Electronic speckle pattern interferometry (ESPI) can provide accurate contour measurements in the micron
range and short measurement times far below one second. Typical surfaces in industrial applications, however,
often show discontinuities, like steps or holes. An unambiguous measurement of such surfaces is possible, if
the synthetic wavelength is chosen larger than the largest surface step. A long synthetic wavelength, however,
introduces a high noise level such that an unambiguous measurement combined with a high accuracy is not
possible in any case. Our preferred solution for this problem is the combination of two or more synthetic
wavelengths. In contrast to other publications (hierarchical, pixel-wise approach) our area-based approach uses
only two synthetic wavelengths minimizing measurement time and device complexity. In this paper we present
different methods for merging the phase images of the two synthetic wavelengths into one measurement result
combining accuracy and unambiguousness.
KEYWORDS: Linear filtering, Image filtering, Gaussian filters, Image fusion, Data fusion, Time metrology, Digital filtering, Nonlinear filtering, Wavefronts, Optical testing
Electronic speckle pattern interferometry (ESPI) can provide
accurate contour measurement in the micron range and short
measurement times far below one second. An advantage of this
method is that illumination axis and observation axis can be
identical in contrast to e.g. triangulation. Therefore ESPI
represents an interesting alternative to other optical measurement
principles used for surface profiling. Typical surfaces in
industrial applications often show discontinuities, like steps or
holes. An unambiguous measurement of such surfaces is only
possible if the synthetic wavelength is chosen larger than the
largest surface step. Since the noise level introduced to the
measurement increases proportional to the synthetic wavelength,
unambiguous measurements suffer from a loss of accuracy. The
solution for this problem is the combination of two or more
synthetic wavelengths. In contrast to other publications
(hierarchical, pixel-wise approach or temporal phase unwrapping)
our novel area-based approach uses only two synthetic wavelengths
minimizing measurement time and device complexity. The use of
areas instead of pixels allows a lower signal to noise ratio and a
smaller number of synthetic wavelengths (in our case only two)
respectively, compared to the hierarchical pixel based approach.
In this paper we present the steps required during pre-processing
(laser tilt and wave front compensation) and the opportunities and
drawbacks of different algorithms used for the fusion of the two
images gained from different synthetic wavelengths.
In some specific applications the electronic speckle pattern
interferometry (ESPI) is superior to other optical surface
metrology methods. The two-wavelength ESPI for surface contouring
can achieve both high accuracy of height resolution in the micron
range and short measurement times far below a second. A further
advantage of this method is that compared to e.g. triangulation,
illumination axis and observation axis can be identical. A problem
of interferometric methods in general are phase ambiguities
originating from discontinuous measurement object surfaces. A
common idea to decrease the range of ambiguity is the fusion of
several interferograms recorded at different wavelengths. This
paper presents a concept for a loss free sequential superposition
of several spatially separated laser beams as well as algorithms
for the determination of measured surface discontinuities. Also a
solution of a stability control for fast wavelength tuning of
laser diodes is presented.
Over the course of the last few years, several readout techniques for fiber Bragg grating (FBG) sensors have been proposed. However, all of them suffer from specific restrictions concerning response speed, accuracy, sensor multiplexibility and cost. In the past, it was often assumed that diffraction grating spectrometers were suitable only for FBG applications with modest resolution. The achievable pixel resolution is nowadays in the range of several tens of pm. For FBG sensors with typical temperature coefficients of 5 pm/K and strain coefficients of 0.7 pm/με this resolution is not sufficient for the majority of applications. We present a discussion on different methods for the subpixel registration of FBG spectra and we introduce a novel detection algorithm: the linear phase operator technique (LPO). Even under extreme noisy conditions LPO ensures a significant resolution enhancement by a factor of three compared to conventional algorithms and is shown to be very efficient in its implementation. The efficiencies of several conventional algorithms and LPO is compared by simulations and by means of a test bench. With slight efforts LPO is adaptable to further applications like spectrometer based Fabry-Perot sensors and other sensors with CCD detectors.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.