Defects detection in rubber made products by pulsed phase thermography

Abstract

in this paper is discussed a powerful method of an enhancement of subsurface defects detection in automotive rubber made products via the pulsed phase thermography technique of the transient thermal data analysis. Hidden defects visibility improvement is attained by the artificial extension of an experimental thermal data set, acquired by a high sensitive infrared scanned detector camera, as well as by its upsampling before taking a one-dimensional discrete Fourier transform via a Goertzel algorithm. Tha data set extension in a periodic mode is originally done by using the two-dimensional discrete wavelet transform-based pattern. A lowpass linear finite impulse response digital filter, designed by using the least-square minimization method with the Kaiser window, is applied to prevent aliasing during the resampling procedure. It is demonstrated that the described approach to the transient experimental thermal data processing considerably amplifies defects detection posibilities of the employed type of a thermographic experimental equipment.