ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69

Contents lists available at ScienceDirect

ISPRS Journal of Photogrammetry and Remote Sensing

journal homepage: www.elsevier.com/locate/isprsjprs

A dual quaternion-based, closed-form pairwise registration algorithm

for point clouds
Yongbo Wang ⇑, Yunjia Wang, Kan Wu, Huachao Yang, Hua Zhang
Key Laboratory for Land Environment and Disaster Monitoring of SBSM, Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of
Mining and Technology, Xuzhou, Jiangsu 221008, China

a r t i c l e i n f o a b s t r a c t

Article history: The representation of similarity transformation in three-dimensional (3D) space, especially of orienta-
Received 30 August 2013 tion, is a crucial issue in navigation, geodesy, photogrammetry, robot arm manipulation, etc. Considering
Received in revised form 31 March 2014 the large amount of computer resources required by iterative algorithms designed for spatial similarity
Accepted 15 April 2014
transformation, the high dependence on initial values of unknown parameters, and the instability of solv-
Available online 14 May 2014
ing transformation parameters for large-angle registration, a closed-form solution for pairwise light
detection and ranging (LiDAR) point cloud registration is proposed. In this solution, dual-number quater-
Keywords:
nions are used to represent the 3D rotation. The relationship between the rotation matrix-based repre-
LiDAR
Registration
sentation of similarity transformation and the dual quaternion-based representation is described first.
Similarity transformation Considering that the same features from two neighboring stations coincide after pairwise registration,
Rigid transformation a dual quaternion-based error norm, which is associated with the sum of the position errors, is con-
Dual quaternions structed. Based on theory of least squares and by extreme value analysis of the error norm, detailed der-
Geodetic datum ivations of the model and the main formulas are obtained. Once the similarities between the same
features from the two neighboring LiDAR stations are constructed, the rotation matrix, the scale param-
eter, and the translation vector are simultaneously derived. Two experiments are conducted to verify the
feasibility and effectiveness of the proposed algorithm. The proposed algorithm has the advantages of
simplicity and ease of implementation, making it better than the traditional methods that use matrices
to describe spatial rotation. Moreover, it solves the transformation parameters without the initial esti-
mates of unknown parameters, making it better than iterative algorithms. Most importantly, in contrast
to unit quaternion-based algorithms, the proposed algorithm solves seven unknown parameters simul-
taneously. Therefore, it effectively avoids the accumulation of introduced error in calculation and the
negative impact from the inappropriate choice of initial values.
Ó 2014 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier
B.V. All rights reserved.

1. Introduction
Reconstructing geographical entities and their neighborhoods
efficiently and accurately is a key problem in digital city and is also
a core research topic in three-dimensional (3D) geographic infor-
mation system technology. Since the former US vice president Al
Core introduced the concept of digital earth in 1998, the seamless
3D representation of large amounts of available geographical
entities in a computer has been a popular topic of research. Among
the newly developed instruments designed for the acquisition of
location-based data, terrestrial light detection and ranging (LiDAR)
⇑ Corresponding author. Tel.: +86 136 55208823.
E-mail addresses: ybwang816@163.com (Y. Wang), wyj4139@cumt.edu.cn
(Y. Wang), wukan6899@263.net (K. Wu), huachao-yang@163.com (H. Yang),
zhhua_79@163.com (H. Zhang).
has been the focus of considerable attention because of its instan-
taneous and direct acquisition of point cloud. However, given the
high spatial complexity of geographical entities and the limited
field of LiDAR’s vision, several stations should be established to
fully acquire the detailed point cloud of some entities of interest.
Therefore, similarity transformation is used to register point clouds
from different stations and integrate point clouds.
Given the matched points, i.e. pairs of conjugate points, from
the two neighbor point clouds as input, many registration
approaches have been proposed to calculate the transformation
parameters from the nonlinear overdetermined equations in the
least-squares sense. Presently, the most frequently used model is
the similarity transformation model with seven parameters (i.e.,
one scale factor, three translation parameters, and three rotation
angles). It is also known as the Helmert transformation or the
Bursa model, which is used in this study. By using different

http://dx.doi.org/10.1016/j.isprsjprs.2014.04.013
0924-2716/Ó 2014 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved.
64 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69
solutions to calculate transformation parameters, available regis-
tration approaches can be divided into two categories, namely,
analytical methods (Horn, 1987, 1988; Walker et al., 1991;
Daniilidis, 1999; Prošková, 2011, 2012) and iterative methods
(Besl and Mckay, 1992; Zheng et al., 2008). The main difference
between these two categories lies in the requirement of initial esti-
mates of the unknown transformation parameters before registra-
tion. Considering that the Bursa model is a nonlinear model, the
linearization of multivariate function is necessary for iterative
methods. However, it may lead to the loss of accuracy to some
extent, and the convergence of the linearized model may be
affected by the closeness of the initial estimates and their corre-
sponding true values. Furthermore, the Bursa model is difficult to
implement in cases of large rotation angles because the initial esti-
mates are difficult to obtain in some cases, such as absolute orien-
tation in photogrammetry. By contrast, analytical methods may
effectively avoid such a case. By using an optimization process,
the analytical method directly calculates the similarity transforma-
tion parameters. The difference between analytical methods and
iterative methods mainly lies in the representation of the 3D rota-
tion. Currently, two operators used in analytical registration meth-
ods, namely, orthonormal matrices (Horn, 1988) and quaternions
(Horn, 1987), are the two prominent representatives of vector
algebra and Clifford algebra. Orthonormal matrices provide a
closed-form solution by singular value decomposition (SVD), and
quaternions provide a closed-form solution by minimizing a cost
function. Compared with matrix-based methods, unit quaternion-
based methods use only four parameters, which is less than the
six parameters used in matrix-based methods, to represent spatial
rotation. Therefore, the demand for computer resources will signif-
icantly decrease. This decreased demand has been the focus of con-
siderable attention worldwide.
The disadvantages of unit quaternion-based methods lie in the
fact that the rotation parameters and displacement parameters
must be calculated in turn, but the calculation may cause error
accumulation in certain cases. To overcome the probable error
accumulation in the process of asynchronous calculation of trans-
formation parameters, dual quaternions are used to describe the
spatial rotation and to ensure the simultaneous calculation of reg-
istration parameters. The representative work on dual quaternions
was first introduced by Walker et al. (1991). However, the scale
parameter in the process of transformation was not considered.
Therefore, the scale parameter cannot be directly introduced into
the coordinate transformation in geodesy, absolute orientation in
photogrammetry, and registration of LiDAR points, which are the
main focus of this paper. To compensate for this deficiency, we
introduce the scale parameter in the process of similarity transfor-
mation and provide the formula derivation process in detail. We
also conduct a comparative analysis of the new algorithm based
on two experiments.
The remainder of the paper is organized as follows. Section 2
briefly reviews the representative work of iterative methods and
closed-form methods. Section 3 introduces the concept and prop-
erties of dual quaternions and then derives the optimization func-
tion based on the Bursa model. The step-by-step and detailed
derivation of the formula is also provided. In Section 4, two exper-
iments are described: the first one verifies the correctness of the
new algorithm and the second one verifies its feasibility in the reg-
istration of LiDAR point clouds. Section 5 concludes the paper.
2. Literature review
Most available solutions to the pairwise registration are based
on Chasles’ theorem, which states that any rotation and translation
can be expressed as a translation along a line and a rotation around
that line (Chasles, 1830). Accordingly, available solution to the
pairwise registration problem can be categorized into two, namely,
iterative methods and closed-form methods. The iterative method
usually uses the rotation matrix to represent rigid transformation
and needs the initial approximate values of transformation for lin-
earization, which often introduces error in calculation. By contrast,
the closed-form method does not need initial approximate values
and avoids linearization, which has been the focus of considerable
attention recently.
2.1. Closed-form solution to the pairwise registration problem
The most important advantage of the closed-form solution is
that it provides the best possible transformation in one step. In
contrast to the iterative approach, the closed-form solution does
not need to determine good initial estimates. Representatives of
closed-form solutions to the pairwise registration problem were
independently proposed by Arun et al. (1987), Horn (1987,1988),
and Haralick et al. (1989), which differ with respect to transforma-
tion representation and alternative ways of minimizing a criterion
function. Arun et al. (1987) used the rotation matrix to represent
spatial transformation and determined the least squares solution
by computing the SVD of the derived matrix (Haralick et al.,
1989). Similarly, Horn used orthonormal matrices to represent
rotation. After the computation of the positive semidefinite square
root of a positive semidefinite matrix, a solution for the rotation
matrix is derived through the direct manipulation of 3  3 matri-
ces (Horn, 1987). The third study, which is also similar to that of
Horn, used unit quaternions to represent rotations. The solutions
for the desired quaternion were the eigenvector of a symmetric
4  4 matrix associated with the largest positive eigenvalue
(Horn, 1988). The fourth study is that of Walker et al. (1991), which
used dual quaternions to represent transform components. By
minimizing a single-cost function associated with the sum of the
orientation and position errors, the rotation matrix and the trans-
lation vector are derived simultaneously.
These closed-form pairwise techniques are theoretically equiv-
alent, but they differ in the way the problem is formulated. An
experimental comparison found that the differences were negligi-
ble in practical applications with nondegenerate data (Eggert et al.,
1997).
2.2. Quaternion-based registration methods
Quaternions were invented by W.R. Hamilton to represent a 3D
vector (Hamilton, 1844), which is a convenient way to describe
rotation on a unit sphere. After Horn successfully applied quater-
nions in absolute orientation (Horn, 1987), their compactness in
describing the rotation matrix and the high efficiency of this
description became the focus of considerable attention. Currently,
quaternions have been successfully used in rigid body motion anal-
ysis (Joseph and Laviola, 2003; Kim and Golnaraghi, 2004) and
datum transformation in geodesy (Yang, 1999; Shen et al., 2006;
Zeng and Yi, 2011; Li et al., 2012). However, with unit quaternions
representing rotation in space, seven transformation parameters
are calculated in turn, that is, rotation angles first, scale parameters
later, and translation vectors last. In the case of introducing error in
calculating the rotation angles, the error may be amplified using
the following two steps. To overcome the aforementioned limita-
tion, Walker et al. (1991) introduced dual quaternions to represent
similarity transformation. A single-cost function is formulated,
with the real part and the dual part of the dual quaternion used
to represent rotation and translation, which enable the simulta-
neous calculation of six parameters (Walker et al., 1991), namely,
three rotation angles and three translation parameters. A similar
work was conducted by Daniilidis (1999). However, the studies
 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69 65

of Walker et al. (1991) and Daniilidis (1999) do not consider the
scale parameter in the process of transformation.
3. Dual quaternion-based registration model
3.1. Dual quaternions
A dual quaternion has eight elements, whereas a 3D object transfor-
mation has six independent variables, indicating that two of the
eight elements in the dual quaternion representation are not inde-
pendent. As shown in Eqs. (5) and (6), the components of any dual
quaternion satisfy the following two constraints:
r_ T r_ ¼ 1; ð7Þ

Dual quaternions are composed of quaternions and dual num-

r_ T s_ ¼ 0: ð8Þ
bers expressed as follows:
_
q ¼ r_ þ es_ ; ð1Þ 3.2. Error norm and its minimization
where r_ and s_ are both real quaternions and are called the real part
and the dual part, respectively. Dual quaternions are similarly inter- The feature points of the reference station and unregistered sta-
preted as real quaternions: tion are represented by f~ p0i g, respectively, and integer
p0i g and f~
2_ 3 NðN P 3Þ denotes the number of corresponding feature point pairs.
cos 2h Based on the similarities in the feature points between the refer-
_ 6 7
q¼4 _ _ 5; ð2Þ ence station and the unregistered station, the error norm for the
sin 2h n similarity measurement of the proposed registration algorithm
_ can be expressed as follows:
where the dual vector n represents a line in 3D space_from which
the coordinate system is rotated and translated, and h is the dual X
N
Fðl; R;~
tÞ ¼ pi þ ~
klR~ p0i k:
t ~ ð9Þ
angle of rotation and translation.
_
Detailed descriptions of the dual i¼1
_
vector n and dual angle h are expressed as follows:
_
The location quaternion of a feature point ~
p is
n þ e~
n ¼~ p ~
n; ð3Þ  
1 0
p_ ¼ : ð10Þ
_ 2 ~
p
h ¼ h þ ed; ð4Þ
Based on Eq. (5), the real part of dual quaternions fully represents
where ~n is a unit vector that specifies the direction of the rotation the whole process of rotation, and the corresponding rotation
axis and the direction of the translation, with the rotation involving matrix is expressed as follows:
the line with direction ~n passing through point P with a rotation
angle of h, and d is the distance of translation along the direction R ¼ ðr 20  ~
r T~
rÞI þ 2~rT þ 2r 0 Kð~
r~ rÞ; ð11Þ
specified by ~n. Comparing the unit quaternions with the dual qua-
2 3
ternion representation, the same transformation can be formed by 0 r 3 r2
first translating the original coordinate frame along the direction where Kð~rÞ ¼ 4 r 3 0 r1 5, which is a skew-symmetric
of ~
n by a distance of d and then rotating it by an angle of h with matrix. r2 r 1 0
respect to a line having a unit vector ~n as its direction and passing Eq. (11) can be expressed as a matrix:
through point P. " #
R ~
0 T
Substituting Eqs. (3) and (4) into Eq. (2), we obtain the follow- ¼ Wðr_ Þ Q ðr_ Þ; ð12Þ
ing equations (see Fig. 1): ~
0T 1
   
" # r0 rT
~ r0 rT
~
cos 2h where Wðr_ Þ ¼ and Q ðr_ Þ ¼ .
r_ ¼ h  ; ð5Þ ~r r0 I  Kð~
rÞ ~r ~
r0 I þ KðrÞ
sin 2 ~n
Translation vector ~
t can be expressed with dual quaternions as
"  h # follows:
 d sin 2
s_ ¼ h 2  : ð6Þ T
d
2
n þ sin 2h ð~
cos 2 ~ p ~
nÞ t_ ¼ Wðr_ Þ s_ ; ð13Þ

where t_ T ¼ 2 0 ~ t T , which is the corresponding form of vector ~
1
t.
Assuming that p_ 0i is the location quaternion of unregistered fea-
ture points and p_ i is the new location of quaternion after registra-
tion, the dual quaternion-based registration process can be
expressed as follows:
T T
p_ i ¼ Wðr_ Þ s_ þ lWðr_ Þ Q ðr_ Þp_ 0i : ð14Þ

Therefore, the sum of the square distances between p_ i and p_ 0 is

2
fi ¼ ðp_ i  p_ 0i Þ : ð15Þ

Substituting Eq. (14) into Eq. (15), we derive the following

equation:
2 T T 2
fi ¼ ðp_ i  p_ 0i Þ ¼ ðWðr_ Þ s_ þ lWðr_ Þ Q ðr_ Þp_ 0i  p_ 0i Þ
T 2 T 2 T T T T
¼ ½Wðr_ Þ s_  þ ½lWðr_ Þ Q ðr_ Þp_ 0i  þ ðp_ 0i Þ ðp_ 0i Þ þ 2l½Wðr_ Þ s_  ½Wðr_ Þ Q ðr_ Þp_ 0i  :
T T T T
2ðp_ 0i Þ ½Wðr_ Þ s_   2lðp_ 0i Þ ½Wðr_ Þ Q ðr_ Þp_ 0i 
Fig. 1. Dual quaternion-based rotation and translation. ð16Þ
66 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69

Then, each item of Eq. (16) can be further expressed as follows: Using Eq. (26), scale parameter l can be represented as the function
2 of r_ and s_ as follows:
T T T T
½Wðr_ Þ s_  ¼ ½Wðr_ Þ s_ ½Wðr_ Þ s_  ¼ Wðr_ Þ Wðr_ Þs_ T s_ ¼ ½r_ T r_ s_ T s_
r_ T C1 r_ þ s_ T C2 r_
¼ s_ T s_ ; ð17Þ l¼ : ð30Þ
2C 4
T 2 T T Substituting Eq. (29) into Eq. (30), we obtain the following
½lWðr_ Þ Q ðr_ Þp_ 0i  ¼ l2 ½Wðr_ Þ Q ðr_ Þp_ 0i ½Wðr_ Þ Q ðr_ Þp_ 0i 
equation:
T T
¼ l2 ½Wðr_ Þ Wðr_ Þ½Q ðr_ Þ Q ðr_ Þ½p_ 0T _0
i pi 
h i
l _T T 1 _T T
2 r_ T C1 r_  2N r C2 C2 r_ þ 2N r C3 C2 r_
_T _
¼ l ½r r I½r _T _rI½p_ 0T _0
i pi  ¼l 2 _ 0T _ 0
pi pi ; ð18Þ l¼ : ð31Þ
2C 4
T T T T T
2ðp_ 0i Þ ½Wðr_ Þ s_  ¼ 2ðp_ 0i Þ Wðr_ Þ s_ ¼ 2ðWðr_ Þp_ 0i Þ As C2 is skew-symmetric, we derive the following equation:
T T
¼ 2ðQ ðp_ 0i Þr_ Þ s_ ¼ 2r _T Q ðp_ 0i Þ s_ CT2 C2 ¼ C 22 E; ð32Þ
¼ 2s_ T Q ðp_ 0i Þr_ ; ð19Þ
where E is a unit matrix.
T T T
Using Eqs. (32), Eq. (31) can be further expressed as
2ðp_ 0i Þ ½ lWðr_ Þ Q ðr_ Þp_ 0i  ¼ 2l ðWðr_ Þp_ 0i Þ ½Wðp_ 0i Þr_ 


1 T T C 22
T
¼ 2lðQ ðp_ 0i Þr_ Þ ½Wðp_ 0i Þr_  l ¼ r_ T C1 r_ þ r_ C3 C2 r_ = 2C 4  : ð33Þ
2N 2N
T
¼ 2lr _T Q ðp_ 0i Þ Wðp_ 0i Þr_ : ð20Þ Substituting Eq. (29) into Eq. (24), we obtain the following
Substituting Eqs. (17)–(20) into Eq. (16), we derive the following equation:
equation:     
@F 1 1
¼ lðC1 þ CT1 Þ  lCT2 ðlC2 þ C3 Þ  CT3 ðlC2 þ C3 Þ þ 2k1 r_
fi ¼ ðp_ i  p_ 0i Þ
2 @ r_ 2N 2N
T ¼ 0:
¼ s_ T s_ þ l2 p_ 0T _0 _0 _0 _T _0 _ _T _ 0 _
i pi þ ðpi Þ ðpi Þ þ 2ls Wðpi Þr  2s Q ðpi Þr
T ð34Þ
 2lr_ T Q ðp_ 0i Þ Wðp_ 0i Þr_ ; ð21Þ
P P P As C3 is skew-symmetric, we derive the following equation:
T
making C1 ¼ 2 Ni¼1 Q ðp_ 0i Þ Wðp_ 0i Þ, C2 ¼ 2 Ni¼1 Wðp_ 0i Þ, C3 ¼ 2 Ni¼1
P P CT3 C3 ¼ C 33 E: ð35Þ
Q ðp_ 0i Þ, C 4 ¼ Ni¼1 ð~ pi Þ, and C 5 ¼ Ni¼1 ð~
pTi ~ p0T ~0
i pi Þ. The similarity measure-
ment, as shown in Eq. (9), can be rewritten in the form of the Using Eqs. (32) and (35), Eq. (34) can be simplified as follows:
quadratic function of quaternions r_ and s_ : 
@F 1 2 1 1 1
F ¼ lr_ T C1 r_ þ Ns_ T s_ þ ls_ T C2 r_ þ s_ T C3 r_ þ l2 C 4 þ C 5 : ð22Þ ¼ lðC1 þ CT1 Þ  l C 22 E  l CT2 C3  l CT3 C2  C 33 E þ 2k1 r_
@ r_ 2N 2N 2N 2N

Using Eqs. (7) and (8) as restrictions, the best dual quaternions for 1 2 1 1 T T
¼  l C 22 E  C 33 E  lC2 C3 þ lðC1 þ C1 Þ þ 2k1 r_
representing rotation can be calculated by minimizing following 2N 2N N
  
error function: 1 l 1 T
¼2  2
ðl C 22 þ C 33 ÞE  C2 C3  ðC1 þ CT1 Þ þ k1 r_ ¼ 0; ð36Þ
4N 2 N
F ¼ lr_ T C1 r_ þ Ns_ T s_ þ ls_ T C2 r_ þ s_ T C3 r_ þ l2 C 4 þ C 5 þ k1 ðr_ T r_  1Þ þ k2 ðs_ T r_ Þ;
ð23Þ making
 
where k1 and k2 are Lagrange multiplier constants. Taking the par- 1 l 1 T
A¼ ðl2 C 22 þ C 33 ÞE þ C2 C3  ðC1 þ CT1 Þ : ð37Þ
tial derivatives, we derive the following equations: 4N 2 N

@F Eq. (36) can be rewritten as

¼ lðC1 þ CT1 Þr_ þ lCT2 s_ þ CT3 s_ þ 2k1 r_ þ k2 s_ ¼ 0; ð24Þ
@ r_ Ar_ ¼ k1 r_ : ð38Þ

@F Based on Eq. (38) and according to Walker et al. (1991), the quater-
¼ 2Ns_ þ lC2 r_ þ C3 r_ þ k2 r_ ¼ 0; ð25Þ
@ s_ nion r_ is an eigenvector of the matrix A and k1 is the corresponding
eigenvalue. In general, four solutions are derived for this equation.
@F Considering that A is real and symmetric, all the eigenvalues and
¼ r_ T C1 r_ þ s_ T C2 r_ þ 2lC 4 ¼ 0: ð26Þ
@l eigenvectors are real and all the eigenvectors are orthogonal. The
desired solution is identified by referring back to the original error
Therefore, the solution of Eqs. (7), (8), (24), (25), and (26) for r_ and s_ shown in Eq. (22).
provides the optimal solution for the registration parameters. Multiplying Eq. (24) by r_ T , we obtain the following equation:
Multiplying Eq. (25) with r_ T , we obtain the following equation:
1 1
k2 ¼ r_ T ðlC2 r_ þ C3 r_ Þ: ð27Þ
lr_ T C1 r_ ¼  lr_ T CT2 s_  r_ T CT3 s_  k1 : ð39Þ
2 2

Considering that C2 and C3 are both skew-symmetric matrices, we Then, multiplying Eq. (25) by s_ T , we obtain:
obtain the following equation: 1
Ns_ T s_ ¼  ðls_ T C2 r_ þ s_ T C3 r_ Þ: ð40Þ
k2 ¼ r_ T ðlC2 r_ þ C3 r_ Þ ¼ 0: ð28Þ 2
Substituting Eqs. (39) and (40) into Eq. (22) gives
Therefore, we can represent the dual quaternion s_ as the function of
r_ as follows: F ¼ l2 C 4 þ C 5  k1 : ð41Þ
1 Clearly, the error will be minimized if we select the eigenvector
s_ ¼  ðlC2 r_ þ C3 r_ Þ: ð29Þ
2N corresponding to the largest positive eigenvalue.
 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69 67

Table 1
Coordinate pairs extracted from the local and WGS84 coordinate systems.

Site Local coordinate system (System A) WGS84 coordinate system (System B)

x y z X Y Z
Solitude 4157222.543 664789.307 4774952.099 4157870.237 664818.678 4775416.524
Buoch Zeil 4149043.336 688836.443 4778632.188 4149691.049 688865.785 4779096.588
Hohenneuffen 4172803.511 690340.078 4758129.701 4173451.354 690369.375 4758594.075
Kuehlenberg 4177148.376 642997.635 4760764.800 4177796.064 643026.700 4761228.899
Ex Mergelaec 4137012.190 671808.029 4791128.215 4137659.549 671837.337 4791592.531
Ex Hof Asperg 4146292.729 666952.887 4783859.856 4146940.228 666982.151 4784324.099
Ex Kaisersbach 4138759.902 702670.738 4785552.196 4139407.506 702700.227 4786016.645

Table 2
Contrast of the proposed dual quaternion-based algorithm and the weighted Procrustes algorithm.

Scheme No.
Dual quaternion-based algorithm
Weighted Procrustes algorithm
Rotation matrix (R) Translation vector (T) Scale (l)
2 3
1:000000000 0:000004815 0:000004333 (641.8805, 68.6554, 416.3982) 1.000005583
4 0:000004815 1:000000000 0:000004841 5
0:000004333 0:000004841 1:000000000
2 3
0:999999999 0:000004814 0:000004332 (641.8804, 68.6553, 416.3982) 1.000005583
4 0:000004814 0:999999999 0:000004841 5
0:000004332 0:000004841 0:999999999

Table 3
Residuals of each site after transformation.

Site Dual quaternion-based algorithm Weighted Procrustes algorithm

Dx Dy Dz DX DY DZ
Solitude 0.0939 0.1353 0.1402 0.0940 0.1351 0.1402
Buoch Zeil 0.0588 0.0500 0.0136 0.0588 0.0497 0.0137
Hohenneuffen 0.0403 0.0883 0.0078 0.0399 0.0879 0.0081
Kuehlenberg 0.0197 0.0213 0.0872 0.0202 0.0220 0.0874
Ex Mergelaec 0.0916 0.0140 0.0059 0.0919 0.0139 0.0055
Ex Hof Asperg 0.0117 0.0067 0.0549 0.0118 0.0065 0.0546
Ex Kaisersbach 0.0292 0.0035 0.0014 0.0294 0.0041 0.0017

 
Making A0 ¼ N1 CT2 C3  C1 þ CT1 and kA0 as its eigenvector, the Table 4
Point features extracted from two neighboring LiDAR point clouds.
eigenvalue of matrix A can be expressed as
No. Reference station Unregistered station
1 l x y z x y z
kA ¼ ðl2 C 22 þ C 33 Þ þ kA0 : ð42Þ
4N 2 1 91.406 53.344 8.320 49.007 54.453 0.978
2 91.297 53.222 0.916 47.365 54.435 6.242
Considering that l, N, C22, and C33 are all constants, kA is maximized
3 60.158 24.280 8.948 36.514 13.733 3.642
only when kA00 is maximized. Therefore, the calculation of the max- 4 60.135 24.278 1.521 34.881 13.859 3.608
imum eigenvalue of matrix A0 and its eigenvector r_ efficiently 5 56.298 19.186 5.700 53.378 25.872 4.187
obtains our desired result 6 13.269 2.677 1.444 7.324 32.695 1.389
By substituting r_ into Eq. (31), (33), l is obtained. Using Eq. (29) 7 4.666 17.245 1.605 9.587 19.650 2.449
8 49.939 14.297 27.119 36.532 0.319 21.980
results in s_ . Therefore, the best registration parameters are 9 52.769 11.523 25.906 39.932 1.307 19.965
acquired. 10 72.929 8.630 27.146 67.051 8.834 15.017
11 46.500 30.291 23.078 54.124 40.688 13.216
12 52.581 22.934 5.676 51.943 30.962 3.965
3.3. Implementation of the Proposed Registration Algorithm
13 58.972 17.511 18.862 57.712 23.376 8.397
14 55.429 26.155 23.077 59.650 32.625 12.037
The detailed implementation of the proposed registration algo- 15 55.313 26.131 23.039 59.512 32.705 12.071
rithm can be summarized as follows: 16 63.467 27.962 26.981 41.466 18.246 21.085
17 57.673 22.069 25.782 39.133 10.234 20.247
18 49.687 14.083 3.666 29.781 0.026 8.062
1. We input the two sets of corresponding feature points f~ p0i g and
0
f~
pi g, which are extracted from the reference station and unreg-
istered station, respectively.
2. We construct matrix A0 based on matrices C1, C2, and C3 as well 4. Experiments
as the two constants C4 and C5. Then, we calculate the maxi-
mum eigenvalue and its corresponding eigenvector r_ of A0 . 4.1. Experimental scheme designs
3. We calculate the scale parameter l using Formula (31), or (33)
and the quaternion s_ using Formula (29). The proposed algorithm is implemented using C++ program-
4. We calculate the rotation matrix R and translation vector ~ t ming language. Two experimental schemes are designed to verify
using Formula (13). the correctness and effectiveness of the algorithm. The first
68 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69
Table 5
Results of the proposed dual quaternion-based registration algorithm.
Scheme No. Rotation matrix (R)
2 3
1 0:850416493 0:494507078
4 0:479380907 0:868981201
0:216761931 0:018287246
2 3
2 0:850416493 0:494507078
4 0:479380907 0:868981201
0:216761931 0:018287246
experiment involves the coordinate transformation between two
different coordinate systems. The results are compared with those
of the well-known weighted Procrustes algorithm (Grafarend and
Awange, 2003). The second experiment involves the use of the pro-
posed algorithm in the registration of LiDAR points. The correct-
ness of the proposed model after the introduction of the scale
parameter is analyzed in detail.
4.2. Coordinate transformation in geodesy
The Cartesian coordinates of the seven stations, as shown in
Table 1, are taken from Grafarend and Awange (2003). Using these
coordinates, we compute the transformation parameters from a
local geodetic system (System A) to WGS84 (System B). The trans-
formation parameters from the local coordinate system to WGS84
are calculated by the proposed dual quaternion-based algorithm
and the weighted Procrustes algorithm. The results are used to ver-
ify the correctness of the dual quaternion-based algorithm.
The results of the two algorithms are shown in Table 2.
The residuals of each site after transformation are shown in
Table 3.
Based on the residuals shown in Table 3, the standard error of
the dual quaternion-based algorithm is 0.011799 and that of the
weighted Procrustes algorithm is 0.011796. The difference is at
the fifth place after the decimal point, which is small enough to
ignore. Therefore, we can conclude that the dual quaternion-based
algorithm is valid to calculate the similarity transformation
parameters.
4.3. LiDAR point cloud registration
The second experimental data are the two LiDAR point clouds
acquired by a terrestrial LiDAR instrument (Riegl LMS-Z420i). Cur-
rently, most registration algorithms are highly efficient, and almost
all of them can calculate real-time or quasi-real-time transforma-
tion parameters. Therefore, in this study, only the correctness of
the presented algorithm is verified. Sample point clouds are
acquired by the Riegl LMS-Z420i series terrestrial LiDAR. Conjugate
point features are extracted by (1) fitting salient points marked by
special reflective material and (2) by taking the intersection of
three planes. Both extraction means are highly accurate. The
extracted conjugate point features used to test the proposed algo-
rithm are shown in Table 4.
The extracted conjugate point features are directly included in
the presented dual quaternion-based registration algorithm. The
transformation parameters are presented in the first row in Table 5.
To further verify the validity of the presented algorithm, all point
coordinates from the unregistered station are scaled to half of their
actual coordinates and are included in the proposed dual quater-
nion-based registration algorithm. The transformation parameters
are shown in the second row in Table 5.
To further verify the correctness of the dual quaternion-based
registration algorithm, the unit quaternion-based algorithm
(Horn, 1987) is used to calculate the transformation parameters.
Comparatively, the results of the unit quaternion-based algorithm
are the same as those in Table 5.
Translation vector (T) Scale factor (s)
0:179595486 22.9656, 29.3962, 2.2652 1.000385435
0:122742085 5
0:976053196
0:179595486 22.9656, 29.3962, 2.2652 2.000770870
0:122742085 5
0:976053196
Therefore, we conclude that the proposed dual quaternion-
based registration algorithm works and is an alternative for the
concise description of similarity transformation.
5. Conclusions
The representation of rigid transformation in 3D space, espe-
cially of orientation, is a crucial issue in navigation, geodesy, pho-
togrammetry, and robot arm manipulation, among others.
Currently, the most popular representations of a 3D rotation are
rotation matrix, Euler angles, Rodrigues vector, and quaternions.
As renormalization is difficult for rotation matrices, representation
by Euler angles may cause singularities, and Rodrigues vectors do
not allow for an easy composition algorithm, quaternions seem
to be a suitable representation of rotations in 3D with few param-
eters. Considering that unit quaternions represent rotation in
space, seven transformation parameters are calculated, namely,
the rotation angles first, the scale parameter later, and three trans-
lation vectors last. In the case of introducing an error in calculating
the rotation angles, the error may be amplified in the following two
steps. Dual quaternions are used to represent similarity transfor-
mation. A single-cost function is formulated in which the real part
and the dual part of the dual quaternions are used to represent
rotation and translation. By minimizing the formulated single-cost
function, seven transformation parameters are calculated simulta-
neously. Detailed derivations of related formulas are given and
analyzed, and experiments are conducted to verify the effective-
ness and the feasibility of the proposed algorithm.
As only point features are used to construct the similarities
between the two neighboring LiDAR stations, future work can
focus on introducing line and face features in pairwise registration.
A newly formulated dual quaternion-based error norm should con-
sider orientation and position errors simultaneously to formulate a
unified pairwise registration model that considers point, line, and
face features simultaneously.
Acknowledgments
This study was funded by the National Natural Science Founda-
tion of China (Project Numbers 41271444/41001297/41001312)
and the Natural Science Foundation of Jiangsu Province (Project
Number BK2012569).
References
Arun, K.S., Huang, T.S., Blostein, S.D., 1987. Least-squares fitting of two 3-D point
sets. IEEE Trans. Pattern Anal. Mach. Intell. 9 (5), 698–700.
Besl, P.J., McKay, N.D., 1992. A method for registration of 3-D shapes. IEEE Trans.
Pattern Anal. Mach. Intell. 14 (2), 239–256.
Chasles, M., 1830. Note sur les propriétés générales du système de deux corps
semblables entr’eux et placés d’une manière quelconque dans l’espace; et sur le
déplacement fini ou infiniment petit d’un corps solide libre. Bull. Sci. Math.
Astron. Phys. Chim. 14 (XIV), 321–326.
Daniilidis, K., 1999. Hand-eye calibration using dual quaternions. Int. J. Robotics
Res. 18 (3), 286–298.
Eggert, D.W., Lorusso, A., Fisher, R.B., 1997. Estimating 3-D rigid body
transformations: a comparison of four major algorithms. Mach. Vis. Appl. 9
(5–6), 272–290.
 Y. Wang et al. / ISPRS Journal of Photogrammetry and Remote Sensing 94 (2014) 63–69
Grafarend, E.W., Awange, L.J., 2003. Nonlinear analysis of the three dimensional
datum transformation [conformal group C7(3)]. J. Geodesy 77 (1), 66–76.
Hamilton, W.R., 1844. On quaternions, or on a new system of imaginaries in algebra.
Phil. Mag. 25 (3), 489–495.
Haralick, R.M., Joo, H., Lee, C.N., et al., 1989. Pose estimation from corresponding
point data. IEEE Trans. SMC 19 (6), 1426–1446.
Horn, B.K., 1987. Closed-form solution of absolute orientation using unit
quaternions. J. Opt. Soc. Am. (Ser. A) 4 (4), 629–642.
Horn, B.K., 1988. Closed form solution of absolute orientation using orthonormal
matrices. J. Opt. Soc. Am. (Ser. A) 5 (7), 1127–1135.
Prošková, J., 2011. Application of dual quaternions algorithm for geodetic datum
transformation. J. Appl. Math. 4 (2), 225–236.
Prošková, J., 2012. Discovery of dual quaternions for geodesy. J. Geometry Graphics
16 (2), 195–209.
Joseph, J., Laviola, J., 2003. A comparison of unscented and extended Kalman
filtering for estimating quaternions motion. IEEE Press. 3, 2435–2440.
69
Kim, A., Golnaraghi, M.F., 2004. A quaternions-based orientation estimation
algorithm using an inertial measurement unit. IEEE Position Location
Navigation Symp., 268–272.
Li, B., Shen, Y., Li, W., 2012. The seamless model for three-dimensional datum
transformation. Sci. China (Earth Sci.) 55 (12), 2099–2108.
Walker, M.W., Shao, L., Volz, R.A., 1991. Estimating 3-D location parameters using
dual number quaternions. CVGIP: Image Understanding 54 (3), 358–367.
Yang, Y., 1999. Robust estimation of geodetic datum transformation. J. Geodesy 73
(5), 268–274.
Shen, Y.Z., Chen, Y., Zheng, D.H., 2006. A quaternions-based geodetic datum
transformation algorithm. J. Geodesy 80 (5), 233–239.
Zeng, H., Yi, Q., 2011. Quaternion-based iterative solution of three-dimensional
coordinate transformation problem. J. Comput. 6 (7), 1361–1368.
Zheng, D., Yue, D., Yue, J., 2008. Geometric feature constraint based algorithm for
building scanning point cloud registration. Acta Geodaetica Et Cartographica
Sin. 37 (4), 464–468.