KisanSeva – An Intelligent Cropping System

In agriculture, small landholders often face challenges in maximizing profits due to limited access to technology and reliable data. This report proposes a technological solution leveraging modern tools and frameworks to address these issues. The solution integrates Flutter for cross-platform application development, TensorFlow for machine learning predictions, and various data sources including Weather APIs for real-time environmental data. Firebase and Firestore are utilized for cloud storage and synchronization, while Firebase Authentication ensures secure access to the application. Key features of the proposed solution include an IoT integration using Node MCU devices to collect real-time agricultural data such as soil pH and rainfall. This data is streamed to a Firebase Real-time Database, where a machine learning model developed in Google Colab predicts optimal crop choices based on historical and real-time data analysis using Pandas and NumPy for data manipulation. Additionally, the application includes a Feed feature aimed at providing curated news relevant to the agricultural industry. This feature leverages REST APIs to aggregate and deliver up-to-date information to users, enhancing the decision-making capabilities. Overall, this report outlines a comprehensive technological solution tailored to empower small landholders by providing actionable insights, predictive analytics, and timely information, ultimately enabling them to make informed decisions that lead to increased profitability in agriculture.