Far East Journal of Ocean Research
Volume 2, Issue 3, Pages 187 - 202
(November 2009)
|
|
DIGITAL IMAGE PROCESSING AND AVICENNIA DETECTION FOR FOREST MAPPING IN PERSIAN GULF: EMPHASIS ON PLANTS INDICES
Saied Pirasteh (Malaysia), Ehsan Esfandyari (Iran), Javad Rashidi (Iran) and Ahmad Savari (Iran)
|
Abstract:
The Naiband is located in the Persian Gulf and south of Bushehr Province, Iran. It is close to Avicennia ecosystem region. The ASTER data has been used to detect and map the Avicennia forests. The attempt has been done to generate false color composition of bands 321 and detect the Avicennia forest region. Using plant indices for Avicennia forests mapping, the study reveals that the ASTER data and the digital image processing is a good technique for enhancement and detection of Avicennia. The focus in this study has been given to fifteen verified indictors. It shows that only six of them (i.e., IR-2, MSAVI, AST2-AST3, DVI IPVI, and BI) have proven to be capable for Avicennia forests mapping. This study shows that kappa coefficients of 77.16% and 84.22% can be considered as the best for measuring the area of Avicennia forests. It is found that indicators such as AST2-AST3 and DVI are having accuracy with 85.76% and 89.95% which are considerable for mapping too. However, the study shows the potential of the remote sensing data (ASTER) for detecting Avicennia. |
| Keywords and phrases: Avicennia forest, plant indices, ASTER data. |
| Communicated by Hyo Choi |
| Number of Downloads: 225 | Number of Views: 560 |
|
