In this paper, a new 2step quadrature phaseshifting digital holographic optical encryption method using orthogonal polarization is proposed and tolerance errors for this method are analyzed. Unlike the conventional technique using a PZT mirror, the proposed optical setup comprises two input and output polarizers, and one λ/4plate retarder. This method makes it easier to get a phase shift of π/2 without using a mechanically driven PZT device for phaseshifting and it simplifies the 2step phaseshifting MachZehnder interferometer setup for optical encryption. The decryption performance and tolerance error analysis for the proposed method are presented. Computer experiments show that the proposed method is an alternate candidate for 2step quadrature phaseshifting digital holographic optical encryption applications.
I. INTRODUCTION
Optical systems have fast parallel information processing and rapid transmission of data. Therefore, optical methods have shown great potential in the field of security applications due to these advantages. Recently, various kinds of optical information processing methods have been investigated for cryptography and security systems.
[1

14]
In implementing an optical encryption system, the phase has been the most widely used function. However, because a photosensitive device like a chargecoupled device (CCD) cannot record the phase information directly, some innovative techniques to obtain the phase information have been studied in recent decades. Among recent methods the phaseshifting digital holographic technique that uses the CCD camera for direct recording of a hologram has an advantage of real time digital information processing, and it is possible to obtain the full complex phase information.
[15

18]
Especially the 2step phaseshifting interferometric technique is applied to cryptography.
[19

24]
In the previous papers
[20
,
21
,
24]
, we proposed an optical encryption system using 2step phaseshifting digital holography. In the proposed optical system, a PZT driven mirror was used to get a quadrature phase shift of π/2 in the reference beam path of the MachZehnder interferometer. Generally in the phaseshifting interferometric method, a phase shift is acquired by moving the PZT mirror generating an optical path difference in the reference beam path of the interferometer, and the accurate amount of phase shift is implemented by applying the proper voltage to the PZT element, which is attached to the mirror. However, this method has to control the accurate expansion or contraction of the PZT device in order to get the exact phase shift and this makes it difficult to get the accurate phase step in the phaseshifting interferometry due to this electrical and mechanical operation, and therefore phase shift error can occur.
In this paper, a new 2step quadrature phaseshifting digital holographic optical encryption method using orthogonal polarization is proposed, and the performance of the decryption for cryptography is evaluated. Tolerance errors for the proposed method are also analyzed. Unlike the conventional technique using the PZT mirror, the proposed optical setup employs simply two polarizers and one phase retarder to get the quadrature phaseshifting. A quarterwave plate(λ /4plate) is used as a phase retarder, which generates ppolarization interference without phase shift along the vertical direction and spolarization interference with phase shift of π/2 along the horizontal direction. These two interference patterns correspond to the two intensities resulting from the 2step quadrature phaseshifting digital holography. An input polarizer makes collimated light into a linear polarized wave. Another output polarizer, called the analyzer, performs separate recording on the CCD according to the polarization direction. Binary image or random generated binary bit data are used as input data to be encrypted, and a random generated binary bit code is used as a security key code for encryption and decryption. The encrypted Fourier transform hologram is obtained by use of a random phase mask pattern attached to a spatial light modulator (SLM), and a 256 graylevel quantized digital hologram is obtained by the CCD. In Section Ⅱ, the principle to acquire the π/2 phase shift using orthogonal polarization and the encryption and decryption process with the 2step phaseshifting digital holography are described. In Section Ⅲ, computer experiments show results of the decryption with the proposed method and the graph to analyze errors from
2step phaseshifting digital holographic optical encryption system: (a) by using the PZT mirror, (b) by using orthogonal polarization.
misalignment of optical components. Finally, brief conclusions are summarized in Section Ⅳ.
II. THEORY
 2.1. Principle
Figure 1
(a) shows the conventional schematic optical setup for the 2step phaseshifting digital holographic optical encryption system by using the PZT mirror, while
Fig. 1
(b) is the proposed schematic setup for the same 2step phaseshifting digital holographic optical encryption system by using orthogonal polarization. Schematically, the optical setup contains a MachZehnder type interferometer. Laser light is collimated by a spatial filter (SF) and a collimating lens (CL), and then passes through a linear polarizer (P1) whose polarization direction is 45° with respect to the horizontal axis. A beam splitter (BS1) divides the collimated light into two linearly polarized plane waves as the reference and the object beams. In the reference beam, the 45° linearly polarized light passes through a λ/4plate which sets the fastaxis along the vertical axis. Then, after passing through the λ/4plate, quadrature phase shift of π /2 occurs only on the horizontal axis. If we align an output analyzer (P2) whose polarization direction is set on the horizontal axis, a π/2 phase shifted reference beam is obtained on the horizontal axis. On the other hand, when we align the output analyzer (P2) whose polarization direction is set on the vertical axis, no phase shift occurs in the reference beam on the vertical axis. This scheme makes it possible to acquire 2step quadrature phaseshifting digital holograms with π/2 phase shift between spolarization and ppolarization on the CCD. Binary image or bit data to be encrypted is displayed on a SLM1 which is attached to a random phase mask, and is Fourier transformed on the CCD by a lens (L1) as the object beam, while a phasetype SLM2 which display a security key code is Fourier transformed on the CCD by a lens (L2) as the reference beam.
The encryption and decryption principle by using the 2step phaseshifting digital holographic method is described as follows.
[24]
Let 
o
(
x, y
) be binary data to be encrypted and exp[
jθ_{o}
(
x, y
) be a random phase mask, where
x
and
y
are transversal coordinates at the input spatial plane. The multiplication of the two is represented as
Let 
r
(
x, y
) be a binary code that can represent a security key code. This key code is multiplied by
π
(radian) to become a binary phase
θ_{r}
(
x, y
) =
π
⋅
r
(
x, y
) if 
r
(
x, y
) is 1 or 0, where
x
and
y
are transversal coordinates at the spatial plane. This random phase pattern with unit amplitude can be displayed on the phasetype SLM and is expressed as
Fourier transformed functions of the binary data and the security key code are supposed to be
O
(
α
,
β
) and
R
(
α
,
β
), where
α
and
β
are transversal coordinates at the spatial frequency plane. Then, the digial holoraphic intensity pattern recorded by the CCD at the spatial frequency plane is given by
where
Δϕ
=
ϕ_{O}

ϕ_{R}
is the phase difference between the object and the reference beams. The 2step quadrature phaseshifting digital holographic method gives two intensity patterns in the form of a digital hologram.
These two digital holograms are encrypted data. After a DCterm removal technique is applied, Eq. (4) is modified as
where
A
(
α
,
β
) is 
S
(
α
,
β
)
^{2}
+ 
R
(
α
,
β
)
^{2}
and
B
(
α
,
β
) is 2
S
(
α
,
β
)
R
(
α
,
β
).
Then, the phase difference of the object beam and the reference beam and the magnitude are calculated as follows.
From Eqs. (6) and (7), the complex hologram with encryption information is expressed as
By using this complex hologram and the security key, the reconstructed complex distribution and the original binary data is decrypted.
 2.2. Error Analysis
The proposed architecture comprises optical components such as a polarizer and a phase retarder. In order to acquire exactly a 2step quadrature phaseshifting interference pattern with π/2 phase shift between spolarization and ppolarization on the CCD, these optical components must be aligned in the right position without any angular error. For the proposed setup, the polarization direction of the input linear polarizer (P1) is set 45° rotated with respect to the horizontal axis and the fastaxis of the λ/4plate is aligned along the vertical axis. Also, the polarization direction of the output linear polarizer (P2) is aligned along the vertical axis for the case of recording ppolarization interference and the polarization direction of P2 is aligned along the horizontal axis for the case of recording spolarization interference. However, a small angular variation of these optical components can cause interference intensity variation error on the CCD.
Figure 2
shows schematic drawing of the proposed optical setup for error analysis of the polarization misalignment. The collimated laser light shown in
Fig. 1
is expressed as a monochromatic plane wave of angular velocity
ω
traveling in the zdirection with wavenumber
k
.
Schematic drawing of the proposed optical setup for error analysis on the polarization misalignment.
Coordinate system representation of the orthogonal linear polarization: (a) the xyaxis linear polarization components due to the misalignment of the input polarizer(P1) with error α , (b) the fast and slow axis linear polarization components due to the misalignment of the input polarizer(P1) with error α and the λ/4plate with error ϕ.
The electric field lies in the xyplane and is generally described by
where
and
ŷ
are unit vectors along the horizontal (xdirection) and vertical(ydirection) axis, respectively. Expressing
E_{x}
and
E_{y}
in terms of their magnitudes and phases,
E_{x}
=
A_{x}
cos(
wt

kz
) and
E_{y}
=
A_{y}
cos(
wt

kz
), we obtain
In
Fig. 2
, when the polarization direction of the input polarizer(P1) is set 45° with respect to the horizontal axis, then each magnitude in the x and y direction is
and
Meanwhile, if the polarizer P1 is misaligned with angle error
α
with respect to the
π
/4axis, the magnitude components along the x and y direction are given by
Figure 3
(a) shows a coordinate system representation of the xyaxis linear polarization components due to the misalignment of the input polarizer(P1) with
α
.
The second error occurs when we misalign the λ/4plate in the reference path. Considering the case where the fastaxis of the λ/4plate is misaligned with angle error
ϕ
with respect to the vertical axis, we obtain the fast axis and slow axis linear polarization components due to the misalignment of the polarizer P1 with
α
and the λ/4plate with
ϕ
.
Fig. 3
(b) shows a coordinate system representation of the fast and slow axis linear polarization components due to the misalignment.
The third error occurs due to the misalignment of the output linear polarizer(P2). When we intend to acquire
Coordinate system representation of the orthogonal linear polarization in the reference path: (a) the fast axis linear polarization components due to the vertical misalignment of the output analyzer(P2) with error β_{1}, (b) the slow axis linear polarization components due to the vertical misalignment of the output analyzer(P2) with error β_{1}, (c) the slow axis linear polarization components due to the horizontal misalignment of the output analyzer(P2) with error β_{2}, (d) the fast axis linear polarization components due to the horizontal misalignment of the output analyzer(P2) with error β_{2}.
ppolarization interference intensity on the CCD, the polarization direction of the polarizer P2 must be aligned along the vertical axis. On the other hand, when we intend to get spolarization interference intensity on the CCD, the polarization direction of P2 must be aligned along the horizontal axis. However, a small angular misalignment of this optical component can cause interference intensity variation error on the CCD.
Fig. 4
shows a coordinate system representation of the orthogonal linear polarization components considering the misalignment in the reference path. If the polarizer P2 is misaligned with angle error
β
_{1}
with respect to the vertical axis when we get ppolarization interference and if the polarizer P2 is misaligned with angle error
β
_{2}
with respect to the horizontal axis when we get spolarization interference, the magnitude components along the orthogonal polarization directions of the P2 are written by
and
Another error occurs when we consider the same misalignment of the output polarizer(P2) in the object path as in the reference path.
Fig. 5
shows a coordinate system representation of the orthogonal linear polarization components
Coordinate system representation of the orthogonal linear polarization in the object path: (a) the orthogonal linear polarization components due to the vertical misalignment of the output analyzer(P2) with error β_{1}, (b) the orthogonal linear polarization components due to the horizontal misalignment of the output analyzer(P2) with error β_{2}.
considering the misalignment in the object path. If the polarizer P2 is misaligned with the same angle errors
β
_{1}
and
β
_{2}
with respect to the vertical and horizontal axes when we get 2step interference intensities, the magnitude components along the orthogonal polarization directions of P2 are written by
In this paper, a digital hologram recorded on the CCD results from interference pattern of two Fourier transformed functions. According to a linear property of the Fourier transform theorem, the transform of a weighted sum of two functions is simply the identically weighted sum of their individual transforms. If
F
[f(x, y)] = F(
α
,
β
) and
F
[g(x, y)] =G(
α
,
β
) then F[sf(x,y) + tg(x, y)] = sF(
α
,
β
) + tG(
α
,
β
). Therefore, a small angular variation of the optical components in the proposed setup changes the magnitude components along the orthogonal polarization directions in the reference and the object paths. These small changes make a different referencetoobject beam ratio, which results in a small interference intensity variation on the CCD and can cause error.
First, we consider the object path. If an object function is represented as Eq. (1) and a monochromatic plane wave is expressed as Eq. (11) and (12), a light function after passing through the object function is defined by
When the misalignment error of the λ/4plate is
α
with respect to the
π
/4axis, we can write
where
A_{x}
and
A_{y}
are given by Eq. (13) and (14).
If the output polarizer(P2) is misaligned with angle error
β
_{1}
with respect to the vertical axis when we get ppolarization interference and if the polarizer P2 is misaligned with angle error
β
_{2}
with respect to the horizontal axis when we get ppolarization interference, then we can write
where
A_{y}
_
_{h}
,
A_{y}
_
_{v}
,
A_{x}
_
_{h}
and
A_{x}
_
_{v}
are given by Eqs. (25) ~ (28). By Fourier transforming Eqs. (32) ~ (35), we have
Second, we consider the reference path. If a reference function is represented as Eq. (2) and a monochromatic plane wave is expressed as Eq. (11) and (12), a light function after passing through the reference function is defined by
When the misalignment error of the input polarizer(P1) is
ϕ
with respect to the vertical axis, we can write
where
A_{s}
and
A_{f}
are given by Eq. (15) and (16). Quadrature phase shift of π/2 is represented in Eq. (41) which occurs only on the slow axis. By Fourier transforming Eq. (41) and (42), we have
Just like the object path case, if the output polarizer(P2) is misaligned with angle error
β
_{1}
with respect to the vertical axis when we get ppolarization interference and if the polarizer P2 is misaligned with angle error
β
_{2}
with respect to the horizontal axis when we get ppolarization interference, then we can write
and
where
A_{y}
_
_{s,h}
,
A_{y}
_
_{s,v}
,
A_{y}
_
_{f,h}
,
A_{y}
_
_{f,v}
,
A_{x}
_
_{s,h}
,
A_{x}
_
_{s,v}
,
A_{x}
_
_{f,h}
, and
A_{x}
_
_{f,v}
are given by Eqs. (17) ~ (24).
Now, considering these misalignment factors in Eq. (3) and (4), the 2step quadrature phaseshifting digital holographic intensity pattern recorded by the CCD can be expressed as
where
I
_{1}
is achieved when we get ppolarization interference along the vertical direction and
I
_{2}
is achieved when we get spolarization interference along the horizontal direction. Then, Eq. (53) and (54) are rewritten by
where
Δ_{ϕ}
=
ϕ_{O}

ϕ_{R}
is the phase difference between the object and the reference beams.
With Eqs. (55) and (56), if we let error terms like as
and apply DCterm removal technique, Eqs. (55) and (56) is expressed as
From Eqs. (20), (21), (26) and (27), 
A_{y_v}

A_{y_f,v}
≅
A_{x_h}

A_{x_s,h}
 for small
α
,
ϕ
,
β
_{1}
and
β
_{2}
. Then the phase difference of object beam and reference beam and the magnitude are calculated as follows.
However, because we cannot exclude the error terms
E
_{1}
(
α
,
β
) and
E
_{2}
(
α
,
β
) in the Eqs. (61) and (62) when we record interference intensities on the CCD, we can only acquire two intensities of Eqs. (59) and (60) containing error. With these two digital holograms, the calculated phase difference and the magnitude with errors are given by
where
ϕ
_{e}
and
A
_{e}
stand for the errors in the phase difference and the magnitude,
I
_{1}
'
and
I
_{2}
'
are expressed as Eqs. (6) and (7). So, a complex hologram which has the error from Eqs. (57) and (58) can be expressed as
By using this complex hologram and the security key, the reconstructed complex distribution and the decrypted binary data are obtained by
where
e
_{1}
(
α
,
β
) and
e
_{2}
(
x
,
y
) are the generated errors.
Binary input data for computer simulations(256×256 pixels): (a) binary image data, (b) a random generated binary bit data, (c) a random generated binary bit code as a security key.
If there are no misalignment errors of the optical components, i.e.
α
=0
ϕ
=0
β
_{1}
=0 and
β
_{2}
=0 then
E
_{1}
(
α
,
β
)=0,
E
_{2}
(
α
,
β
)=0 given by the Eqs. (61) and (62). Therefore,
e
_{1}
(
α
,
β
)=0 and
e
_{2}
(
x
,
y
)=0 in the Eqs. (66), (67) and (68).
In the proposed system the number of error pixels between the original data and the decrypted data is defined as
where
N_{X}
,
N_{Y}
are the entire pixel number.
III. COMPUTER EXPERIMENTS
Computer simulations show the decryption performance and error analysis of the proposed 2step quadrature phaseshifting digital holographic optical encryption method using orthogonal polarization. Binary image and random generated binary bit data of size 256×256 pixels shown in
Figs. 6
(a) and (b) are used as input data to be encrypted, and
Fig. 6
(c) shows a random generated binary bit code as a security key code for encryption and decryption.
Figure 7
shows an example of the decrypted binary image data for the misalignment of optical components.
Fig. 7
(a) is the decrypted binary image when the misalignment error of the input polarizer(P1) is 5 degrees. In this case, the reconstructed image is the same as the original image and is not affected by the small misalignment of the input polarizer.
Fig. 7
(b) is the decrypted binary image when the misalignment error of the λ/4plate is 5 degrees. Similarly, the reconstructed image is the same as the original image and is not affected by the small misalignment of the λ/4plate.
Figs. 7
(c) and (d) are the decrypted binary images when the misalignment error of the output polarizer(P2) is
An example of the decrypted binary image data for the misalignment of optical components: (a) when the misalignment error of the input polarizer(P1) is 5 degrees, (b) when the misalignment error of the λ/4plate is 5 degrees, (c) when the misalignment error of the output analyzer(P2) is 5 degrees on the vertical axis, (d) when the misalignment error of the output analyzer(P2) is 5 degrees on the horizontal axis.
Error graphs of the decrypted binary bit data according to the misalignment of optical components: (a) when the misalignment error of the input polarizer(P1) is 0 to 10 degrees, (b) when the misalignment error of the λ/4plate is 0 to 5 degrees, (c) when the misalignment error of the output analyzer(P2) is 0 to 5 degrees on the vertical and horizontal axis, (d) the magnified graph of the FIG. 8(c) when the misalignment error of the output analyzer(P2) is 0 to 4 degrees on the vertical and horizontal axis.
5 degrees on both the vertical and horizontal axes. However, in this case, the reconstructed image is affected by even the 5 degrees misalignment of the output polarizer and gets damaged in several regions. For some example regions, circular areas shown in
Figs. 7
(c) and (d) include such an error in the decrypted image. In these circular areas, some dark pixels instead of the correct white pixels and some white pixels instead of the correct dark pixels compared to the reconstructed original image shown in
Figs. 7
(a) and (b) represent error pixels.
In order to know how many pixels are damaged in the decrypted data, the mismatching error to the original data is investigated and analyzed according to the misalignment of optical components. The random generated binary bit data shown in
Fig. 6
(b) is used as input data to be encrypted for convenience. All graphs shown in
Fig. 8
are an average value obtained by 100 times evaluations to the random generated binary data(256×256 pixels).
Fig. 8
(a) is the error graph when the misalignment error of the input polarizer(P1) is from 0 to 10 degrees, where the number of error pixels is less than 1 pixel. This means that the small misalignment of the input polarizer(P1) does not affect the decryption data.
Fig. 8
(b) is the error graph when the misalignment error of the λ/4plate is from 0 to 5 degrees, where the number of error pixels is also less than 1 pixel. This means that the small misalignment of the λ/4plate does not affect the decryption data seriously.
Fig. 8
(c) is the error graph when the misalignment error of the output polarizer(P2) is from 0 to 5 degrees, where the number of error pixels is about 70 pixels for the case of 5 degrees misalignment and is above 2 pixels even if the case of 3 degrees misalignment.
Fig. 8
(d) shows the magnified graph of the result. This means that the small misalignment of the output polarizer(P2) does affect mainly the decryption data error in the proposed setup. However, the number of error pixels is less than 1 pixel if we control the misalignment error of the polarizer P2 less than 2 degrees. The alignment of the optical components within this angle can be achieved in the setup.
IV. CONCLUSIONS
In this paper, a new 2step quadrature phaseshifting digital holographic optical encryption method using orthogonal polarization is proposed. The π/2 phase shift of the 2step quadrature phaseshifting digital holography is achieved by constructing a MachZehnder interferometer with two polarizers and one λ/4plate retarder in the proposed optical setup. This scheme provides a good decryption performance which achieves the same result as the conventional phaseshifting method using the PZT mirror, and has advantages of compactness and easy configuration of the optical system. The error analysis of the proposed method shows that even small misalignment of the output polarizer is the main error source in this system, meanwhile small misalignments of the input polarizer and λ/4plate do not affect the decryption performance. However, the maximum tolerance error of the output polarizer is about 2 degrees and this amount is acceptable and controllable in the optical setup. Computer experiments verified that the proposed method is a good alternate for cryptography for security applications.
Javidi B.
,
Horner J. L.
(1994)
“Optical pattern recognition for validation and security verification”
Opt. Eng.
33
1752 
1756
Heanue J. F.
,
Bashaw M. C.
,
Hesselink L.
(1995)
“Encrypted holographic data storage based on orthogonalphasecode multiplexing”
Appl. Opt.
34
6012 
6015
Refregier P.
,
Javidi B.
(1995)
“Optical image encryption based on input plane and Fourier plane random encoding”
Opt. Lett.
20
767 
769
Javidi B.
,
Sergent A.
,
Ahouzi E.
(1998)
“Performance of double phase encoding encryption technique using binarized encrypted images”
Opt. Eng.
37
565 
569
Weber D.
,
Trolinger J.
(1999)
“Novel implementation of nonlinear joint transform correlators in optical security and validation”
Opt. Eng.
38
62 
68
Cuche E.
,
Bevilacqua F.
,
Depeursinge C.
(1999)
“Digital holography for quantitative phasecontrast imaging”
Opt. Lett.
24
291 
293
Javidi B.
,
Nomura T.
(2000)
“Securing information by means of digital holography”
Opt. Lett.
25
28 
30
Unnikrishnan G.
,
Singh K.
(2000)
“Double random fractional Fourier domain encoding for optical security”
Opt. Eng.
39
2853 
2859
Lin G.S.
,
Chang H. T.
,
Lie W.N.
,
Chuang C.H.
(2003)
“Publickeybased optical image cryptosystem based on data embedding techniques”
Opt. Eng.
42
2331 
2339
Arizaga R.
,
Torroba R.
(2003)
“Validation through a binary key code and a polarization sensitive digital technique”
Opt. Commun.
215
31 
36
Nomura T.
,
Okazaki A.
,
Kameda M.
,
Morimoto Y.
(2005)
“Image reconstruction from compressed encrypted digital hologram”
Opt. Eng.
44
2313 
2320
Lee H. J.
,
Gil S. K.
(2006)
“Error analysis for optical security by means of 4step phaseshfting digital hologaphy”
J. Opt. Soc. Korea
10
118 
123
Gil S. K.
,
Jeon S. H.
,
Kim N.
,
Jeong J. R.
(2006)
“Successive encryption and transmission with phaseshifting digital holography”
Proc. SPIE
6136
339 
346
Jeon S. H.
,
Gil S. K.
(2010)
“QPSK modulation based optical image cryptosystem using phaseshifting digital holography”
J. Opt. Soc. Korea
14
97 
103
Hariharan P.
(1987)
“Digital phaseshifting interferometry: a simple error compensating phase calculation algorithm”
Appl. Opt.
26
2504 
2506
Yamaguchi I.
,
Zhang T.
(1998)
“Phaseshifting digital holography”
Opt. Lett.
22
610 
612
Jeong M.O.
,
Kim N.
,
Park J.H.
(2008)
“Elemental image synthesis for integral imaging using phaseshifting digital holography”
J. Opt. Soc. Korea
12
275 
280
Liu J.P.
,
Poon T.C.
,
Jhou G.S.
,
Chen P.J.
(2011)
“Comparison of two, three, and fourexposure guadrature phaseshifting holography”
Appl. Opt.
50
2443 
2450
Meng X.
,
Cai L. Z.
,
Yang X. L.
,
Shen X. X.
,
Dong G. Y.
,
Wang Y. R.
(2006)
“Twostep phaseshifting interferometry and its application in image encryption”
Opt. Lett.
31
1414 
1416
Gil S. K.
,
Byun H. J.
,
Lee H. J.
,
Jeon S. H.
,
Jeong J. R.
(2007)
“Optical encryption of binary data information with 2step phaseshifting digital holography”
Proc. SPIE
6488
648812 
Jeon S. H.
,
Hwang Y. G.
,
Gil S. K.
(2008)
“Optical encryption of graylevel image using onaxis and 2f digital holography with twostep phaseshifting method”
Opt. Rev.
15
181 
186
Awatsuji Y.
,
Tahara T.
,
Kaneko A.
,
Koyama T.
,
Nishio K.
,
Ura S.
,
Kubota T.
,
Matoba O.
(2008)
“Parallel twostep phaseshifting digital holography”
Appl. Opt.
47
D183 
D189
Lee H.C.
,
Kim S.H.
,
Kim D.S.
(2010)
“Twostep onaxis digital holography using dualchannel MachZehnder interferometer and matched filter algorithm”
J. Opt. Soc. Korea
14
363 
367
Jeon S. H.
,
Gil S. K.
(2011)
“2step phaseshifting digital holographic optical encryption and error analysis”
J. Opt. Soc. Korea
15
244 
251