A global interactive attention-based lightweight denoising network for locating internal defects of CFRP laminates

Systematic Evaluation of Ultrasonic In-Line Inspection Techniques for Oil and Gas Pipeline Defects Based on Bibliometric Analysis

“…In GIALDN, a thresholdbased denoising method was used to eliminate noise-related features and enhance feature discriminability. The results showed that GIALDN achieved a location accuracy of 98.68%, which was more than 15% higher than VGGnet11 and FaultNet, and outperformed LSTM, RNN, Rsenet18, SEresnet18, and Densenet121 [115]. For ultrasonic defect detection, which typically involves a larger volume of data compared to conventional eddy current testing, compressing input data into latent features can replace the indiscriminate retention of raw data.…”