Comparison of Image Analysis Methods on the Example of Ultrasonic Thermography of an Aramid Composite

Abstract

Abstract Infrared thermography using ultrasound thermal excitation of the tested material is one of the most effective methods in non-destructive testing of a multi-layer aramid composite. This type of material is very popular in the construction of light ballistic armours. Typical defects are delamination between layers of aramid fabric joined by resin. They are usually filled with air. Delamination located deep under the surface of the test generates very weak temperature signals. They are often at the level of noise. To reduce the impact of noise on the detection of a defect, special methods of image analysis (thermograms) are used. Such methods include principal component analysis and wavelet analysis. Principal Component Analysis is a relatively new procedure of statistical data treatment, which is becoming increasingly popular in non-destructive testing. Mathematically, it is often regarded as implementation of the so-called singular values decomposition technique, which allows extracting of spatial information from a matrix of source data. The wavelet analysis is an integral transform, which represents the convolution of an analysed process with a special mother function called wavelet. Wavelets are characterized by two parameters: scale and shift. The paper presents a comparison of the efficacy of these methods in the detection of defects in the multilayer composite reinforced aramid fibre.