remote Sensing (ACRS 2018) |
| Organiser |
Malaysian Remote Sensing Agency (MRSA) Ministry of Energy, Science, Technology, Environment & Climate Change (MESTECC) Asian Association on Remote Sensing (AARS) |
| Conference Place |
Renaissance Kuala Lumpur Hotel |
| Province/State |
Malaysia |
| Conference Date |
15 October 2018 |
| To |
19 October 2018 |
| Proceeding Paper |
| Volume |
2018 |
| Issue |
39 |
| Page |
198 |
| Editors/edition/publisher |
|
| Abstract |
In general, a surface of land cover types is composed of a variety of natural mixtures.
Mixed pixels have been recognized as a problem affecting the use of remotely sensed data in
land cover classification. Spectral Mixture Analysis (SMA) approach is one of the most
commonly used methods for handling the mixed pixel problem. It can be used to provide a full
spectrum measurement of vegetation response, which makes it more robust than vegetation
indices. This study aims to compare SMA and three vegetation indices – Normalized Different
Vegetation Index (NDVI), Enhanced Vegetation Index (EVI) and Soil Adjusted Vegetation
Index (SAVI) – in forest classification based on Maximum Likelihood Classifier (MLC). The
area of study is Phukhieo Wildlife Sanctuary, Chaiyaphum province located in Northeast of
Thailand, covering an area of approximately1560 km2. THEOS multispectral images with 15m
resolution acquired 2013 was used for analysis process. The process of SMA included minimum
noise fraction (MNF), pixel purity index (PPI), and n-dimensional visualization. The
endmember extraction consists of four end member: Green Vegetation (GV),
nonphotosynthetic vegetation (NPV), Shade and Soil. The result shows that the accuracy and
kappa coefficient of 88.06% and 0.86 respectively were from SMA. The multispectral THEOS
data with 15-m resolution proved to be effective in analyzing through SMA technique for
classifying various forest types in high accuracy. |
| Author |
|
| Peer Review Status |
มีผู้ประเมินอิสระ |
| Level of Conference |
นานาชาติ |
| Type of Proceeding |
Full paper |
| Type of Presentation |
Oral |
| Part of thesis |
true |
| Presentation awarding |
false |
| Attach file |
|
| Citation |
0
|
|
|
|
|
|
|
|
|
Journal Publication
ไทย