Composite delamination detection using infrared images

This paper summarizes our results of detecting composite delamination using infrared camera. Rivets in airframe are usually made of composite materials. It is very difficult to detect rivet delamination as conventional optical methods can not identify the delaminations. Here a neural net-based image- processing tool was developed by Intelligent Automation, Inc. (IAI) to process the infrared images. The tool consists of Fast Fourier Transform (FFT), Principal Component Analysis (PCA), and Fuzzy CMAC (Cerebellar Model Arithmetic Computer) neural networks. Results show that our tool can accurately pinpoint those delaminated rivet heads.