Optical voice recorder based on digital holography and usage for nondestructive testing

In this study, studies on optical voice recorder based on off-axis digital holography technique are presented. The work is presented in three main sections; 1-The reconstruction of sound wave (normal sound and ultrasonic sound waves) from recorded sound hologram obtained by optical voice recorder, listening reconstructed sounds and obtaining the frequency spectrum of sound waves. 2-The implementation of different signal processing algorithms for reconstruction process of sound waves obtained from sound holograms, the investigation of noise suppression capabilities of choosen algorithms and the calculation of the signal-to-noise ratio (SNR) of system for each signal processing algorithm. 3-The design of a detector that can be detect cracks in materials occurred with stroke or over time by using optical voice recorder based on off-axis digital holography technique. In the first section, Fourier transform method (FTM) is used for reconstruction. The sound waves obtained from sound holograms by using FTM are recorded into multimedia files and these recorded sound waves are listened. When the frequency spectrums are compared, it is proved that the reconstructed sound waves and the original sound waves – the sound waves applied to systems as input for recording- are exactly the same. In the second section, Fourier transform method (FTM), Fresnel-Transformation method (F-TM) and one dimensional-continuous wavelet transform method (1D-CWT) are implemented as signal processing algorithms. To compare the success of the above-mentioned algorithms in terms of supressing the noise in the system, the signal-to-noise ratios (SNR) are calculated in the system in cases each algorithm is implemented. In order to calculate the SNR values, the temporal phase profile obtained by using signal processing algorithm in reconstruction process and the input sound waves at the system are used. The SNR values are calculated as 66.90%, 8.70% and 4.78% for FTM, F-TM and 1D-CWTM respectively. One can easily see that FTM is the best method to suppress the noise in the system. In the last section, the studies are presented to design a detector that can detect the cracks in the materials formed by various reasons by using the proposed optical sound recorder. For crack detection, glass and wood materials in which the crack can be easily seen and metal in which the crack cannot be seen with the naked eye are used. This detection is performed by using the frequency spectrum obtained from sound hologram and obtaining the 3 dimensional behavior of sound wave while passing through the material. It is observed in the obtained frequency spectrums that the frequency of the cracked material is shifted more than the frequency of the solid material. In addition, from the 3D images, it is seen that the sound wave passes through the non-cracked part with more intensity than the cracked part. When comparing the 3D images for two cases, the cracked area is seen as a bump.