An integrated smart surveillance system for diseases monitoring in tropical plantation forests

Abstract

The plantation forests are more susceptible to damage by pest and diseases than are natural forest. Monitoring of plantation forest is critical input to the Integrated Pest Management (IPM) process. This study proposes an integrated smart surveillance system for diseases monitoring in tropical plantation forests. The system, called Dismon, is an online and mobile application for monitoring and diseases identification in plantation forests. The application contains three components: digital library, monitoring system and disease identification system. In this work we build the intelligent system for disease monitoring by developing computer vision technology for disease identification through digital image processing. We have examined 1766 leaf samples containing five diseases: leaf spot, leaf blight, leaf curl, phyllode rust and anthracnose leaf spot. In this work, we train Support Vector Machine (SVM) for leaf diseases identification. The experimental results show that the proposed system obtained accuracy of 91% in differentiating healthy leaves and acacia leaf diseases. The ROC curve of acacia leaf identification indicates that the system is reliable to distinguish the leaf diseases. The resulting surveillance system is promising for plantation forests. This application can help surveyors, forest rangers or public users for gathering information, record observation and diseases identification in forest plantation.