ArticlePublisher preview available

Improvement of accuracy in digital holography by use of multiple holograms

Optica Publishing Group
Applied Optics
Authors:
  • Rheinmetall Defence Electronics GmbH
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Speckle pattern decorrelation reduces the accuracy of interferometric shape and deformation measurements. We introduce a technique for the reduction of speckle noise in digital holography. The method is not based on classical filtering techniques such as median filters. Instead it utilizes the shift theorem of the Fourier transform. For this method several holograms of the same object under test are recorded. The reconstruction leads to a set of object wave fields with different speckle patterns. A proper averaging procedure, taking into account the properties of the wrapped phases, leads to an improvement of the accuracy in the resulting phase difference. The theory of the applied method is described and our first results for technical components with an improvement of accuracy up to 1 / 57 of the wavelength are presented.
This content is subject to copyright. Terms and conditions apply.
Improvement of accuracy in digital holography by use
of multiple holograms
Torsten Baumbach, Ervin Kolenovic´ , Volker Kebbel, and Werner Jüptner
Speckle pattern decorrelation reduces the accuracy of interferometric shape and deformation measure-
ments. We introduce a technique for the reduction of speckle noise in digital holography. The method is
not based on classical filtering techniques such as median filters. Instead it utilizes the shift theorem of
the Fourier transform. For this method several holograms of the same object under test are recorded. The
reconstruction leads to a set of object wave fields with different speckle patterns. A proper averaging
procedure, taking into account the properties of the wrapped phases, leads to an improvement of the
accuracy in the resulting phase difference. The theory of the applied method is described and our first
results for technical components with an improvement of accuracy up to 157 of the wavelength are
presented. © 2006 Optical Society of America
OCIS codes: 030.6140, 030.6600, 070.2590, 090.1760, 090.2880, 120.5050.
1. Introduction
Holographic interferometry has been used success-
fully for many years for the inspection of technical
components. The technique uses holographic plates
for the recording of holograms. Processing and han-
dling of the hologram plates are complex and time-
consuming. The introduction of digital holography
eliminated these drawbacks by using a CCD camera
for recording the holograms.
1,2
Using a CCD camera
and hence having direct access to the complex wave
field introduced a new degree of flexibility to optical
metrology.
3
In digital holography the reconstruction
of the holograms is done numerically in the computer.
For this, different reconstruction algorithms like the
convolution,
4
the Fresnel,
1
or the Fourier
5
approach
have been developed.
Despite the advantages of digital holography, it has
the same drawback as most coherent methods:
speckle noise due to the roughness of the surface.
Speckle noise reduces the spatial resolution and the
signal-to-noise ratio of the measurements.
6
Although
many investigations are dealing with the reduction of
speckle noise, the problem has yet to be solved
satisfactorily.
7–13
A significant reduction of speckle
noise for the investigation of transparent objects was
achieved by Kebbel et al.
14
They used an averaging
technique over several speckle fields, but the tech-
nique is not applicable to opaque objects.
The proposed technique in this paper uses the
Fourier approach in digital holography to reduce
speckle noise for opaque objects with rough surfaces.
Several digital holograms of the object are recorded
from different lateral positions of the CCD camera.
According to the shift theorem of the Fourier trans-
form, a lateral shift of an infinite-sized hologram
would only cause an additional phase factor in the
object plane. However, because of the finite size of the
CCD camera, the reconstructed wave field results
from a convolution of the object wave with a sinc
function. This convolution causes different speckle
patterns to arise from different hologram positions.
Consequently, for a set of laterally shifted holograms,
we receive a set of phase differences that can be used
for an averaging process.
2. Lensless Fourier Holography
In lensless Fourier holography
15,16
the object and ref-
erence source point are located in the same plane
parallel to the hologram plane. Therefore both source
points have the same distance from the hologram (see
Fig. 1).
The reconstruction algorithm for lensless Fourier
holography is based on the Fresnel reconstruction.
The reference source point ris located at x,y
0, 0in the object plane, and the object wave field in
The authors are with the Bremer Institut für angewandte
Strahltechnik, Klagenfurter Strasse 2, D-28359 Bremen, Ger-
many. The e-mail address for T. Baumbach is baumbach@bias.de.
Received 25 October 2005; revised 1 March 2006; accepted 13
April 2006; posted 19 April 2006 (Doc. ID 65601).
0003-6935/06/246077-09$15.00/0
© 2006 Optical Society of America
20 August 2006 Vol. 45, No. 24 APPLIED OPTICS 6077
... Several approaches have been developed in DH and DHM to diminish the adverse speckle effects to widen the use of these imaging methods; however, these ruin the image contrast and reduce the spatial resolution. In general, the methods to reduce speckle noise can be grouped into two categories: physical approaches [8][9][10][11][12][13][14][15][16] and computational approaches. [17][18][19][20][21][22][23][24][25][26][27] The physical techniques are commonly based on two methodologies. ...
... Therefore, the average of the reconstructed images provides a numerical reconstruction with reduced speckle noise following a 1∕ ffiffiffiffi Q p law. 28 Such holograms can be acquired by changing the angle of illumination through a rotating mirror, 8 slightly rotating the object, 9 and lateral shifting the object 10 or the sensor. 11 Another reported physical technique consists of using different wavelengths to register multiple holograms. 12 In addition, other physical methods use some optical elements to reduce the light's spatial coherence partially. ...
... Several approaches have been developed in DH and DHM to diminish the adverse speckle effects to widen the use of these imaging methods; however, these ruin the image contrast and reduce the spatial resolution. In general, the methods to reduce speckle noise can be grouped into two categories: physical approaches [8][9][10][11][12][13][14][15][16] and computational approaches. [17][18][19][20][21][22][23][24][25][26][27] The physical techniques are commonly based on two methodologies. ...
... Therefore, the average of the reconstructed images provides a numerical reconstruction with reduced speckle noise following a 1∕ ffiffiffiffi Q p law. 28 Such holograms can be acquired by changing the angle of illumination through a rotating mirror, 8 slightly rotating the object, 9 and lateral shifting the object 10 or the sensor. 11 Another reported physical technique consists of using different wavelengths to register multiple holograms. 12 In addition, other physical methods use some optical elements to reduce the light's spatial coherence partially. ...
Article
Full-text available
Images recorded by coherent imaging systems, including laser-based photography, digital holography (DH), and digital holographic microscopy (DHM), are severely distorted by speckle noise. This work presents a single-shot image processing method to reduce the speckle noise, coined hybrid median-mean filter (HM2F). The HM2F is based on the average of conventional median-filtered images with different kernel size. The synergic combination of the median filter and mean approach provides a denoised image with reduced speckle contrast while the spatial resolution is kept up to 97% from the original value. The HM2F method is compared with the conventional median filter approach (CMF), the 3D Block Matching (BM3D) filter, the non-local means (NLM) filter, the 2D windowed Fourier transform filter (WFT2F), and the Wiener filter using different speckle-distorted images to benchmark its performance. Based on the experimental results and the simplicity of the technique, HM2F is proposed as an effective denoising tool for reducing the speckle noise in laser-based photography, DH, and DHM.
... Over the years, a large variety of methods and approaches has been published on speckle filtering is introduced. Boumbach et al. [29] Introduced an approach based on the use of several recorded holograms of the same specimen for the reconstructed phase map quality improvement. Whereas, Rong et al. [30] proposed another approach where multiple off-axis holograms are recorded using a circularly polarized illumination beam and a rotating linearly polarized reference beam, and the speckle noise is removed in the reconstructed images by averaging these fields. ...
Article
Speckle-noise filtering is an extensive and key process in coherent interferometric techniques to obtain important and accurate information from the recorded interferogram or fringe pattern. The speckle-noise inherent to these interferograms, recorded by digital image sensors in these optical techniques, is eliminated by an appropriate image processing method. Since the beginning and further development of these techniques, a wide variety of speckle noise filtering methods and techniques have been proposed. The present research work aims to exploit the performance of the non-local sparse principal component analysis (NLS-PCA) method for speckle noise reduction as applied to the interferometric fringe patterns obtained from digital holographic and digital speckle pattern interferometric techniques. The performance of the NLS-PCA is evaluated on numerical simulations using different types of speckle fringe patterns resulting in the method being highly effective in filtering the speckle noise and providing superior results evaluated in terms of peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), Edge preservation Index (EPI), and Equivalent Number of look (ENL), in comparison to widely known methods such as windowed Fourier transform method, Lee filter, Weiner filter. Furthermore, the proposed method maintains the finer details and preserves the fringe edges effectively. The performance of the NLS-PCA method is also demonstrated on experimental speckle fringe patterns and digital holograms.
... However, due to the use of coherent light sources, DHM recordings are subject to the deleterious effects of coherent noise, commonly known as speckle, which is derived from the roughness that most real-world materials present on the optical scale [11]. Multiple strategies for the reduction of speckle noise in Digital Holography can be found in the literature, ranging from physical alterations to the recording system [12,13], numerical processing of the reconstructions [14,15], or a combination of both [16]. Specifically for DHM, the Optics and Opto-Digital Processing group has also made several denoising proposals by both physical [17][18][19] and numerical [20][21][22][23] approaches which take into account the specific experimental conditions of microscopy imaging systems. ...
Article
In this work, the contributions that the research group of Optics and Opto-Digital Processing have developed to democratize digital holographic microscopy (DHM) are reviewed. The said group, based on the Universidad Nacional de Colombia Sede Medellin, has focused its work on developing tools, both of software and hardware, for DHM architectures, both without and with lenses, with the aim of making these cutting-edge technologies of microscopy accessible to many people. Cost-effective hardware designs, open-source processing software, and user-friendly modeling platforms are the main contributions of our research group to the democratization of DHM. The fundamentals of DHM, both without and with lenses, are briefly recalled to introduce the above-mentioned review.
... In optical ways, multiple acquisitions are made which provide different speckle patterns, and a series of uncorrelated noisy holograms are reconstructed and superimposed on the intensity to reduce speckle noise. These holograms were obtained by time diversity with a moving diffuser [8,18] or double diffusers [11], different wavelengths [10,15], varying light polarization [20,22], angular variation of the light beam [19,21], spatial varying position of the record camera [2,16] or an object [17]. In addition, the recorded hologram was split into several subholograms, and the superposition of sub-holograms reconstruction was utilized to reduce speckle noise [7]. ...
Article
Full-text available
Digital holography is a promising imaging technology. However, there is speckle noise in the reconstructed image of a digital hologram. Speckle degrades the quality of the reconstructed image. Suppression of speckle noise is a challenging problem in digital holography. A novel method is proposed to reduce speckle by a single-shot digital hologram in this paper. In the proposed method, a single-shot digital hologram is obtained by a conventional experiment setup without additional requirements. At the same time, different binary masks are designed and generated on Bernoulli process in a computer. Then the single-shot digital hologram is sampled by these binary masks to generate multiple holographic patterns. Eventually, these holographic patterns are reconstructed to suppress speckle noise in the reconstruction image. Simulation and experiment are made to verify the proposed method. Results show that it is effective and feasible for the proposed method to reduce speckle in the digital holography. The proposed method can be applied into various fields such as three-dimensional imaging, holographic diagnosis, and art display.
... In real conditions, there are several degradation sources that may induce more decorrelation noise than expected if all is perfect. As examples, the reconstruction of holographic data might not be perfectly in focus [30], the pixels could have a large active surface [3], the recording could have a low number of pixels or saturated pixels [31], the number of useful quantization bits could be insufficient [32], or there also could be wavelength changes between exposures [33]. As a consequence, all of these degradation sources have an increase in speckle decorrelation and then an increase in noise. ...
Article
Full-text available
Digital holography is a very efficient technique for 3D imaging and the characterization of changes at the surfaces of objects. However, during the process of holographic interferometry, the reconstructed phase images suffer from speckle noise. In this paper, de-noising is addressed with phase images corrupted with speckle noise. To do so, DnCNN residual networks with different depths were built and trained with various holographic noisy phase data. The possibility of using a network pre-trained on natural images with Gaussian noise is also investigated. All models are evaluated in terms of phase error with HOLODEEP benchmark data and with three unseen images corresponding to different experimental conditions. The best results are obtained using a network with only four convolutional blocks and trained with a wide range of noisy phase patterns.
Article
Full-text available
In this work, we present a new method to reduce the shot noise in phase imaging of digital holograms. A spatial averaging process of phase images reconstructed at different reconstruction distances is performed, with the reconstruction distance range being specified by the numerical focus depth of the optical system. An improved phase image is attained with a 50% shot noise reduction. We use the integral of the angular spectrum as a reconstruction method to obtain a single-object complex amplitude that is needed to perform our proposal. We also show the corresponding simulations and experimental results. The topography of a homemade TiO 2 stepwise of 100 nm high was measured and compared with the atomic force microscope results.
Article
Digital Holography (DH) is a popular and practical technology for recording the complex-valued information of the object field. However, speckle, caused by the coherent light, greatly reduces the quality of the reconstructed image. In this paper, a speckle-suppression algorithm called Gauss speckle elimination mask (GSEM) that only needs a single-shot DH is proposed. This method utilizes the Gauss window function to extract the sub-holograms. Besides, the gray value compensation is proposed in this method for restoring the grayscale distribution. All sub-holograms are operated by the gray value compensation, then a new stack of sub-reconstructed images is acquired. The output compounded by this stack of sub-reconstructed images avoids the additional noise caused by the ringing effect. It could be shown in this paper that this method has a better performance in speckle-suppression while maintaining the image contrast.
Article
Full-text available
A new method of phase determination in hologram interferometry is described. The Fresnel holograms, which represent the undeformed and the deformed states of the object, are generated on a CCD target and stored electronically. No lens or other imaging device is used. The reconstruction is done from the digitally stored holograms with mathematical methods. It is shown that the intensity as well as the phase can be calculated from the digitally sampled holograms. A comparison of the phases of the undeformed and the deformed states permits direct determination of the interference phase.
Article
Full-text available
Generally, speckles impair the reconstruction of digital and optical holograms of diffusely scattering objects. With the help of an iterative method, it is possible to introduce diffusers in digital holography that do not suffer from this disadvantage. Optically obtained speckle-free reconstructions of digital holograms are presented.
Article
Full-text available
The out-of-plane displacement of a cantilever is measured by phase-shifting digital holography. From four phase-shifted holograms of a cantilever recorded by a CCD image sensor, a complex amplitude at each pixel of the CCD plane is obtained using the conventional phase-shifting method. The complex amplitude of a cantilever is reconstructed from the complex amplitude of the CCD plane using the Fresnel diffraction integral. The phase difference distribution on the cantilever before and after deformation, i.e., the out-of-plane displacement distribution, is calculated. In order to decrease the effect of speckle noise, a new method using divided holograms is proposed. The theory and experimental results are shown.
Conference Paper
Fresnel or Fraunhofer hologram s can be recorded by CCD arrays as long as the sampling theorem is fulfilled. The numerical reconstruction of the wavefields is performed either by the discrete finite Fresnel transform or by solving the diffraction integral using the convolution theorem. The availability of the digitally evaluated complex wavefield instead of only the intensity field in optical reconstruction allows the subtraction of the phase distributions belonging to two states of the object to yield the sign-correct interference phase distribution with high accuracy. A numerical compensation for excess motion as well as effective filtering of the interference phase modulo 2(pi) are feasible.
Article
We present a noise reduction filter for the phase unwrapping of data from phase-shifting speckle pattern interferometry. This filter is effective in reducing the noise in wrapped phase maps without affecting the edges of the phase jumps, so that a simple and fast unwrapping algorithm can be used. We also discuss a data analysis strategy, including a routine for the recovery of unwrapping errors. The filter performance has been evaluated on real data from a fiber optic phase-shifting speckle interferometer, which we built for the analysis of out-of-plane deformations. The phase shifting is introduced by current modulation of the diode laser source. Some results relative to the analysis of very noisy phase measurements are presented.
Article
In speckle pattern interferometry the starting phase is distributed randomly. Furthermore the background intensity and the modulation also are structured stochastically. This leads to a significant amount of noise in the interferogram analysis. Low modulated speckles are more difficult to evaluate than high modulated ones. We report on a method to reduce this problem. The principle is to use more than one wavelength at the same time and to separate the different wavelengths in the recording plane. With this method the precision can be improved by factor up to 2.3. Our first experiments show the high potential of this method.
Article
A coherent optical imaging system consisting of recording a digital hologram by a CCD array and numerical reconstruction of the complex wave field in a computer is subjected to a frequency analysis. This analysis recognizes Fresnel and lensless Fourier transform holography; collimated and divergent reference waves; as well as the real image, the virtual image, and the dc term. The influences of finite sampling and the fill factor of the CCD pixels are examined. The impulse response of the system is a shifted Fraunhofer diffraction pattern of the aperture defined by the CCD. A fill factor below unity leads to a contrast decrease, which is quantitatively described in the modulation transfer function.
Article
The quantitative evaluation of fringe patterns obtained by holographic interferometry of a deformed surface is discussed in relation to the possible automatic image processing techniques that may be applied. The static and dynamic methods of determining the displacement vectors of the surface points from their holographic interference patterns are presented, and practical means of obtaining image data in a form suitable for automatic processing are examined. Major sources of interference pattern disturbances, including speckle noise, varying background intensity, and the presence of dust particles on the optical surfaces, are indicated. Digital image processing techniques which can be applied to holographic interferograms to suppress these disturbances are then presented, with attention given to grey-scale modification, smoothing, background correction, image enhancement by phase-shifted patterns, and geometric correction. It is noted that the above-mentioned techniques have been implemented on an eight-bit microprocessor system, yielding a considerable improvement of disturbed and degraded images without any a priori knowledge of the information contained in the original pattern.
Article
The noise in sawtooth fringes generated by electronic speckle pattern interferometry (ESPI) is investigated. When deformations of depolarising objects are studied, the scattered object light can be decomposed into two orthogonal linearly polarised speckle patterns which are partially decorrelated. Their correlation coefficient decreases with increasing depolarisation coefficient of the object. By suitable merging of the phase distributions of these two speckle fields on the basis of a modulation depth analysis, the rms phase error in the ESPI sawtooth fringes can be reduced significantly.