Abstract—The rapid spread of diseases and pests has significantly impacted global agricultural productivity. Farmers often struggle with pest identification, leading to the overuse of pesticides, which causes environmental harm and incurs high costs. This work presents an early real-time insect identification system using deep learning for real-time mobile insect image detection. By applying the YOLOv5-S model to a 10-species insect dataset, the system achieved a mAP@0.5 accuracy of 70.5%, and 42.9% on the IP102 dataset, optimized for low-end mobile devices. Additionally, it provides farmers with vital information on insect biology, distribution, and management to reduce production costs and promote sustainable farming.
 
Keywords—deep learning, real-time insect identification system, YOLOv5, low-end mobile devices


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).