Independent component analysis enhanced pulse thermography for high silicon oxygen phenolic resin (HSOPR) sheet with subsurface defects

As an important part of solid-propellant missile engine, the quality assurance of cladding layer 'high silicon oxygen phenolic resin (HSOPR) sheet' is very important. In this paper, independent component analysis enhanced pulse thermography (ICA-PT) was introduced to detect the subsurface defect of HSOPR. The basic principle of ICA was described. The experiment of HSOPR sheet with simulated flat bottomed holes (FBHs) detected by ICA-PT was carried out. Moreover, the comparative experiments of different algorithms such as pulse phase thermography, continuous wavelet transform and principal component analysis were carried out. The SNR of the characteristic image was analyzed. The experimental results demonstrated that ICA algorithm can improve the detection SNR significantly. Therefore, the research on ICA-PT is of significant importance to quantitative evaluation the quality of the high silicon oxygen phenolic resin sheet.