FOREST TYPE CLASSIFICATION USING AIRSAR IMAGE DATA

摘要

In the era of rapid and advance information and communication technology, more precise information can be captured using remote sensing satellite, Large amounts of data can be gathered cost-effectively and utilised to formulate and develop effective forest management plans (Lillesand & Kiefer 2000). The input of large amounts of inventory data using remote sensing technology helps enhance planning for sustainable management, especially in hilly areas where bulk of the remaining production and protection forests are located. Remote sensing data has wide application in forestry and is being commonly used in mapping of different forest types and in monitoring changes in the extent of forest cover. However, the application of remote sensing in assessingthe quality of forest, such as tree species composition and distribution, stocking (volume, height and dbh), growth and yield, is still very limited (Mohd Ibrahim et al. 2001). This is crucial for an effective decision-making process especially after logging where forecasting future timber production and length of the next rotation by forest managers depends largely on the compilation of these data.