Design and Development of an AI-based LINE Chatbot for Detection and Identification of Major Chili Plant Diseases

¹ Faculty of Agricultural Innovation, College of Agricultural Innovation and Food, Rangsit University, Pathum Th
² Department of Farm Mechanics, Faculty of Agriculture, Kasetsart University, Bangkok, Thailand

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

This study explored the use of deep learning techniques, particularly variations of the YOLOvl 1 architecture, for the detection and classification of major diseases in chili plants. Chili is an economically significant crop in Thailand: however, its productivity is often hindered by common foliar diseases. A dataset comprising 7.500 annotated images of three prevalent chili diseases. Cercospora leaf spot, pepper yellow leaf curl, and anthracnose. was used to train and evaluate four YOLOvl 1 variants: YOLOvl In. YOLOvl Is. YOLOvllm, and YOLOvl 11. Among these. YOLOvl In demonstrated the highest performance, achieving a mean Average Precision (mAP) of 89%. To support real-time disease monitoring in the field, the best-performing model was integrated into the LINE chatbot platform. This Al-powered tool enables rapid and accessible disease identification, thereby enhancing early detection and promoting more effective disease management practices with the potential to improve crop yields and reduce agricultural losses.