Wavelet co-efficient of thermal image analysis for machine fault diagnosis

Abstract

The ultimate goal of this study is to introduce a new method of machine fault diagnosis using different machine conditions data such as normal, misalignment, mass-unbalance and bearing-fault from infrared thermography (IRT). Using thermal image, it is easy to obtain information about the machine condition rather than other conventional methods of machine condition diagnostic technique. Thermal image technique can be successfully applied in the field electrical and electronics system, mechanical system, energy system and medical diagnosis. To get information from the image many techniques of image processing such as discrete Fourier transformation, discrete cosine transformation, neural networks, wavelet transform and many others methods is being used. In this study, our main focal point is to analysis thermal image by discrete wavelet decomposition and tries to find out significant result of machine condition monitoring. In this work, decomposition level of 2 shows satisfactory result for machine condition diagnosis.