A system for automatic rice disease detection from rice paddy images serviced via a Chatbot

A LINE Bot System to diagnose rice diseases from actual paddy field images was developed and presented in this paper. It easy-to-use automatic system designed help farmers improve the yield quality. The targeted were taken environment without special sample preparation. We used a deep learning neural networks technique detect images. an object detection model training refinement process performance of our previous research on leave detection. based analyzing model’s predictive results could be repeatedly quality database next model. deployment for created selected best paper, YOLOv3, trained by refined data set. measured 5 target classes found that Average True Positive Point improved 91.1% paper 95.6% study. Therefore, we Rice Disease system. Our worked automatically real-time suggest primary diagnosis users group, which included disease specialists. They communicate freely via chat. In real deployment, own defined measurement average 78.86%. fast took only 2–3 s server.