Título: Evaluation of Forest Extent Changes: Multi-temporal Satellite Images
Autores: Podduwa Kankanamge Subash Chaninda Jayasinghe; Tokyo University of Agriculture and Technology
Masao Yoshida; Ibaraki University
Fecha: 2011-01-18
Publicador: Applied Remote Sensing
Fuente:
Tipo:

Tema: Environmental conservation; remote sensing
Remote sensing, monitoring forest, GIS, Neural network, Sri Lanka
Forest cover
Descripción: The main purpose of the present study was to develop a high accuracy forest extent map using spatial data mining techniques (Artificial Neural Network (ANN)), supervised and unsupervised classification and GIS overlay processing methods. Another purpose of this study was to monitor forest extent changes using multi-temporal satellite images (Land sat 5 TM 1992 and Aster 2006). Study area is located in Nuwareliya of Sri Lanka. The first approach consisted of supervised classification and the second approach consisted of unsupervised classification. Use of a combination of several classifications mapping provides better results than use of single classification. This is the new idea of this research and we propose a GIS overly processing technique that combines mapping with previous supervised and unsupervised classification to produce an improved forest cover map. For this purpose, derived thematic maps (supervised and unsupervised) were combined with GIS overlay techniques to generate a new map. Then, these three maps were reclassified and converted to ASCII format and suitable format for ANN modeling was prepared. Back-propagation algorithm was used for the implementation of ANN modeling. The overall accuracy of Aster was 96.2 whereas that of land sat TM was 94.6.  Results revealed that the extent of forest cover was lost by 5.28% in the present study area in the period between 1992 and 2006. The results of this study are expected to be useful for researchers, managers and policy makers for updating existing forest maps, detecting forest changes and planning.
Idioma: Inglés

Artículos similares:

Correlation of Multispectral Satellite Data with Plant Species Diversity vis-à-vis Soil Characteristics in a Landscape of Western Himalayan Region, India por Amit Chawla; Institute of Himalayan Bioresource Technology (CSIR),Amit Kumar,S Rajkumar,Rakesh Deosharan Singh,Ashwani Kumar Thukral,Paramvir Singh Ahuja
A case-study of in-stream juvenile salmon habitat classification using decision-based fusion of multispectral aerial images por Christine Louise Woll; University of Alaska Fairbanks, School of Fisheries and Ocean Sciences,Anupma PRAKASH,Trent SUTTON
Analysis of Spectral Vegetation Indices Related to Soil-Line for Mapping Mangrove Forests Using Satellite Imagery por Ibrahim Kasawani; Universiti Malaysia Terengganu, Malaysia,Usali Norsaliza; Universiti Putra Malaysia,Ismail Mohd Hasmadi; Universiti Putra Malaysia
A Neural Network Model for Mapping and Predicting Unconventional Soils at A Regional Level por El-Sayed Ewis Omran; Soil and Water Department, Suez Canal University, Ismailia, Egypt
Retrieving Tiger Habitats: Conserving Wildlife Geospatially por Suman Sinha; Research Fellow,Laxmikant Sharma; Assistant Professor,Mahendra Singh Nathawat; Professor & Head
10 
Mapping Spatial and Temporal Distributions of Kuwait SST Using MODIS Remotely Sensed Data por Mohammad M M M Alsahli; Geography Department, Kuwait University,Mohammad M M M Alsahli; Geography Deprtment, Kuwait University,Kevin P Price; Geography Department, University of Kansas,Robert Buddemeier; Kansas Geological Survey, University of Kansas,Daphne G. Fautin; Ecology & Evolutionary Biology, University of Kansas,Stephen Egbert; 2Geography Department, University of Kansas