**1. Introduction**

98 Advances in Wavelet Theory and Their Applications in Engineering, Physics and Technology

[1] García, B., Vicente, J. & Aramendi, E. "Time-Spectral Technique for Esophageal Speech

[2] García, B., Vicente, J., Ruiz, I., Alonso, A. & Loyo, E. "Esophageal Voices: Glottal flow

[3] Baken, R. & Orlikoff, R. "Clinical measurement of speech and voice" Second Edition. San

[4] Brown, J.C. & Puckette, M.S. "A high resolution fundamental frequency determination

[5] Mallat, S "A Wavelet Tour of Signal Processing" Second Edition.. Academic Press, ISBN:

[6] Hooper, C.R. "Using evidence-based research in speech-language pathology: a project

[7] Cnockaert, L.; Grenez, F.; Schoentgen, J. "Fundamental Frequency Estimation And Vocal Tremor Analysis By Means Of Morlet wavelet Transforms", ICASSP 2005. [8] J. K. Maccallum, L. Cai, L. Zhou, Y. Zhang, and J. J. Jiang, "Acoustic analysis of aperiodic

[9] M. Carello and M. Magnano, "A first comparative study of oesophageal and voice

[10] D. Rudoy and T. Quatieri, "Time-varying autoregressions in speech: detection theory

[11] S. Aviyente and A. Yener Mutlu, "A time–frequency based approach to phase and

based on phase changes of the Fourier transform" J. Acoust. Soc. Amer., vol.94, pp.

that changed my thinking"American – Speech – Language – Hearing -Association,

voice: perturbation and nonlinear dynamic properties in esophageal phonation.,"

prosthesis speech production," EURASIP Journal on Advances in Signal

and applications," IEEE Trans. Audio, Speech, and Language Process., vol. 19, no.

phase synchrony estimation," IEEE Transactions on Signal Processing, vol. 59, no.

Diego, CA: Singular Publishing Group, ISBN:1565938690, 2000.

Journal of Voice, vol. 23, no. 3, pp. 283-90, May 2009.

Processing, vol. 2009, no. 821304, pp. 1-7, 2009.

**6. References** 

Regeneration" Biosignal '02, 2002.

Restoration" ICASSP 2005.

662-667, August 1993.

0-12-466606-X.

January 2003.

4, pp. 977-989, 2011.

7, pp. 3086-3098, 2011.

This chapter investigates the use of advanced signal processing techniques especially wavelet transforms to extract additional information from a two dimensional surface profile. The wavelet transform is able to aid the user in quickly assessing, visually, if the surface profile has a periodic or non-periodic component as well as if the profile signal is stationary or non-stationary. In addition, thresholds could be set at different frequencies of interest to automatically determine for the user if a periodic signal is present and if its magnitude is acceptable or not. The basis of this chapter is a doctoral dissertation by Lemaster (2004). A laser based, non-contact profilometer was used for all the surface profiles presented in this chapter though contact profilometers could also benefit from this type of analysis. The original work was conducted for wood and wood-based composites; however the signal processing techniques discussed in this chapter are applicable to all types of surfaces. In fact, an industry that would also like to determine if a surface profile is stationary or not or has periodic components is the road surface industry. They routinely use laser based optical profilometers very similar to the type used in this study except for the optics used to obtain the desired range and sensitivity. They are interested in detecting and quantifying pot holes, ruts, and washboard which are very similar to the surface characteristics of interest to the wood industry but on a different scale.

Traditional time domain analysis that is commonly used in the analysis of surface quality does not adequately show if a periodicity exits on the surface. While frequency domain analysis can reveal if the surface has a periodicity component it cannot adequately determine if the periodicity continues across the entire surface (stationary) or if it only extends across a portion of the surface (non-stationary). This information is of importance if the user wants to extend the capability of traditional surface quality analysis and not only quantify surface irregularities but classify them to both type and source.
