Tensor-based ultrasonic signal processing for defect detection in fiber reinforced polymer (FRP) structures

Ultrasonic testing (UT) technique has been widely used in defect detection of composite materials. To better identify the defective regions, a number of one- or two-dimensional signal processing methods have been adopted for defect signal enhancement. However, the application of these methods is limited by their complex operation. Most of the existing methods cannot deal with the entire three-dimensional tensor of UT data in an efficient manner. In order to solve this problem, a third-order tensor decomposition method, Tucker3, is adopted in this paper for UT-based defect detection in fiber reinforced polymer (FRP) structures. After Tucher-3 decomposition, the defect information is extracted by a small number of factors, which is further summarized by the leverages. The candidate defective regions are then identified from the leverages, based on which the locations and the shapes of the defects can be calculated by clustering.