FarmIQ: An Intelligent Agricultural Solution

Authors: Kukku Anna Mathew, Lakshmi P P, Mala H S, Jasira B

Abstract: Agriculture is becoming increasingly vulnerable to changing weather conditions, soil nutrient imbalance, and the rapid spread of crop diseases, all of which place considerable pressure on farmers to make proper decisions with narrowed resources. In many cases, farmers continue to depend on intuition or general advice rather than structured analysis, resulting in poor crop selection, inefficient fertilizer application, and delayed recognition of plant infections. These challenges indicate a rising requirement for smart solution that simplify field-level decision- making while improving precision. This study presents FarmIQ, an integrated Agricultural Decision Support System designed to provide real-time, data-driven guidance for sustainable crop management. The system brings together multiple computational techniques, applying machine learning for crop recommendation, fertilizer optimization, and yield forecasting, along with a deep-learning-based method for automated leaf disease classification. FarmIQ evaluates essential soil and environmental parameters, including npk, pH, temperature, humidity, and rainfall, to find crops that are suitable with current field conditions. A fertilizer module analyzes nutrient deficiencies and generates targeted recommendations to maintain soil health, while yield prediction is done using regression models trained on historical production data. For disease detection, the system uses a fine-tuned VGG16 convolutional neural network capable of identifying common infections in tomato, potato, grape, and corn plants and providing treatment suggestions immediately after classification. All functionalities are delivered through a lightweight Flask based user facing platform to input soil details, upload leaf images, and obtain results instantly and clearly. Experimental evaluation shows that FarmIQ performs reliably across varied agricultural scenarios, demonstrating strong potential for practical use in field environments. The system contributes to improved decision accuracy, reduced dependency on agricultural experts, and more efficient resource utilization. By integrating predictive analytics and automated diagnosis into a unified platform, FarmIQ supports the transition toward data-driven and sustainable agriculture for farmers, students, and researchers.