APPLICATION OF THE MMAC ALGORITHM TO TREE HEIGHT AND CROWN DIAMETER ESTIMATION IN MOUNTAINOUS FOREST

摘要

This paper applied the multi-level morphological active contour (MMAC) algorithm to the estimation of diameter at breast height (DBH), total height, and crown width of trees in mountainous forest based on rasterized airborne lidar data. The MMAC algorithm comprises three steps: a bottom up erosion (BUE) process to identify stand candidates, a top down dilation (TDD) process to estimate the crown periphery, and an active contour model (ACM) process to delineate crown contours. The total height (LH) and crown width (LCW) can be directly calculated by the MMAC method and then used as regressors in a multiple regression model for the estimation of diameter at breast height (LDBH). The results showed that the average estimation bias of LH, LCW, and LDBH is around 0.50 m, 2.54 m, and 8.7 cm respectively.