AI Powered Geo Poverty Analysis System using Deep Learning

AI Powered Geo Poverty Analysis System using Deep Learning is an advanced geospatial intelligence platform developed to estimate regional poverty levels using artificial intelligence and satellite imagery. The system integrates high-resolution daytime satellite images with nighttime VIIRS light intensity data to identify socioeconomic patterns and accurately predict regional wealth indices. Built using TensorFlow, Python, and Flask, the platform employs a Convolutional Neural Network (CNN) to perform automated feature extraction and poverty classification. The responsive React + Tailwind CSS dashboard enables users to upload satellite imagery, visualize predictions through interactive heatmaps, monitor analytics, and generate comprehensive reports. The backend provides secure REST APIs with JWT authentication, efficient model inference, dataset management, and automated report generation. Designed for researchers, government organizations, NGOs, and academic institutions, the platform demonstrates how artificial intelligence can support large-scale poverty assessment while reducing the dependency on expensive field surveys. This project is ideal for final-year engineering students, AI researchers, and organizations working in geospatial intelligence, remote sensing, sustainable development, and social impact analytics.