Impact damage characterization on CFRP parts using laser line scanning active thermography

This study presents a dual-path data processing framework for the detection and characterization of barely visible impact damage (BVID) in carbon-fiber- reinforced polymer (CFRP) structures using laser line thermography (LLT). A robotic LLT system was used to scan impacted CFRP specimens, and the resulting thermal sequences were analyzed using two complementary methods: full thermogram reconstruction followed by Pulse Phase Thermography (PPT) to detect subsurface delaminations, and Time-Summed Gradient Fil- tering (TSGF) to enhance surface-breaking cracks. Both processing paths produced interpretable results that were fused into a unified combined image and overlay mask, enabling simultaneous visualization of different defect types from a single scan. Quantitative analysis was performed on the binary masks to extract defect dimensions and Signal-to-noise ratio (SNR) values. The results demonstrated that delaminations and multiple cracks could be accurately detected and spatially distinguished, with good agreement to reference methods such as flash thermography and vibrothermography. This work highlights the potential of LLT as a versatile and scalable inspection technique, where multimodal defect detection and segmentation can be achieved through targeted processing and data fusion strategies. • A dual-path processing framework was developed for laser line thermography (LLT). • Subsurface delaminations and surface cracks were detected from a single LLT scan. • Combined and overlay images enabled multimodal visualization and interpretation. • Binary masks supported quantitative defect analysis and signal-to-noise estimation.