Introduction Rice crops are highly susceptible to diseases such as bacterial leaf blight, brown spot, and leaf smut. Early detection of these diseases is critical in order to avoid loss of rice crops. In this project, we have developed an AI-based system that detects rice leaf diseases from images and gives suggestions for treatment through generative AI . What Inspired Us Farmers face difficulty in identifying diseases in crops and may not have access to agricultural experts at all times. This inspired us to develop an intelligent system that detects diseases in rice leaves and gives suggestions for treatment.
How We Built the Project First, we stored our trained model in Amazon S3. Then we used Amazon SageMaker to run the prediction system. When an image of the rice leaf is uploaded, the system uses the deep learning model to detect the disease in the rice leaf. Once the disease is detected, the result is passed on to Amazon Bedrock, which uses Amazon Nova Lite to generate suggestions for treatment, including chemical treatment, organic treatment, and prevention tips.
What We Learned Through this project, we learned how to integrate deep learning models with cloud services as well as generative AI technology. We also learned how AI can be used to solve real-world problems in agriculture.
Challenges We Faced Some of the challenges faced during the project were how to integrate the model with cloud services, AWS permissions, as well as how to create prompts for the use of generative AI to provide useful agricultural advice to farmers.
Future Work In the future, we will integrate IoT devices such as temperature, humidity, and moisture sensors to provide environmental conditions for more precise disease detection.
Conclusion Through the project, we were able to show how AI technology can be used to help farmers detect diseases in their crops as well as provide them with smart treatment options to enhance agricultural productivity.
Log in or sign up for Devpost to join the conversation.