The paper presents an improved algorithm, Alpha-EIOU-YOLOv8, for detecting rice leaf diseases using deep learning. The authors modify the YOLOv8 model by replacing the original Box Loss function with a combination of EIoU loss and alpha-IoU loss to enhance the detection accuracy. The system collects images of rice leaves with three common diseases—leaf blast, leaf folder, and brown spot—and uses a two-stage approach: image collection and AI-based disease identification. The trained YOLOv8 model is deployed on IoT devices to detect and alert farmers about the diseases. The proposed method achieves an accuracy of up to 89.9% on a dataset of 3175 images, outperforming existing methods. The system's hardware design includes a micro-controller, USB camera, and GSM module, and it is tested on a Raspberry Pi 4 Model B. The results show that the modified YOLOv8 model significantly improves detection accuracy, making it suitable for real-world applications in rice fields.The paper presents an improved algorithm, Alpha-EIOU-YOLOv8, for detecting rice leaf diseases using deep learning. The authors modify the YOLOv8 model by replacing the original Box Loss function with a combination of EIoU loss and alpha-IoU loss to enhance the detection accuracy. The system collects images of rice leaves with three common diseases—leaf blast, leaf folder, and brown spot—and uses a two-stage approach: image collection and AI-based disease identification. The trained YOLOv8 model is deployed on IoT devices to detect and alert farmers about the diseases. The proposed method achieves an accuracy of up to 89.9% on a dataset of 3175 images, outperforming existing methods. The system's hardware design includes a micro-controller, USB camera, and GSM module, and it is tested on a Raspberry Pi 4 Model B. The results show that the modified YOLOv8 model significantly improves detection accuracy, making it suitable for real-world applications in rice fields.