This study explores the use of Deep Learning (DL) techniques for detecting diseases in coconut leaves. Traditional methods of disease detection in agriculture are often time-consuming, subjective, and prone to errors, particularly in large-scale operations. The study highlights the efficiency of Convolutional Neural Networks (CNN) in image-based tasks, such as plant disease identification and management. Specifically, it applies EfficientNet, a pre-trained CNN model, to enhance the accuracy and efficiency of detecting diseases in coconut leaves. The approach involves thorough data collection and preprocessing strategies, along with data augmentation to improve the model's robustness. The architecture includes convolutional layers for feature extraction, pooling layers to reduce dimensionality, and fully connected layers with activation functions to improve class separability. After training and testing the model on various coconut leaf disease datasets, the system demonstrated improved identification accuracy. This approach aims to reduce the reliance on manual detection methods, improve pest management, ultimately enhancing agricultural productivity.