Several methods for the analysis of white light interferograms are presented and their performance is compared to a new method employing the wavelet transform in connection with sub-Nyquist sampling. As the result of computer simulations and experiments the wavelet method proves to be best suited to the problem. Furthermore, a method is proposed to characterize the microtopography of surfaces on the basis of fractal parameters, whereby these parameters are calculated by means of a wavelet method. The advantages of using wavelets in comparison to the well known fractal description by means of the power spectral density is, that this new method is less affected by noise and offers the ability to separate the roughness from the structure. Simulations show that even in the otherwise disturbing presence of sharp edges or spikes the separation of micro and macro structure is successfully performed by the proposed wavelet method.

Date Published: 17 September 1997
PDF: 11 pages
Proc. SPIE 3098, Optical Inspection and Micromeasurements II, (17 September 1997); doi: 10.1117/12.281153

Published in SPIE Proceedings Vol. 3098:
Optical Inspection and Micromeasurements II
Christophe Gorecki, Editor(s)