IoT cloud integration with EfficientNet-B7 for real-time pest monitoring and leaf-based classification

Abstract

The increasing prevalence of pest infestations poses a significant threat to global agricultural productivity, often resulting in substantial yield losses and economic damage. To address this challenge, this paper proposes an intelligent, cloud-enabled pest detection and classification framework leveraging state-of-the-art deep learning techniques. The proposed system integrates YOLOv8 for rapid and accurate pest detection with EfficientNet-B7 for fine-grained species-level classification. The framework is trained and evaluated using the Pestopia dataset, which contains annotated images representing diverse pest species. To enhance data diversity, robustness, and model generalization, data augmentation techniques such as center cropping and horizontal flipping are applied during preprocessing. YOLOv8 is employed to detect and localize pest instances within images, while EfficientNet-B7 extracts high-level discriminative features from detected regions to enable precise species identification. Furthermore, the system incorporates cloud-based real-time monitoring through Adafruit IO, enabling scalable, remote access to pest information for timely decision-making. The performance of the proposed framework is evaluated using standard metrics, including accuracy, precision, recall, and F1-score, achieving values of 97.8%, 98.9%, 98.4%, and 98.9%, respectively. The experimental results demonstrate the effectiveness and reliability of the proposed approach for real-time pest management. The cloud-integrated architecture facilitates proactive pest control strategies, supporting smarter, data-driven agricultural practices, and improved crop protection.