A comparative study of template matching, ISO cluster segmentation, and tree canopy segmentation for homogeneous tree counting

摘要

Counting trees can be challenging due to the crowded environment, time-consuming, and expensive operation. The information on the locations and the number of oil palm trees in a plantation area is important in many aspects. First, it is important to predict the yield of palm oil, which is the most widely used vegetable oil in the world. Second, it provides essential information to understand the growing situation of palm trees after plantation, such as the age or the survival rate of the palm trees. As such, this research investigated tree counting extraction of oil palm plantation. The research area is located at an oil palm plantation area (Felda Pasir Raja) in Johor, Malaysia. Three methods of extraction had been used in this research, i.e. Template Matching Algorithm (TMA), ISO Cluster Unsupervised Classification (ICUC), and Tree Canopy Segmentation (TCS). The results obtained using TCS emerged as the best method for tree counting in this research. The number of the trees detected by the TCS method was 77,963 trees, while its percentage was 96%. As for TMA and ICUC, the percentages were 89% and 82%, respectively. Therefore, this research could be used amongst the plantation organisations, especially the oil palm industries, which are responsible to monitor the status of oil palm trees for effective palm oil production.