Inspiration

Heritage monuments are valuable cultural assets, yet many face gradual deterioration due to environmental exposure, biological growth, and human activities. Traditional inspection methods are often periodic, time-consuming, and resource-intensive. Meanwhile, tourists capture thousands of photographs of these monuments every day, but these images are rarely used for conservation purposes. This inspired us to develop ANCESTRA, a platform that transforms crowdsourced photographs into a potential resource for intelligent heritage monitoring and preservation.

What it does

ANCESTRA is an AI-powered heritage informatics platform that combines computer vision with predictive conservation analytics. It aims to analyze heritage site images, identify structural damage such as cracks, erosion, and biological growth, and assess the severity of detected anomalies. The platform also supports heritage site management, damage history tracking, reporting, and notifications to help authorities make informed conservation decisions.

How we built it

ANCESTRA is designed as a modular web application integrating frontend, backend, database, and artificial intelligence components. React.js is used for the user interface, while Node.js/Express or Python Flask supports backend services. MongoDB manages application data, and Python with OpenCV and deep learning frameworks supports image processing and damage detection. The system is designed to connect uploaded images with analysis results and heritage conservation records.

Challenges we ran into

Key challenges included understanding the technical requirements of heritage conservation, handling photographs captured under different lighting conditions and viewing angles, identifying suitable approaches for damage detection, and planning reliable integration between the application and AI modules. Another challenge was designing a system that could support future monitoring and comparison of images collected over time.

Accomplishments that we're proud of

We established a clear framework for crowdsourced heritage monitoring and developed a structured plan connecting image uploads, AI-based damage analysis, severity assessment, and conservation management. We are particularly proud of the idea of repurposing everyday tourist photographs as a potential source of visual evidence for heritage preservation.

What we learned

Through this project, we gained experience in requirement analysis, system architecture, web application development, database design, computer vision, and AI integration. We also learned about the importance of image quality, data consistency, modular development, and testing when building AI-powered applications for real-world problems.

What's next for ANCESTRA

Our future plans include improving damage detection accuracy, expanding the training dataset across multiple heritage sites, implementing cross-temporal image comparison, and developing more reliable severity assessment. We also aim to enhance predictive conservation scheduling, automated work-order generation, cloud deployment, and authority notifications to support scalable, data-driven heritage preservation.