Design of an Intelligent System for Defect Recognition in Composite Materials using Lock-In Thermography

This paper examines the lock-in thermographic technique for detecting Teflon defects within the composite material with a polymer matrix (Carbon Fiber-reinforced polymers, CFRP). In particular, a deep learning based network, made of a succession of convolutional layers, is implemented to process single thermal sequences generated in a simulation environment. As a result, the proposed methodology can accurately identify subsurface defects.