This paper presents design and development of a Acoustic Resonance Testing machine which is a Non-Destructive Test (NDT).
Acoustic Resonance Testing (ART) is a rapidly growing NDT method for 100% testing of products inline.
It works on the principle of Acoustic Resonance, the object is impacted and its acoustic response is recorded and
analysed.
The acoustic response of an object is unique like a finger print and changes to the acoustic response
are used to identify defective parts.
In this paper we have developed a low cost automated inspection system using ART principles, improving both the accuracy and the efficiency of the existing ART systems.
We have chosen to implement the transform decomposition method of Discrete Fourier Transform.
It is computationally more efficient than the FFT as only a small subset of output points are required from a signal sampled at a very high rate for spectral analysis.
To improve the accuracy of the classification system (for defect detection) we have implemented the support vector machine based expert system to allow intelligent threshold detection.