Completed
InfraScan
MATLAB
Machine learning
Image processing
Deep learning
CNN

Project Overview
InfraScan is a hands-on, AI-powered project that guides you through developing a real-time structural health monitoring system using infrared thermal imaging and deep learning. By combining MATLAB’s powerful image processing and neural network capabilities, you’ll build a complete solution to detect structural defects—like cracks and moisture—without requiring costly sensors or hardware access. By the end, you will have: Processed and analyzed infrared images using MATLAB. Built and trained convolutional neural networks (CNNs) to detect anomalies. Developed a real-time visualization interface for live defect detection. Gained practical experience in image preprocessing, model tuning, and deployment.
Project Steps
- Established MATLAB development environment
- Applied fundamental Image processing techniques
- Implemented deep learning with CNNs
- Collected and preprocessed Infrared datasets
- Developed and trained CNN model for defect detection and evaluated its performance
- Developed real-time monitoring interface
- Documented and presented professionally (Canva and OBS Studio)
Project Details
Mentors
- Vikash Sunda (Y23)
- Shauya Srivastava (Y23)
- Kshitij Srivastava (Y23)
- Mohit Meena (Y23)