Damage monitoring in fibre-reinforced polymer composites using adaptive threshold methods and geometric features

Laminated composites are prone to impact-induced damage during their service operations, which is significant to assess the extent of damage. This study proposes a framework to assess the intensity of impact-induced damage in composites using digital image correlation (DIC) technique. Initially, glass fibre-reinforced polymer (GFRP) composite specimens were subjected to different indentation displacements of 5, 6 and 7 mm, and digital images were acquired on the front and rear sides of the indented specimens. Images are pre-processed using median and Gaussian filters, and their results were compared. Damage regions are segmented through Gaussian threshold (AGT) and adaptive mean threshold (AMT) methods, and the extent of damage was calculated using geometrical features. Pre-processing results show that the Gaussian filter performs better than the median filter. It was also found that AGT and AMT methods performed equally well in segmenting the damaged zone. Geometric features such as area and perimeter were able to quantify the induced damage under different indentation displacements. The result shows a better correlation (R = 0.85) between the dent depth and rear surface area obtained from the AGT segmented damage regions. This framework produces higher performance with simple instrumentation facilities; thus, it can be implemented in aerospace and automobile industries to evaluate the impact damage resistance in various structural components.