ABHINAV MANASAN (STM21CS003)
ADITHYAN P (STM21CS008)
NASEEM SWABAH V (STM21CS044)
SAYANTH RAJ K K (STM21CS055)

This project introduces a comprehensive wall crack and mold detection system powered by Convolutional Neural Networks (CNN), designed to provide accurate, real-time insights for building maintenance and structural integrity. By utilizing CNN-based image processing and feature extraction techniques, the system can identify different types of cracks—such as masonry, shear, and corrosion cracks—and link each to specific repair solutions, allowing users to take targeted preventive actions. This precise categorization helps building administrators and homeowners prioritize repairs, address underlying causes, and reduce long-term structural damage costs. In addition, the system features early mold detection, which is crucial for preventing material degradation, indoor air quality deterioration, and health risks. By enabling timely intervention, mold-related expenses and damage can be minimized. An integrated repair cost estimation module further enhances decision-making by providing approximate repair projections based on crack or mold severity, thereby supporting effective planning and budget management. Combining structural health monitoring with preventive maintenance, this dual-function system strengthens building resilience, promotes safety, and safeguards investments in both residential and commercial properties, while its mobile-friendly design ensures easy, user-centered access to high-precision diagnostics.