Early Detection of Coconut Black-Headed Caterpillar Using Machine Learning Techniques

Abstract

Abstract: Coconut (Cocos nucifera L.) cultivation supports millions of farmers in tropical regions and plays a significant role in agricultural economies. However, coconut productivity is severely affected by the black-headed caterpillar (Opisina arenosella), a destructive leaf-feeding pest that reduces photosynthetic efficiency and nut yield. Conventional pest monitoring methods rely on manual inspection, which is labor-intensive and often fails to detect infestations at an early stage. This study proposes a machine learning–based framework for early detection of black-headed caterpillar infestation in coconut palms. The system integrates image-based leaf analysis and environmental parameters such as temperature and humidity. Supervised machine learning algorithms including Support Vector Machine (SVM), Random Forest, and Convolutional Neural Network (CNN) were implemented and evaluated using standard performance metrics. Experimental results indicate that CNN achieved the highest accuracy in detecting early-stage infestations. The proposed approach offers a scalable, cost-effective, and sustainable solution for intelligent pest monitoring in coconut plantations.