Oil palm tree enumeration based on template matching

Remote sensing imagery is one of the methods for agricultural monitoring provided with a proper technique of application and implementation. Since Malaysia is one of the top nations providing the palm oil product with best quality, thus research of sustainable utilization and development associating with oil palm tree and fruits should be the forefront of agenda. Currently, Malaysia is the world price benchmark for crude palm oil (CPO) even though Indonesia has taken the title of being the world's largest CPO producer in 2006. Thus, Malaysia need to enhance the research and development by utilizing remote sensing technology where oil palm plantation can be monitored and managed in a more effective, efficient and low-cost maintenance. This can be possible if the numbers of oil palm tree can be automatically enumerated and then categorized into healthy and disease trees based on remote sensing images. Observation from on-site plantation request higher cost and may lead to human error in providing accurate statistics of oil palm trees. This study intend to tackle this problem by integrating template matching analysis on WorldView-2 imagery data to discriminate the features in remote sensing imageries for enumerating oil palm trees. Thus, this study will focus on investigating the implementation of correlation coefficient of template matching techniques that best fit for WorlView-2 imagery data for recognition and characterization of oil palm tree.