THz Signal Identification for Intelligent Characterization Under High-Resolution Mode based on RFECNet

Artificial intelligence (AI) technology has shown great potential in the automatic and intelligent identification of internal defects in composites based on terahertz (THz) spectroscopy. Based on the powerful feature extraction capability of deep learning, the proposed deep learning framework-based three-dimensional intelligent characterization system is proposed to detect the glass fiber reinforced polymer (GFRP) debonding defects in terahertz nondestructive testing (THz NDT), in which the defect datasets are firstly established by the THz time domain spectroscopy (THz- TDS), and then the robust feature extraction capability network (RFECN et) is adopted to realize the automatic and intelligent defect location and imaging by accurately classifying different THz signals. A series of experiments have been performed to validate the effectiveness of proposed system, which will provide a new solution for intelligent and automatic THz characterization of internal debonding defects of composites.