Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{177532,
        author = {DEREK JOSE G and SURESH M and MUTHUKUMARAN S and EDWIN DHAS P},
        title = {ONLINE MONITORING OF UNAUTHORIZED CONSTRUCTION ACROSS THE CITY},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {812-816},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=177532},
        abstract = {Urbanization has accelerated across the globe, bringing with it the challenge of regulating and managing land use effectively. Unauthorized construction activities not only violate municipal planning norms but also pose significant risks to public safety, environmental sustainability, and the equitable distribution of city resources. Traditional approaches to monitoring such developments—such as physical inspections by municipal staff, drone-based surveillance, or remote sensing using satellite data—often suffer from high operational costs, limited coverage, slow response times, and an overreliance on manual validation. These limitations create opportunities for undetected and unregulated construction to persist. To address these critical issues, this project introduces a comprehensive, AI-powered web-based monitoring system that automates the detection of unauthorized constructions throughout a city. The system harnesses deep learning algorithms—primarily a lightweight and efficient MobileNetV2-based Convolutional Neural Network (CNN)—to analyze and compare satellite imagery captured at different time intervals. By detecting structural differences between historical and current images of urban plots, the system can effectively identify changes that may signify illegal building activity. The architecture of the solution is modular and scalable. The process begins with image acquisition, wherein satellite images are retrieved from sources such as Google Earth Engine. These images undergo preprocessing steps such as noise reduction and contrast enhancement to improve clarity and usability. The images are then segmented into discrete land parcels for focused analysis. Users—including homeowners, builders, or real estate developers—can register on the platform and submit property details and construction plans for approval. Administrators have access to a dedicated review dashboard, where they can verify submissions, view satellite imagery comparisons, and oversee flagged discrepancies. The system also integrates location services through Google Maps APIs and offers visual geo-mapping of detected violations. Technologically, the solution is built using Python for backend logic, incorporating TensorFlow for AI model training and OpenCV for image processing. Geographic Information Systems (GIS) tools such as ArcGIS and QGIS support spatial data analysis, while a web interface is developed using frameworks like Flask or Django. The database layer is supported by PostgreSQL with PostGIS extensions for managing geo-spatial queries and storage. This AI-driven platform offers a paradigm shift from reactive to proactive governance in urban development monitoring. It not only minimizes human error and reduces surveillance costs but also improves the transparency and accountability of municipal operations. By bridging the gap between AI technologies and urban governance, this system stands as a model for smart city applications, enabling authorities to maintain regulatory compliance, protect public infrastructure, and promote sustainable city growth.},
        keywords = {},
        month = {May},
        }