Turmeric Farm Monitoring and Automation Using Deep Learning and Fuzzy Logic on Raspberry Pi: A Low-cost and Energy Efficient Solution

Turmeric is valuable in various food, cosmetics, and medicine products. On the other hand, turmeric plants are susceptible to multiple diseases and pests, which can reduce crop yields. The proposed system combines deep learning, fuzzy logic, and Raspberry Pi to improve the efficiency and accuracy of turmeric farm management. The first component of our approach is a deep learning model that categorizes turmeric leaf images as healthy, diseased, or drought-stressed. The model is trained on a large dataset of turmeric leaf images and deployed for real-time classification on a Raspberry Pi. A fuzzy system for controlling grow lights, pesticide spray, air circulation fans, and water pumps is the second component of our system. The fuzzy system receives the deep learning model output as input and generates control signals for the actuators. The fuzzy system is intended to optimize the growth conditions for turmeric plants while reducing pesticide and water use. The system also considers environmental factors such as temperature, humidity, light intensity, and soil moisture. Our approach is built on Raspberry Pi, a low-cost, energy-efficient computing platform. As a result, our system is suitable for deployment in remote areas with limited access to electricity and other resources. The proposed system has achieved a classification accuracy of 95.47%. Simulink is used to build a standalone application for Raspberry Pi using the proposed deep learning model and Fuzzy system. Our system can increase the efficiency of turmeric farm management while lowering production costs. This can assist farmers in increasing their profits and improving their living conditions.