Ami 99 G
Ami 99 G
Ami 99 G
Graph Method
Tomas Suk and Jan Flusser
Institute of Information Theory and Automation
Academy of Sciences of the Czech Republic
Pod vod
arenskou vez 4, 182 08 Praha 8, Czech Republic
E-mail: [email protected], [email protected]
Abstract
These tables include all irreducible affine moment invariants up to the 10th
order. They are generated by a newly developed graph method. The technique
of elimination of reducible invariants is described. There are also mentioned a few
approaches to the selection of an independent set of the invariants. Finally, an
example of a few dependencies among irreducible invariants is presented.
Keywords: affine invariants, moments, graphs, independent features.
Acknowledgment
This work has been supported by the grant No. 201/03/0675 of the Grant
Agency of the Czech Republic.
Introduction
Recognition of objects and patterns that are deformed in various ways has been a goal of
intensive research for many years. The degradations use to be introduced during the image
acquisition process by such factors as imaging geometry, systematic and random sensor
errors, illumination changes, object occlusion, etc. Finding a set of invariant descriptors
is a key step to recognizing degraded objects regardless of the particular deformations.
Moment invariants have become a powerful tool for recognizing objects regardless of
their particular position, orientation, viewing angle, and other variations. There is a
well elaborated theory on rotation moment invariants [1, 2, 3, 4, 5, 6, 7, 8, 9], which
have been successfully used in numerous applications. In practice, however, we often face
object deformations that are beyond the rotation-translation-scaling model. An exact
model of photographing a planar scene by a pin-hole camera whose optical axis is not
perpendicular to the scene is projective transform. Since the projective transform is not
linear, its Jacobian is a function of spatial coordinates and projective moment invariants
from a finite number of moments cannot exist [10, 11].
For small objects and large camera-to-scene distance is the perspective effect negligible and the projective transform can be well approximated by affine transform. Thus,
having powerful affine moment invariants for object description and recognition is in great
demand.
A pioneer work on this field was done independently by Reiss [12] and Flusser and Suk
[13, 14], who introduced affine moment invariants (AMIs) and proved their applicability
in simple recognition tasks. In their papers, the derivation of the AMIs originated from
the classical theory of algebraic invariants from 19th century, e.g. [15, 16], or newer
[17, 18]. The book [19] contains original notes of the course held by famous German
mathematician David Hilbert in 1897. The notes were edited, translated to English and
first published 50 years after Hilberts death.
In recent years a novel approach to the affine invariants has arisen: normalization
[21, 22]. The normalized moments have some advantages in comparison to the AMIs,
namely easy creation of complete and independent sets, and one drawback, they are
discontinuous on symmetric objects. While the AMIs are simply zero, but continuous,
and the problem of recognition of the symmetric objects is problem of selection of proper
invariants, in case of the normalized moments we must select not only proper moments,
but also proper thresholds in a test of their zeroness. Therefore AMIs remain important
mean for recognition of the affinely distorted objects. There is also a compromise, we can
normalize moments up to the last rotation and then compute rotation invariants from
them. The creation of complete and independent sets is easier then in case of AMIs, e.g.
[23] and problems with discontinuities fall off. They have also theoretical meaning: their
comparison with AMIs can be used for verification of the independency of the AMIs,
how we will see later.
2
The goal of these tables is to compute and publish affine moment invariants of higher
orders and weights. Reiss in [20] presents 17 selected invariants of the 6th order in
maximum. These tables present all irreducible 362 invariants up to weight 10. They are
connected with our previous paper [13]. They also deal with the problem of selection of
the independent invariants. They are organized by the following way: Section 2 deals with
derivation of the AMIs from algebraic invariants, Section 3 discusses various methods,
how to find the number of invariants, Section 4 describes the graph method used for
derivation of the AMIs, Section 5 concerns with looking for dependencies between the
invariants and Section 6 describes a method, how to verify independency of the chosen
invariants. It is followed by the tables of all irreducible AMIs up to the 10th order and
some dependencies between them.
(1)
The general two-dimensional (p + q)-th order moments of an image f (x, y) are defined
as
mpq =
Z Z
xp y q f (x, y)dxdy
p, q = 0, 1, 2, . . .
(2)
The fundamental theorem deals with the relation of the AMIs and algebraic invariants.
It was published in [1] and revisited independently by Reiss [12] and Flusser and Suk [14].
It can be formulated as follows.
Theorem 1: If the binary form of order p has an algebraic invariant of weight w and
degree k
I(ap,0 , , a0,p ) = w I(ap,0 , , a0,p )
( denotes the determinant of the respective affine transform) then the moments of order
p have the same invariant but with the additional factor |J|k :
I(p0 , , 0p ) = w |J|k I(p0 , , 0p ),
where |J| is the absolute value of the Jacobian. Actually = J = a1 b2 a2 b1 .
This theorem makes possible to use results of the theory of algebraic invariants for
computation of the AMIs. The algebraic invariants and the AMIs differ by the normalization to the scaling only. The AMIs can be derived directly, without the theory
of the algebraic invariants, by the following way. The affine transformation (1) can be
decomposed into six one-parameter transformations.
3
Horizontal translation:
Vertical translation:
u=x+
v = y.
(3)
u=x
v = y + .
(4)
u = x
v = y.
(5)
u = x
v = 1 y.
(6)
u = x + ty
v = y.
(7)
u=x
v = y + sx.
(8)
Scaling
Stretching:
Horizontal skew:
Vertical skew:
Any function F of moments is invariant under these six transformations if and only if it
is invariant under the general affine transformation (1). Each of these transformations
implies one constraint on the invariants. For completeness, the affine transform composed
from these six elementary transforms cannot have negative Jacobian. To enable it, we
would have to insert possible mirror reflection to the decomposition
u=x
v = zy,
(9)
where z can be either 1 or -1. We will explain consequences of not doing it later.
Invariancy to translation can be provided by using central moments instead of general
ones (2), any function of them is invariant under the translations (3) and (4). Central
moments are defined as
pq =
Z Z
(x xt )p (y yt )q f (x, y)dxdy
p, q = 0, 1, 2, . . . ,
(10)
where xt = m10 /m00 and yt = m01 /m00 are the coordinates of the centroid.
The constraint of the scaling implies the correct normalization of the moments. The
moments after scaling
pq = p+q+2 pq ,
(11)
particularly
00 = 2 00 .
(p+q+2)/2
(12)
is invariant under scaling.
(13)
That is why the products of moments, where the sum of p-th indices equals the sum of q-th
indices, are invariant under the stretching. The algebraic invariants are functions of binary
form coefficients and have similar form as the AMIs. They have the same coefficients
and the only difference is the normalization to scaling. The moment invariants use the
exponent (p + q + 2)/2 while the algebraic invariants use (p + q)/2. The weight w of the
invariant equals the sum of the p-th indices and the sum of q-th indices of one term of
the corresponding moment invariant. It is important characterization of the invariant.
Other attributes of the invariant are an order, that is the highest order of a moment
in the invariant, and a degree. If we suppose the invariant in form of a polynomial of
the moments, then the degree is the number of moments in one term of the polynomial.
The more precise characteristics of the invariant is a structure, that is the orders of all
moments in one term.
Now, something about the mirror reflection. The invariants with odd weight change
their signs, when the affine transform has negative Jacobian. We can either use absolute
values of them or suppose the affine transform has always positive Jacobian and two
images differing by the mirror reflection are two different images, how is suitable in many
practical tasks.
Two differential equations can be derived from the skew transformations (7) and (8):
F
=0
pq
(14)
F
= 0.
pq
(15)
XX
pp1,q+1
XX
qp+1,q1
and
p
In the theory of algebraic invariants, they are called Cayley - Aronhold differential equations. We can use these equations for computation of coefficients of terms of the invariants
as in [13], but another method for generation of the AMIs was used in this report, see
Section 4, where the algorithms are described.
The number of independent invariants is very important for their using in pattern recognition. It can be computed by a few ways. The basic method is rule of thumb: The
number n of independent invariants equals
n = m p,
(16)
satisfied (see e.g. [10]). Mostly it equals the number of parameters of the transformation.
This equation is called rule of thumb, because often it is not easy to find, which measurements and constraints are independent and which not. Each dependency between the
measurements decreases the number of invariants by one, and each dependency between
the constraints increases the number of invariants by one. In our case, the moments are
independent. The zero-order moment is used to satisfy the scaling constraint and the two
first-order moments are used to satisfy two translation constraints. If we use moments of
the second order only, then the differential equations (14) and (15) become:
211
F
F
+ 02
=0
20
11
(17)
211
F
F
+ 20
= 0.
02
11
(18)
and
If the derivative with respect to the 11 is excluded from these equations, we acquire
02
F
F
= 20
.
02
20
(19)
then
and
(20)
F
= kr1 r201 1 r112 r023
20
(21)
F
= kr3 r201 r112 r023 1 .
02
(22)
(23)
(24)
where w = kr/2. The invariant is homogeneous, if it contains moments of one order only.
The non-homogeneous invariants are sometimes called simultaneous. A similar relation
is satisfied for the simultaneous invariants:
N (k1 , r1 , k2 , r2 , , kn , rn , w) =
= A(k1 , r1 , k2 , r2 , , kn , rn , w) A(k1 , r1 , k2 , r2 , , kn , rn , w 1),
(25)
A(k, r, w) =
"
k+r
r
"
k+r
k
(27)
where [ ]w denotes the coefficient at w-th power of the polynomial. Then the number of
the invariants is
N (k, r, w) = (1 x)
"
k+r
k
=
w
"
"
k+r
k
(28)
generally
A(k1 , r1 , k2 , r2 , , kn , rn , w) =
"
k1 + r1
k1
#"
N (k1 , r1 , k2 , r2 , , kn , rn , w) =
"
k1 + r1
k1
# "
k2 + r2
k2
k2 + r2
k2
"
kn + rn
kn
kn + rn
kn
(29)
(30)
If we need to know the whole number of the independent invariants of some orders, we
can sum the numbers over all structures, but how we will explain later the possibilities
of the Cayley-Sylvester theorem in this case are limited. The entire number of independent non-constant affine invariants from the moments of the 2nd and 3rd order equals
X
N (2, r, 3, 2s; w). We add over all nonnegative integers r, s, which satisfy the conr,s
r+3s=w
straint r+3s = w. We can choose, whether we add over r, i.e. r = z, z+3, z+6, , w3, w
and s = (w r)/3 or over s, i.e. s = 0, 1, 2, , [w/3] 1, [w/3] and r = w 3s. [w/3]
means integer part of w/3 and z is the reminder after dividing w by 3, i.e. z = w 3 [w/3].
7
N (2, r, 3, 2s; w) =
r,s
r+3s=w
(1 x
2+r
r,s
r+3s=w
(1 x
r,s
r+3s=w
1+r
r,s
r+3s=w
"
2+r
2
3+2s
3+2s
#"
3 + 2s
3
)(1 x )(1 x
)(1 x
)(1 x1+2s )
=
(1 x)(1 x2 )(1 x2 )(1 x3 )
w
r,s
r+3s=w
2+2s
2+2s
=
w
(31)
=
w
The members with the indices, which are always greater than w can be omitted.
r,s
r+3s=w
[w/3]
X
x + x2 + x3
1
=
2
2
3
2
2
3
s=0 (1 x)(1 x ) (1 x ) w2s
s=0 (1 x)(1 x ) (1 x ) w
[w/3]
[w/3]
s=0
[w/3]
[w/3]
x +x +x
2
2
3
(1 x)(1 x ) (1 x )
w4s
X
x + x2
2
2
3
(1 x)(1 x ) (1 x )
s=0
[w/3]
X
x3
+
(1 x)(1 x2 )2 (1 x3 )
s=0
s=0
[w/3]
3s
=
w
(32)
x
2
2
3
(1 x)(1 x ) (1 x )
X x2 + 2x3 + 2x4 + x5
+
(1 x)(1 x2 )2 (1 x3 )
s=0
[w/3]
w6s
x4 + x5 + x6
.
2 )2 (1 x3 )
(1
x)(1
x
s=0
6sw
4sw
X
1
1
([w/3] + 1) =
=
2 )2 (1 x3 )
2 )2 (1 x3 )
(1
x)(1
x
(1
x)(1
x
s=0
w
w
X
(33)
The derivatives can be used for transformation of this formula. If we define P (x) =
P
d P
d
n+1
n
))|w = ( n (n+1)an xn ))|w =
n an x , then P (x)|w = aw and dx (xP (x))|w = dx ( n an x
(w+1)aw = P (x)|w (w+1). The weight w can be expressed by integer division and modulo
as w = 3p + z, where p = [w/3].
1
(p + 1) =
=
2
2
3
(1 x)(1 x ) (1 x ) 3p+z
(1 x3 )(1 x6 )2
(p + 1) =
=
2
2
3
2
6
2
(1 x)(1 x ) (1 x ) (1 x ) 3p+z
(1 + x + x2 )(1 + x2 + x4 )2
(p + 1) =
=
(1 x3 )2 (1 x6 )2
3p+z
1 + 2x + 5x2 + x3
(p + 1); z = 0
2
2 2
(1 x) (1 x ) p
(34)
1 + 5x + 2x2 + x3
(p + 1); z = 1
=
2 (1 x2 )2
(1
x)
3 + 3x + 3x2
(p + 1); z = 2.
(1 x)2 (1 x2 )2
p
If z = 0:
1 + 2x + 5x2 + x3
d x + 2x2 + 5x3 + x4
(p + 1) =
=
(1 x)2 (1 x2 )2 p
dx (1 x)2 (1 x2 )2 p
=
p
(1 + 4x + 15x + 4x )(1 x ) + 2(x + 2x + 5x + x )(1 + 3x)
=
=
(1 x)2 (1 x2 )3
p
1 + 6x + 24x2 + 22x3 + 17x4 + 2x5
1 + 6x3 + 24x6 + 22x9 + 17x12 + 2x15
=
=
.
(1 x)2 (1 x2 )3
(1 x3 )2 (1 x6 )3
p
3p
(35)
If z = 1:
1 + 5x + 2x2 + x3
d x + 5x2 + 2x3 + x4
(p + 1) =
=
(1 x)2 (1 x2 )2 p
dx (1 x)2 (1 x2 )2 p
=
p
(1 + 10x + 6x + 4x )(1 x ) + 2(x + 5x + 2x + x )(1 + 3x)
=
=
(1 x)2 (1 x2 )3
p
4
7
10
2
3
4
5
x + 12x + 21x + 28x + 8x13 + 2x16
1 + 12x + 21x + 28x + 8x + 2x
=
=
(1 x)2 (1 x2 )3
(1 x3 )2 (1 x6 )3
p
.
3p+1
(36)
If z = 2:
d 3x + 3x2 + 3x3
3 + 3x + 3x2
(p + 1) =
=
(1 x)2 (1 x2 )2 p
dx (1 x)2 (1 x2 )2 p
(37)
(3 + 6x + 9x )(1 x ) + 2(3x + 3x + 3x )(1 + 3x)
=
=
(1 x)2 (1 x2 )3
p
2
5
8
2
3
4
3x + 12x + 30x + 18x11 + 9x14
3 + 12x + 30x + 18x + 9x
=
=
(1 x)2 (1 x2 )3
(1 x3 )2 (1 x6 )3
p
.
3p+2
1
=
2
2
3
s=0 (1 x)(1 x ) (1 x ) w
X
(38)
.
w
Another term:
[w/3]
[w/3]
X
x + x2 + x3
(x + x2 + x3 )x2s
=
=
2 )2 (1 x3 )
2 )2 (1 x3 )
(1
x)(1
x
(1
x)(1
x
s=0
s=0
w2s
w
Pp
s=0
x2s =
1x2(p+1)
.
1x2
(39)
The partition w = 3p + z,
(x + x2 + x3 )(1 x2([w/3]+1) )
=
=
(1 x)(1 x2 )3 (1 x3 ) w
(40)
(x + x2 + x3 )x2p+2
x + x2 + x3
=
.
(1 x)(1 x2 )3 (1 x3 ) w (1 x)(1 x2 )3 (1 x3 ) 3p+z
(x + x2 + x3 )x2p+2
x3 + x4 + x5
=
=
2
3
3
2
3
3
(1 x)(1 x ) (1 x ) 3p+z (1 x)(1 x ) (1 x ) p+z
z=0:
x9 + x12 + x15
x3 + x4 + x5
=
(1 x)(1 x2 )3 (1 x3 ) p
(1 x3 )(1 x6 )3 (1 x9 ) 3p
x7 + x10 + x13
x2 + x3 + x4
=
z
=
1
:
=
=
2
3
3
3
6
3
9
(1 x)(1 x ) (1 x ) p
(1 x )(1 x ) (1 x ) 3p+1
z=2:
x + x2 + x3
(1 x)(1 x2 )3 (1 x3 )
x5 + x8 + x11
=
(1 x3 )(1 x6 )3 (1 x9 )
10
3p+2
(41)
x + x2 + x3
=
2
2
3
s=0 (1 x)(1 x ) (1 x ) w2s
X
x + x2 + x3
x5 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x15
.
(1 x)(1 x2 )3 (1 x3 ) w
(1 x3 )(1 x6 )3 (1 x9 )
w
(42)
[w/3]
X
x3 + x4 + x5
(x3 + x4 + x5 )x4s
=
=
2
2
3
2
2
3
s=0 (1 x)(1 x ) (1 x ) w4s
s=0 (1 x)(1 x ) (1 x ) w
x3 + x4 + x5
(x3 + x4 + x5 )(1 x4([w/3]+1) )
,
=
=
(1 x)(1 x2 )2 (1 x3 )(1 x4 ) w (1 x)(1 x2 )2 (1 x3 )(1 x4 ) w
(43)
4([w/3]+1) is always greater than w, that is why the second term can be omitted. Another
term of (32) is
[w/3]
X
x6+6s
x6
=
=
2 )2 (1 x3 )
2 )2 (1 x3 )
(1
x)(1
x
(1
x)(1
x
s=0
s=0
w6s
w
[w/3]
x6 (1 x6([w/3]+1) )
x6
=
=
,
2
2
3
6
2
2
3
6
(1 x)(1 x ) (1 x )(1 x ) w (1 x)(1 x ) (1 x )(1 x ) w
(44)
6([w/3]+1) is always greater than w, that is why the second term can be omitted. Another
term of (32) is
[w/3]
X
(x + x2 )x3s
x + x2
=
=
2 )2 (1 x3 )
2 )2 (1 x3 )
(1
x)(1
x
(1
x)(1
x
s=0
s=0
3s
0
[w/3]
(x + x2 )(x3([w/3]+1) 1)
(x + x2 )(x3[w/3] x3 )
=
=
=
(1 x)(1 x2 )2 (1 x3 )(x3 1) 0 (1 x)(1 x2 )2 (1 x3 )2 0
x + x2
(x + x2 )(1 + x + x2 )(1 + x2 + x4 )2
=
=
=
(1 x)(1 x2 )2 (1 x3 )2 3[w/3]
(1 x3 )3 (1 x6 )2
3[w/3]
x + 2x2 + 4x3 + 5x4 + 7x5 + 8x6 + 8x7 + 7x8 + 5x9 + 4x10 + 2x11 + x12
=
=
(1 x3 )3 (1 x6 )2
3[w/3]
11
(45)
Another term
[w/3]
[w/3]
X (x2 + 2x3 + 2x4 + x5 )xs
x2 + 2x3 + 2x4 + x5
=
=
2 2
3
2 2
3
s=0 (1 x)(1 x ) (1 x ) s
s=0 (1 x)(1 x ) (1 x ) 0
x2 + 2x3 + 2x4 + x5
x6 + 2x9 + 2x12 + x15
=
=
=
(1 x)2 (1 x2 )2 (1 x3 ) [w/3] (1 x3 )2 (1 x6 )2 (1 x9 ) 3[w/3]
(x6 + 2x9 + 2x12 + x15 )(1 + x + x2 )
x6 + 2x9 + 2x12 + x15
=
=
.
3 )(1 x6 )2 (1 x9 )
(1 x3 )2 (1 x6 )2 (1 x9 )
(1
x)(1
x
w
w
(46)
[w/3]
X
x3
x3+w6s
=
=
2 )2 (1 x3 )
2 )2 (1 x3 )
(1
x)(1
x
(1
x)(1
x
s=0
s=0
6sw
0
x3+w (x6[w/3] x6 )
x3+w (x6([w/3]+1) 1)
=
=
=
2
2
3
6
2
2
3
6
(1 x)(1 x ) (1 x )(x 1) 0 (1 x)(1 x ) (1 x )(1 x ) 0
x3+2w6[w/3]
x3+2z
=
=
=
(1 x)(1 x2 )2 (1 x3 )(1 x6 ) w (1 x)(1 x2 )2 (1 x3 )(1 x6 ) 3p+z
x3+z (1 + x + x2 )(1 + x2 + x4 )2
=
=
(1 x3 )2 (1 x6 )3
3p
z=0:
= z=1:
z=2:
x + 2x2 + 5x3 + x4
(1 x)2 (1 x2 )3 p
3x2 + 3x3 + 3x4
(1 x)2 (1 x2 )3 p
x2 + 5x3 + 2x4 + x5
(1 x)2 (1 x2 )3
x3 + 2x6 + 5x9 + x12
=
(1 x3 )2 (1 x6 )3 3p
3x7 + 3x10 + 3x13
=
(1 x3 )2 (1 x6 )3 3p+1
x8 + 5x11 + 2x14 + x17
=
(1 x3 )2 (1 x6 )3
3p+2
12
(47)
[w/3]
X
x4 + x5 + x6
(x4 + x5 + x6 )xw4s
=
2 2
3
2 2
3
s=0 (1 x)(1 x ) (1 x ) 4sw s=0 (1 x)(1 x ) (1 x ) 0
(x4 + x5 + x6 )x3p+z4p
(x4 + x5 + x6 )xz
=
=
=
2
2
3
4
2
2
3
4
(1 x)(1 x ) (1 x )(1 x ) w (1 x)(1 x ) (1 x )(1 x ) p
z=0:
x12 + x15 + x18
(1 x3 )(1 x6 )2 (1 x9 )(1 x12 ) 3p
x16 + x19 + x22
3
6
2
9
12
(1 x )(1 x ) (1 x )(1 x ) 3p+1
x20 + x23 + x26
(1 x3 )(1 x6 )2 (1 x9 )(1 x12 )
= z=1:
z=2:
3p+2
(48)
N (2, r, 3, 2s; w) =
r,s
r+3s=w
x5 + x7 + x8 + x9 + x10 + x11 + x12 + x13 + x15
x+x +x
+
(1 x)(1 x2 )3 (1 x3 ) w
(1 x3 )(1 x6 )3 (1 x9 )
w
3
4
5
6
x +x +x
x
+
2
2
3
4
2
2
3
6
(1 x)(1 x ) (1 x )(1 x ) w (1 x)(1 x ) (1 x )(1 x ) w
4x3 + 8x6 + 5x9 + x12
x6 + 2x9 + 2x12 + x15
+
(1 x)(1 x3 )2 (1 x6 )2 w (1 x)(1 x3 )(1 x6 )2 (1 x9 ) w
x3 + 2x6 + 3x7 + x8 + 5x9 + 3x10 + 5x11 + x12 + 3x13 + 2x14 + x17
+
(1 x3 )2 (1 x6 )3
w
12
15
16
18
19
20
22
23
26
x +x +x +x +x +x +x +x +x
=
(1 x3 )(1 x6 )2 (1 x9 )(1 x12 )
2
13
(49)
x+x +x
2
3
3
(1 x)(1 x ) (1 x )
5
(x3 + x4 + x5 )(1 x6 ) x6 (1 x4 )
+
(1 x)(1 x2 )2 (1 x3 )(1 x4 )(1 x6 )
6
12
(50)
w
=
(1 x)(1 x3 )2 (1 x6 )2 (1 x9 )
w
14
=
w
=
w
(51)
=
w
(52)
=
w
1 2x3 + 2x9 2x12 + 2x15 x18 + x24 2x27 + 2x30 2x33 + 2x39 x42
=
=
(1 x2 )(1 x3 )2 (1 x4 )(1 x6 )3 (1 x9 )(1 x12 )
w
1 x6 x12 + x24 + x30 x36
=
=
2
4
6
3
9
12
(1 x )(1 x )(1 x ) (1 x )(1 x ) w
1 x12 x18 + x30
=
=
(1 x2 )(1 x4 )(1 x6 )2 (1 x9 )(1 x12 ) w
1 x18
=
=
2
4
6
2
9
(1 x )(1 x )(1 x ) (1 x ) w
1 + x9
=
.
(1 x2 )(1 x4 )(1 x6 )2 w
(53)
The result in form 1/(1 x )|w can be understood as the sum of a geometric series
1 + xn + x2n + x3n + |w . It means an invariant with a weight n and all its powers.
n
15
Similarly, the result in form 1/((1 xn1 )(1 xn2 ))|w means two independent invariants
with weights n1 and n2 , all their powers and products. The polynomials in the numerator
express invariants, which are linearly independent, but some of their powers or products
are linearly dependent. Such invariants together with the independent ones are called
irreducible. Generally, the reducible invariant can be expressed as a polynomial of other
invariants, the irreducible one cannot.
In our case there is one invariant of weight 2, one invariant of weight 4 and two
invariants of weight 6. There is also one invariant of weight 9, but its second power
is dependent. Therefore its sign is independent, but its absolute value is algebraically
dependent on the other four invariants.
There is another method, how to acquire expressions as the last one in (53). We can
compute the beginning of the sequence in (30) for adequate amount of w and then search
a combination of weights, which causes in polynomial multiplication that last members
of the sequence are zero and all members are nonnegative. The following expressions
were acquired by this method. The invariants from the moments of the 2nd, 3rd and 4th
orders:
XX
N (2, r, 3, 2s, 4, t; w) =
r,s,t
r+3s+2t=w
XX
r,s,t
r+3s+2t=w
"
2+r
2
#"
3 + 2s
3
#"
4+t
4
=
#
1 + x4 + 4x6 + 2x7 + 5x8 + 9x9 + 9x10 + 11x11 + 18x12 + 20x13 + 21x14 + 34x15
+30x16 + 36x17 + 44x18 + 45x19 + 46x20 + 60x21 + 54x22 + 57x23 + 66x24
+59x25 + 59x26 + 66x27 + 57x28 + 54x29 + 60x30 + 46x31 + 45x32 + 44x33
+36x34 + 30x35 + 34x36 + 21x37 + 20x38 + 18x39 + 11x40 + 9x41 + 9x42 + 5x43
+2x44 + 4x45 + x47 + x51
=
(1 x2 )(1 x4 )2 (1 x6 )2 (1 x8 )(1 x10 )(1 x12 )(1 x18 )
.
w
(54)
The numbers of invariants from the moments of the 2nd, 3rd, 4th and 5th orders were
acquired by similar method:
16
XXX
N (2, r, 3, s, 4, t, 5, u; w) =
r,s,t,u
2r+3s+4t+5u=2w
XX
r,s,t,u
2r+3s+4t+5u=2w
"
2+r
2
#"
3+s
3
#"
4+t
4
#"
5+u
5
.
w
(55)
These expressions are not too useful, because they do not say anything about the
structure of the invariants and the weights of the invariants in the denominator can be
changed. The expressions for homogeneous invariants, i.e. invariants from the moments of
the same order, are more interesting. They can be calculated by the simpler way, because
the summation is omitted:
N (2, r; w) =
N (3, 2s; w) =
N (4, r; w) =
N (5, 2s; w) =
"
"
5 + 2s
5
"
2+r
2
"
3 + 2s
3
4+r
4
#
w=r
(56)
w = 3s
(57)
1
=
1 x 2 w
1
=
1 x 6 w
1
=
4
6
(1 x )(1 x ) w
w = 2r
1 + x45
=
(1 x10 )(1 x20 )(1 x30 ) w
17
w = 5s
(58)
(59)
N (6, r; w) =
"
6+r
6
N (7, 2s; w) =
"
1 + x45
=
(1 x6 )(1 x12 )(1 x18 )(1 x30 ) w
7 + 2s
7
N (8, r; w) =
8+r
8
w = 4r
(60)
w = 3r
(61)
w
1 + x32 + x36 + x40 + x72
=
8
12
16
20
24
28
(1 x )(1 x )(1 x )(1 x )(1 x )(1 x ) w
(62)
N (9, 2s; w) =
"
9 + 2s
9
N (10, r; w) =
"
10 + r
10
(63)
(64)
It corresponds with results published in [15]. Table 1 presents weights of the independent homogeneous invariants falling under the limit 10.
The way of expressing the number of invariants in (53), (54) and (55) does not express
the structure of the invariants. We can use another way. E.g. the number of invariants
18
2 3 4
5 6 7 8 9 10
2 6 4,6 10 6 - 8 - 10
1 + x3 z 6
=
2
4
6
2
2
2
4
(1 x )(1 z )(1 z )(1 x z )(1 x z ) r,,2t
1 + x3 z 3
,
=
(1 x2 )(1 z 2 )(1 z 3 )(1 x2 z)(1 x2 z 2 ) r,,t
(66)
where t = 2t. It means there is one invariants of the second order with weight 2, two
invariants of the fourth order with weights 4 and 6, two simultaneous invariants second
19
and fourth order with weights 4 and 6 and one dependent simultaneous invariant with
weight 9.
The number of invariants of the 3rd and 4th orders
1 + y 6 z 4 + 2y 6 z 6 + 2y 6 z 8 + y 6 z 10 + y 9 z 4 + 3y 9 z 6 + 2y 9 z 8 + y 9 z 10
y z 2y 12 z 10 3y 12 z 12 y 12 z 14 y 15 z 8 2y 15 z 10 2y 15 z 12 y 15 z 14
x21 y 18
(1 y 6 )(1 z 4 )(1 z 6 )(1 y 6 z 2 )(1 y 6 z 4 )(1 y 3 z 6 )(1 y 6 z 6 )
12 8
=
,3s,2t
1 + y 4 z 2 + 2y 4 z 3 + 2y 4 z 4 + y 4 z 5 + y 6 z 2 + 3y 6 z 3 + 2y 6 z 4 + y 6 z 5
y z 2y 8 z 5 3y 8 z 6 y 8 z 7 y 10 z 4 2y 10 z 5 2y 10 z 6 y 10 z 7 x14 y 9
=
(1 y 4 )(1 z 2 )(1 z 3 )(1 y 4 z)(1 y 4 z 2 )(1 y 2 z 3 )(1 y 4 z 3 )
8 4
.
,
s,t
(67)
We were not successful in removing negative coefficients in the numerator. It means
that not all invariants in the denominator are independent. These results correspond with
that published in [16].
There is used number notation in the next expressions, not weight one. The number
of invariants of the 2nd, 3rd and 4th orders
20
.
r,
s,t
(68)
The negative coefficients in the numerator again mean not all invariants in the denominator are independent. The denominator also suggests there is no other independent
invariant from moments of all 3 of 2nd, 3rd and 4th orders, but all independent invariants
are either homogeneous or from moments of two orders only.
21
v
v2
v4
v6
v8
v 10
v 12
v 14
v 16
v 18
v 20
v 22
v 24
x0 x1 x2 x3 x4 x5 x6 x7 x8
1 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
0 0 1 1 1 1 0 1 0
0 1 1 1 1 0 1 0 -1
0 1 1 1 0 1 0 -1 0
0 1 1 0 1 0 -1 0 0
0 1 0 1 0 -1 0 -1 0
0 0 1 0 -1 0 -1 -1 0
0 1 0 -1 0 -1 -1 -1 0
1 0 -1 0 -1 -1 -1 -1 0
0 -1 0 -1 -1 -1 -1 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 -1
22
.
r,,,
u
(69)
n35 (y, v)
.
4
4
8
12
3
5
2
2
3
5
3
(1 y )(1 v )(1 v )(1 v )(1 y v)(1 y v)(1 y v )(1 yv )(1 y v ) ,s,,u
(70)
The numerator n35 (y, v) can be seen in Table 3.
y 0 y 1 y 2 y 3 y 4 y 5 y 6 y 7 y 8 y 9 y 10 y 11 y 12 y 13 y 14 y 15 y 16
1 0 0 0 0 0 0 0 0 0
0
0
0
0
0
0
0
0 0 0 0 0 0 0 0 0 0
0
0
0
0
0
0
0
0 0 1 0 1 0 1 0 0 0
0
0
0
0
0
0
0
0 0 0 3 0 2 0 1 0 0
0
0
0
0
0
0
0
0 0 1 0 4 0 2 0 -1 0 -1
0 -1
0
0
0
0
0 1 0 3 0 5 0 1 0 -2
0 -2
0
0
0
0
0
0 0 3 0 4 0 3 0 -2 0 -4
0 -1
0
0
0
0
0 1 0 4 0 3 0 0 0 -4
0 -3
0
0
0 -1
0
0 0 2 0 4 0 2 0 -2 0 -3
0 -2
0 -1
0
0
0 2 0 3 0 4 0 0 0 -3
0 -3
0 -3
0
0
0
0 0 2 0 3 0 1 0 -2 0 -4
0 -3
0 -1
0
1
0 1 0 2 0 2 0 -1 0 -4
0 -5
0 -1
0
0
0
0 0 2 0 2 0 0 0 -3 0 -5
0 -2
0
1
0
0
0 1 0 1 0 0 0 -3 0 -5
0 -1
0
0
0
1
0
0 0 0 0 1 0 -2 0 -4 0 -2
0
1
0
0
0
0
0 1 0 0 0 -1 0 -5 0 -3
0
0
0
1
0
1
0
0 0 1 0 -2 0 -5 0 -3 0
0
0
2
0
2
0
0
0 0 0 -1 0 -5 0 -4 0 -1
0
2
0
2
0
1
0
1 0 -1 0 -3 0 -4 0 -2 0
1
0
3
0
2
0
0
0 0 0 -3 0 -3 0 -3 0 0
0
4
0
3
0
2
0
0 0 -1 0 -2 0 -3 0 -2 0
2
0
4
0
2
0
0
0 -1 0 0 0 -3 0 -4 0 0
0
3
0
4
0
1
0
0 0 0 0 -1 0 -4 0 -2 0
3
0
4
0
3
0
0
0 0 0 0 0 -2 0 -2 0 1
0
5
0
3
0
1
0
0 0 0 0 -1 0 -1 0 -1 0
2
0
4
0
1
0
0
0 0 0 0 0 0 0 0 0 1
0
2
0
3
0
0
0
0 0 0 0 0 0 0 0 0 0
1
0
1
0
1
0
0
0 0 0 0 0 0 0 0 0 0
0
0
0
0
0
0
0
0 0 0 0 0 0 0 0 0 0
0
0
0
0
0
0
1
.
,,t,
u
(71)
z0 z1 z2 z3 z4 z5 z6
1 0 0 0 0
0
0
0 0 0 0 1
0
0
0 1 3 4 6
5
4
0 1 3 8 8
9
7
0 2 4 6 8
7
3
0 2 5 7 7
6
1
0 1 2 4 2
0 -4
0 2 3 1 0 -4 -8
0 0 0 -1 -5 -7 -11
1 0 -1 -4 -7 -11 -12
0 0 -1 -4 -6 -9 -9
0 -1 -2 -4 -6 -8 -7
0 0 0 0 -1 -2 -1
0 0 0 1 1
2
2
0 0 0 0 0
0
1
0 0 0 0 0
0
0
0 0 0 0 0
0
0
0 0 0 0 0
0
0
0 0 0 0 0
0
0
0 0 0 0 0
0
0
v0
v2
v4
v6
v8
v 10
v 12
v 14
v 16
v 18
v 20
v 22
v 24
v 26
v 28
v 30
v 32
v 34
v 36
v 38
z7
0
0
2
3
0
-3
-10
-12
-13
-14
-7
-5
1
5
3
2
0
0
0
0
z 8 z 9 z 10 z 11 z 12 z 13 z 14 z 15
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0 -2 -2 -1
0
0
0
0
-5 -6 -6 -4 -2
0
0
0
-8 -9 -8 -5 -3 -1
0
0
-13 -13 -12 -7 -5 -2 -2 -1
-15 -15 -10 -6 -1
1
2
1
-13 -11 -6 -1
5
7
8
6
-11 -7 -1
4
7
9
9
6
-4
1
7 11 14 12 11
7
1
6 11 13 13 11
7
5
6 10 15 15 12
8
4
0
7 12 13 13 10
4
0 -2
5
8
9
8
3 -1 -6 -7
4
6
6
5
0 -3 -7 -8
1
2
2
0 -3 -7 -9 -8
0
0
0 -1 -2 -4 -5 -6
0
0
0
0
0
0
0 -1
0
0
0
0
0
0
0
0
z 16 z 17 z 18 z 19
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-1
0
0
0
0
0
0
0
4
2
1
0
4
1
0
0
4
1
0 -1
1
0
0
0
-1 -3 -2
0
-4 -2 -1
0
-7 -5 -2
0
-6 -4 -2
0
-8 -3 -1
0
-4 -3 -1
0
0
0
0
0
0
0
0 -1
25
Table 5: The numbers of the irreducible invariants up to the 5th order and up to the
weight 10. They are presented in the form weight : (structure) number of invariants.
orders
2
3
2,3
4
2,4
3,4
2,3,4
5
2,5
3,5
2,3,5
2,4,5
3,4,5
2,3,4,
5
w : (s) n
2 : (2, 0, 0, 0) 1
6 : (0, 4, 0, 0) 1
4 : (1, 2, 0, 0) 1 6 : (3, 2, 0, 0) 1
4 : (0, 0, 2, 0) 1 6 : (0, 0, 3, 0) 1
4 : (2, 0, 1, 0) 1 4 : (2, 0, 2, 0) 1
8 : (0, 4, 1, 0) 1 9 : (0, 2, 3, 0) 1
6 : (1, 2, 1, 0) 2 7 : (2, 2, 1, 0) 2
9 : (1, 4, 1, 0) 2 9 : (2, 2, 2, 0) 3
10 : (2, 4, 1, 0) 1 10 : (3, 2, 2, 0) 1
10 : (0, 0, 0, 4) 1
6 : (1, 0, 0, 2) 1 8 : (3, 0, 0, 2) 1
7 : (0, 3, 0, 1) 1 8 : (0, 2, 0, 2) 2
5 : (1, 1, 0, 1) 1 6 : (2, 1, 0, 1) 1
8 : (4, 1, 0, 1) 1 9 : (1, 2, 0, 2) 2
10 : (2, 2, 0, 2) 3 10 : (3, 3, 0, 1) 2
8 : (1, 0, 1, 2) 2 9 : (2, 0, 1, 2) 2
6 : (0, 1, 1, 1) 1 8 : (0, 1, 2, 1) 2
10 : (0, 2, 1, 2) 5
7 : (1, 1, 1, 1) 3 8 : (2, 1, 1, 1) 4
10 : (1, 3, 1, 1) 7 10 : (2, 1, 2, 1) 6
26
9 : (3, 4, 0, 0) 1
9 : (3, 0, 3, 0) 1
10 : (0, 4, 2, 0) 2
8 : (1, 2, 2, 0) 3 8 : (3, 2, 1, 0) 1
9 : (4, 2, 1, 0) 1 10 : (1, 2, 3, 0) 2
10 : (5, 0, 0, 2) 1
9 : (0, 1, 0, 3) 1 10 : (0, 5, 0, 1) 1
7 : (3, 1, 0, 1) 1 8 : (1, 3, 0, 1) 2
9 : (2, 3, 0, 1) 2 10 : (1, 1, 0, 3) 2
10 : (1, 0, 2, 2) 4 10 : (3, 0, 1, 2) 2
9 : (0, 3, 1, 1) 3 10 : (0, 1, 3, 1) 2
9 : (1, 1, 2, 1) 5
10 : (4, 1, 1, 1) 2
9 : (3, 1, 1, 1) 3
Table 6: The numbers of the irreducible invariants with the highest 6th order and
the weight up to 10. They are presented in the form weight : (structure)
number of invariants.
orders
6
2,6
w : (s) n
6 : (0, 0, 0, 0, 2) 1
6 : (3, 0, 0, 0, 1) 1
10 : (4, 0, 0, 0, 2) 1
3,6
6 : (0, 2, 0, 0, 1) 1
4,6
8 : (0, 0, 1, 0, 2) 1
5,6
8 : (0, 0, 0, 2, 1) 1
2,3,6
7 : (1, 2, 0, 0, 1) 1
10 : (1, 2, 0, 0, 2) 3
2,4,6
6 : (1, 0, 1, 0, 1) 1
8 : (3, 0, 1, 0, 1) 1
9 : (4, 0, 1, 0, 1) 1
10 : (3, 0, 2, 0, 1) 1
3,4,6
8 : (0, 2, 1, 0, 1) 2
2,5,6
9 : (1, 0, 0, 2, 1) 1
3,5,6
7 : (0, 1, 0, 1, 1) 1
4,5,6
10 : (0, 0, 1, 2, 1) 3
2,3,4,6
9 : (1, 2, 1, 0, 1) 5
2,3,5,6
8 : (1, 1, 0, 1, 1) 3
3,4,5,6
9 : (0, 1, 1, 1, 1) 4
2,3,4,5,6 10 : (1, 1, 1, 1, 1) 11
8 : (2, 0, 0, 0, 2) 1
10 : (1, 0, 0, 0, 3) 1
9 : (0, 4, 0, 0, 1) 1
9 : (0, 0, 3, 0, 1) 1
10 : (0, 0, 2, 0, 2) 2
8 : (2, 2, 0, 0, 1) 2 9 : (3, 2, 0, 0, 1) 1
10 : (1, 4, 0, 0, 1) 2 10 : (4, 2, 0, 0, 1) 2
7 : (2, 0, 1, 0, 1) 1 8 : (1, 0, 2, 0, 1) 2
9 : (1, 0, 1, 0, 2) 1 9 : (2, 0, 2, 0, 1) 2
10 : (1, 0, 3, 0, 1) 2 10 : (2, 0, 1, 0, 2) 2
10 : (0, 2, 2, 0, 1) 4
10 : (2, 0, 0, 2, 1) 3
10 : (0, 1, 0, 1, 2) 2 10 : (0, 3, 0, 1, 1) 3
10 : (2, 2, 1, 0, 1) 6
9 : (2, 1, 0, 1, 1) 4 10 : (3, 1, 0, 1, 1) 4
27
Table 7: The numbers of the irreducible invariants with the highest 7th order and
the weight up to 10. They are presented in the form weight : (structure)
number of invariants.
orders
2,7
3,7
6,7
2,3,7
2,4,7
3,4,7
2,5,7
3,5,7
4,5,7
3,6,7
5,6,7
2,3,4,7
2,3,5,7
2,4,5,7
2,3,6,7
3,4,6,7
2,5,6,7
w : (s) n
8 : (1, 0, 0, 0, 0, 2) 1
10 : (0, 2, 0, 0, 0, 2) 2
10 : (0, 0, 0, 0, 1, 2) 1
7 : (2, 1, 0, 0, 0, 1) 1
9 : (4, 1, 0, 0, 0, 1) 1
10 : (1, 0, 1, 0, 0, 2) 2
7 : (0, 1, 1, 0, 0, 1) 1
7 : (1, 0, 0, 1, 0, 1) 1
10 : (4, 0, 0, 1, 0, 1) 1
9 : (0, 2, 0, 1, 0, 1) 2
8 : (0, 0, 1, 1, 0, 1) 1
8 : (0, 1, 0, 0, 1, 1) 1
9 : (0, 0, 0, 1, 1, 1) 1
8 : (1, 1, 1, 0, 0, 1) 2
10 : (3, 1, 1, 0, 0, 1) 4
10 : (1, 2, 0, 1, 0, 1) 5
9 : (1, 0, 1, 1, 0, 1) 3
9 : (1, 1, 0, 0, 1, 1) 3
10 : (0, 1, 1, 0, 1, 1) 4
10 : (1, 0, 0, 1, 1, 1) 3
10 : (3, 0, 0, 0, 0, 2) 1
8 : (3, 1, 0, 0, 0, 1) 1 9 : (1, 3, 0, 0, 0, 1) 2
10 : (2, 3, 0, 0, 0, 1) 3 10 : (5, 1, 0, 0, 0, 1) 1
9 : (0, 1, 2, 0, 0, 1) 2
8 : (2, 0, 0, 1, 0, 1) 1
10 : (0, 3, 1, 0, 0, 1) 3
9 : (3, 0, 0, 1, 0, 1) 1
10 : (0, 1, 0, 2, 0, 1) 2
10 : (0, 0, 2, 1, 0, 1) 2
9 : (2, 1, 1, 0, 0, 1) 3
10 : (2, 0, 1, 1, 0, 1) 4
10 : (2, 1, 0, 0, 1, 1) 4
28
10 : (1, 1, 2, 0, 0, 1) 5
Table 8: The numbers of the irreducible invariants with the highest 8th order and
the weight up to 10. They are presented in the form weight : (structure)
number of invariants.
orders
8
2,8
3,8
4,8
6,8
2,3,8
2,4,8
3,4,8
2,5,8
3,5,8
2,6,8
3,6,8
4,6,8
3,7,8
5,7,8
2,3,4,8
2,3,5,8
3,4,5,8
2,4,6,8
2,3,7,8
w : (s) n
8 : (0, 0, 0, 0, 0, 0, 2) 1
8 : (4, 0, 0, 0, 0, 0, 1) 1
10 : (0, 4, 0, 0, 0, 0, 1) 1
8 : (0, 0, 2, 0, 0, 0, 1) 1
10 : (0, 0, 0, 0, 2, 0, 1) 1
8 : (1, 2, 0, 0, 0, 0, 1) 1
8 : (2, 0, 1, 0, 0, 0, 1) 1
10 : (2, 0, 2, 0, 0, 0, 1) 2
9 : (0, 2, 1, 0, 0, 0, 1) 1
10 : (1, 0, 0, 2, 0, 0, 1) 2
8 : (0, 1, 0, 1, 0, 0, 1) 1
8 : (1, 0, 0, 0, 1, 0, 1) 1
10 : (0, 2, 0, 0, 1, 0, 1) 2
9 : (0, 0, 1, 0, 1, 0, 1) 1
9 : (0, 1, 0, 0, 0, 1, 1) 1
10 : (0, 0, 0, 1, 0, 1, 1) 1
10 : (1, 2, 1, 0, 0, 0, 1) 4
9 : (1, 1, 0, 1, 0, 0, 1) 2
10 : (0, 1, 1, 1, 0, 0, 1) 3
10 : (1, 0, 1, 0, 1, 0, 1) 3
10 : (1, 1, 0, 0, 0, 1, 1) 3
10 : (2, 0, 0, 0, 0, 0, 2) 1
10 : (0, 1, 0, 0, 0, 0, 2) 1 10 : (0, 0, 3, 0, 0, 0, 1) 1
9 : (2, 2, 0, 0, 0, 0, 1) 1 10 : (3, 2, 0, 0, 0, 0, 1) 2
9 : (1, 0, 2, 0, 0, 0, 1) 1 9 : (3, 0, 1, 0, 0, 0, 1) 1
10 : (4, 0, 1, 0, 0, 0, 1) 1
9 : (2, 0, 0, 0, 1, 0, 1) 1
10 : (2, 1, 0, 1, 0, 0, 1) 3
29
10 : (3, 0, 0, 0, 1, 0, 1) 1
Table 9: The numbers of the irreducible invariants with the highest 9th order and
the weight up to 10. They are presented in the form weight : (structure)
number of invariants.
orders
2,9
3,9
2,3,9
w : (s) n
10 : (1, 0, 0, 0, 0, 0, 0, 2) 1
9 : (0, 3, 0, 0, 0, 0, 0, 1) 1
9 : (3, 1, 0, 0, 0, 0, 0, 1) 1
10 : (4, 1, 0, 0, 0, 0, 0, 1) 1
3,4,9 10 : (0, 1, 2, 0, 0, 0, 0, 1) 1
2,5,9
9 : (2, 0, 0, 1, 0, 0, 0, 1) 1
3,5,9 10 : (0, 2, 0, 1, 0, 0, 0, 1) 1
4,5,9
9 : (0, 0, 1, 1, 0, 0, 0, 1) 1
3,6,9
9 : (0, 1, 0, 0, 1, 0, 0, 1) 1
5,6,9 10 : (0, 0, 0, 1, 1, 0, 0, 1) 1
2,7,9
9 : (1, 0, 0, 0, 0, 1, 0, 1) 1
4,7,9 10 : (0, 0, 1, 0, 0, 1, 0, 1) 1
3,8,9 10 : (0, 1, 0, 0, 0, 0, 1, 1) 1
2,3,4,9 9 : (1, 1, 1, 0, 0, 0, 0, 1) 1
2,4,5,9 10 : (1, 0, 1, 1, 0, 0, 0, 1) 2
2,3,6,9 10 : (1, 1, 0, 0, 1, 0, 0, 1) 2
10 : (1, 3, 0, 0, 0, 0, 0, 1) 1
10 : (3, 0, 0, 1, 0, 0, 0, 1) 1
10 : (2, 0, 0, 0, 0, 1, 0, 1) 1
10 : (2, 1, 1, 0, 0, 0, 0, 1) 2
Table 10: The numbers of the irreducible invariants with the highest 10th order
and the weight up to 10. They are presented in the form weight : (structure)
number of invariants.
orders
10
2,10
5,10
2,3,10
2,4,10
3,4,10
2,6,10
4,6,10
3,7,10
2,8,10
2,3,5,10
w : (s) n
10 : (0, 0, 0, 0, 0, 0, 0, 0, 2) 1
10 : (5, 0, 0, 0, 0, 0, 0, 0, 1) 1
10 : (0, 0, 0, 2, 0, 0, 0, 0, 1) 1
10 : (2, 2, 0, 0, 0, 0, 0, 0, 1) 1
10 : (3, 0, 1, 0, 0, 0, 0, 0, 1) 1 10 : (1, 0, 2, 0, 0, 0, 0, 0, 1) 1
10 : (0, 2, 1, 0, 0, 0, 0, 0, 1) 1
10 : (2, 0, 0, 0, 1, 0, 0, 0, 1) 1
10 : (0, 0, 1, 0, 1, 0, 0, 0, 1) 1
10 : (0, 1, 0, 0, 0, 1, 0, 0, 1) 1
10 : (1, 0, 0, 0, 0, 0, 1, 0, 1) 1
10 : (1, 1, 0, 1, 0, 0, 0, 0, 1) 1
30
from it. Let us suppose we have zero at a moment at some place. It can mean that there is
no other invariant with this structure, but it can also mean there is a dependency between
invariants and another irreducible invariant and we cannot decide, what possibility from
these two has occurred. The dependency with the least weight has weight 12, it is the
third in our tables and really, while Reiss [20] found 65 irreducible invariants up to the
4th order by means of the Cayley-Sylvester theorem, the actual number of them is 66,
the extra invariant has the same structure as the dependency. It means we must compute
all invariants up to some limit to find all irreducible ones between them and using the
Cayley-Sylvester theorem for the computation of the structure of all irreducible invariants
is unreliable for weights 12. On the other hand, we have weight limit 10 in these tables,
so we can verify the numbers of the invariants by means of the Cayley-Sylvester theorem
and it really agrees.
We generate the invariants by a way that we call the graph method. Let us consider an
image f and two arbitrary points (x1 , y1 ), (x2 , y2 ) from its support. Let us denote the
cross-product of these points as C12 :
C12 = x1 y2 x2 y1 .
I(f ) =
R QN
k,j=1
Ckjkj
QN
i=1
(72)
Note that it is meaningful to consider only j > k, because Ckj = Cjk and Ckk = 0.
After an affine transform, I becomes
I = J w |J|N I,
where w = k,j nkj is the weight of the invariant and N is the degree of the invariant.
w+N
If I is normalized by 00
we get a desirable affine invariant
P
I
w+N
00
=
31
I
w+N
00
(x1 y2 x2 y1 )2 f (x1 , y1 )f (x2 , y2 )dx1 dy1 dx2 dy2 = 2(m20 m02 m211 ).
(73)
I(f ) =
=
(x1 y2 x2 y1 )2 (x1 y3 x3 y1 )2 f (x1 , y1 )f (x2 , y2 )f (x3 , y3 )dx1 dy1 dx2 dy2 dx3 dy3
m220 m04 4m20 m11 m13 + 2m20 m02 m22 + 4m211 m22
4m11 m02 m31 + m202 m40 .
(74)
Each invariant generated by formula (72) can be represented by a planar connected
graph, where each point (xk , yk ) corresponds to one node and each cross-product Ckj
n
corresponds to one edge of the graph. If nkj > 1, the respective term Ckjkj corresponds
to nkj edges connecting k-th and j-th nodes. Thus, the number of nodes equals the
degree of the invariant and the total number of the graph edges equals the weight w of
the invariant. From the graph one can also learn about the orders of the moments the
invariant is composed from and about its structure. The number of edges originating
from each node equals the order of the moments involved, the number of nodes equals the
order of the invariant.
Particularly, for the invariants (73), (74) given above the corresponding graphs are
shown in Fig. 1 and the list of edges for (73) is
1 1
2 2
(75)
1 1 1 1
2 2 3 3.
(76)
Now one can see that the problem of derivation of the AMIs up to the given weight
w is equivalent to generating all connected graphs with at least two nodes and at most w
edges. Let us denote this set of graphs as Gw .
All graphs with 10 edges or less were generated and corresponding invariants were
computed. More precisely, if we need to generate all invariants of some weight w, it is
satisfactory to generate all graphs from
1 1 1 ... 1 1 1
2 2 2 ... 2 2 2
to
32
(77)
Figure 1: The graphs corresponding to the invariants (73) (top) and (74) (bottom)
1 1 3 4 ... w 2 w 1 w 1
2 3 4 5 ... w 1
w
w,
(78)
1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2
(79)
1 1 3 4 5 6 7 8 9 9
2 3 4 5 6 7 8 9 10 10.
(80)
to
33
1 1 3 3 ... w 3 w 3 w 1 w 1
2 2 4 4 ... w 2 w 2
w
w
(81)
1 1 1 ... 1 1 1
2 2 2 ... 2 3 3
(82)
1 1 3 3 ... w 4 w 4 w 3 w 3 w 3
2 2 4 4 ... w 3 w 3 w 2 w 2 w 2,
(83)
it. There were 1519 invariants generated after the elimination of the zero and identical
ones. First, the invariants were multiplied and invariants identical with some product were
eliminated. Then the invariants linearly dependent on other invariants were eliminated.
The rest of 362 irreducible invariants are published in the appendix.
The invariants are sorted by the highest order of moments. Another criterion is the
number of orders of moments (less first), then the second highest order, the third one etc.
If all orders are the same, then the invariants are sorted by their weights. The invariants
with the same orders of moments and the same weights are not sorted. The labelling
numbers are only temporary (from I11 ), because the invariants with higher weights can
be inserted. If some different (i.e. non-isomorfic) graphs generate identical invariants,
only one of them is published in this case.
Dependencies
Now we have 362 irreducible invariants, we can calculate that only 60 of them at maximum
can be independent by the rule of thumb and we would like to select them. There is a few
ways, how to do it. One approach is using the Cayley-Sylvester theorem for tipping invariants that may be independent, but it is both complicated and unreliable for higher orders,
see Section 3. Another approach can be computing dependencies among the invariants
and then eliminating the dependent invariants. The basic idea of looking for dependencies
is following: if we find an invariant that is both product and linearly dependent, then we
can write an equation: the product equals the linear dependency. Then we can substitute reducible invariants by irreducible. We are running up against two problems, the
dependencies can be dependent by similar way how the invariants themselves and weight
limit. The dependencies can be zero, identical, products and linearly dependent, they
can be eliminated, and also polynomially dependent and these dependencies would have
to be searched again. Hilbert [19] calls dependencies among dependencies second-order
dependencies. The number n of independent invariants is then
n = n0 n1 + n2 n3 + ,
(84)
where n0 is the number of irreducible invariants, n1 is the number of first-order dependencies, n2 is the number of second-order dependencies, etc. and Hilbert proves that this
chain is always finite, when the order of the invariants is finite. Another terminological remark: in [19], the dependency among invariants is called syzygy. After hesitation,
we decided to use general term dependency in this work for better understandability for
non-experts in invariants.
Another problem is weight limit. A weight of a product of two invariants is the sum of
their weights, so if we have some limit to the weights of the invariants, we would need to
compute the products up to a multiple of it and we do not know this limit in advance. We
computed the products up to the limit the computer enabled. Finally, we have computed
36
products until weight 15. The computing time exceeded 9 days and the original number
1519 invariants had to be appended by 17151 products. This computation was more timeand memory-consuming than that of generation of original invariants. In spite of this,
many important dependencies did not fit to the limit and remained hidden.
The algorithm was similar to that for elimination of reducible invariants from previous
section with small modifications. The limit of the weight of the product was changed
(from 10 to 15) and a list of invariants, which are simultaneously products and linearly
dependent, was created before their elimination. The found dependencies must be farther
processed. If some factor is also product, it must be substituted by its factors, if it is
linearly dependent, it must be substituted by this dependency so we obtain dependency
among irreducible invariants only. By this way, we obtain relatively complicated tree data
structure, which must be simplified. It is done by the special recursive algorithm. The
tree is searched from its root. If some product of sums is found, then the multiplication
is executed so one sum of products is obtained. If some factor is not mere invariant, but
still subtree, the same procedure is called recursively on it. The factors of the original
product are moved to the other side of the equation and identical terms are added so we
obtain the dependency in form of sum of products of irreducible invariants.
It often happens that no term remains after last addition and we obtain dependency
in form 0=0. Some dependencies are identical and some differ only in multiplicative
constant. All these dependencies are eliminated and only one is preserved from the set of
identical dependencies. We obtained 122 dependencies after this elimination in our case.
To remove non-integer coefficients, the special procedure was proposed. The tolerance
was set to 0.001. Its inverse value is 1000. The coefficients of the dependencies are divided
by all integers from 1 to this limit 1000 and the divider with minimum deviation is used
as denominator. So, all coefficients of the dependency are converted to fractions. Then,
least common multiple from all denominators is found and the dependency is multiplied
by it. Integer coefficients are the result.
The algorithm for elimination of the dependent dependencies is similar to that for
elimination of linearly dependent invariants. All products of dependencies and invariants
with weight limit 22 are computed (the limit 22 is the least, when all reducible dependencies are eliminated). Then the dependencies and products originated from products with
the same structure of orders of moments are found. The matrix of their coefficients is
created. The coefficients of one dependency are in one column and the coefficients of identical terms are in one row. The columns are removed until their number equals the rank
of the matrix. If the rank decreased, the removed column must be returned back. The
dependencies with the biggest number of terms are removed first. Some dependencies are
expressed in form invariant multiplied by another dependency. These dependencies are
removed last from the matrix, but they are eliminated from the set after this computation.
As the result, we have found second-order dependencies. We decided not to compute
higher-order dependencies, but to eliminate one first-order dependency for each second37
order dependency by a special heuristic algorithm: we find the first-order dependency with
maximum weight in each second-order dependency. If there is more such dependencies, we
find that with maximum number of terms. This dependency is eliminated and also each
multiple that contains it is eliminated from following search. 54 dependencies remained
after this algorithm, 12 of them of 4th order, 36 of the 5th order and 6 of the 6th order.
To append this number of dependencies, other computations with limited order of
moments were carried out to decrease the number of products. When the limit of the
order was 3 and the limit of the weight was 18, one new dependency was found. When
the limit of the order was 4 and the limit of the weight was 16, 194 new dependencies
were found after elimination of the identical ones. Another dependency was obtained,
when the selection conditions were defined as to use invariants of 2nd and 4th orders
only and the limit of the weight was 18. From these 196 dependencies 31 ones remained
after elimination of the dependent ones. It would be 85 dependencies together, but if 12
overlaying dependencies of 4th order are omitted, the result is 73 dependencies.
Now, the algorithm for elimination of the dependent dependencies in detail:
1. Compute the product of each pair invariant dependency and dependency dependency satisfying some conditions, e.g. the sum of weights is less than or equal
to some limit, in our case 22.
2. All dependencies with the same structure (the sum of the structures of the invariants in one term) are found. The linear combination of such dependencies can be
computed. The structures with minimum weight are processed first.
3. The matrix of all coefficients is constructed. It has coefficients of one dependency
in one column. The coefficient corresponding the term, which is not included in the
dependency, is zero. There are the coefficients of one term in different dependencies
in one row of the matrix. The rank of the matrix of coefficients equals the number
of the linearly independent dependencies of the current structure.
4. The dependencies of the current structure are divided into two groups: the products
invariant dependency and the others. The dependencies are also sorted by the
number of terms. Now, the dependency, which is not the product, with the maximum number of terms is removed from the matrix and its rank is checked. If it
decreased, then the dependency is returned to the matrix, else another dependency
is removed. If no other non-product dependency can be removed, then the products
are removed, that with the maximum number of terms first. The removing finishes,
when the matrix is regular, i.e. the number of columns equals its rank.
5. If there are some dependencies, which were not removed and are not products, then
they are linearly independent. All other dependencies are eliminated from the set.
38
6. If the removed dependencies are used as right side of the system of linear equations,
solving it we can obtain coefficients of the second order dependencies.
7. The second order dependencies, whose each term contains the same dependency or
invariant, are expunged. They were processed before.
8. In each second order dependency, the first order dependency with maximum weight
is found. If there is more such dependencies, we find that with maximum number
of terms. This dependency is eliminated and also each multiple that contains it is
eliminated from the following search.
9. The search is stopped, when all multiples are processed.
During multiplication of the dependencies in this algorithm we work with labels of the
invariants only, not with original coefficients and moments.
Now we have subtracted the second-order dependencies from the first-order ones and
still too much dependencies remained. More precisely, we have 31 dependencies of the 3rd
and 4th orders for 32 invariants with 9 of them independent. Thereto the invariant I14
does not occur in the dependencies, therefore only 22 dependencies are useful. We decided
not to compute higher-order dependencies, but to eliminate the redundant dependencies
by the following heuristic algorithm in which each invariant supposed to be dependent
chooses its dependency. The dependencies were sorted by the highest order, by the number
of orders, by the weight and by the number of terms. Then each invariant gets assigned
the first dependency, which it is included in. When all such dependencies are occupied,
some invariant gets assigned a substitutional dependency. The dependencies without any
assignation are then omitted.
The remainder 64 dependencies are published in the appendix only for the sake of
interest. In detail, there is 1 dependency of the 3rd order, 21 dependencies of the 4th
order, 36 dependencies of the 5th order and 6 dependencies of the 6th order. The set is
incomplete and cannot be used for selection of the independent invariants even the 4th
order.
There is another approach to the selection of independent invariants. We can select some
of them and verify their independency by comparison with the normalized moments.
Firstly, we should choose irreducible invariants only and respect the rule of thumb not
only for the whole number of chosen invariants, but also for each subset. Let us suppose
we have no restriction on the order of the invariants. For instance, we have 10 moments
up to the 3rd order, i.e. 10-6=4 independent invariants. We begin with I1 , I2 , I3 and
I4 . Now we would like to insert invariants of 4th order. There is 5 moments of 4th order
plus 3 ones of zero and first orders, i.e. 8-6=2 homogenous invariants of of 4th order at
39
maximum (I6 and I7 ). If we take simultaneous invariants of the 4th and 2nd orders, there
are 11 moments, i.e. 11-6=5 invariants. The invariants used must be subtracted from
this number, if we use I1 , I6 and I7 , then 2 new invariants remain (I8 and I9 ). I10 must
necessarily be dependent and we need at least one simultaneous invariant of orders 3 and
4 or 2, 3 and 4.
Generally, If we have invariants up to the order r 1 and need to insert invariants
of the order r, we need to insert r + 1 invariants. The r of them should be homogenous
or simultaneous of orders r and 2, with maximally r 2 of them homogenous and one
simultaneous of order r and some order higher than 2. We choose the invariants with
minimum weight. In Table 11, there are the maximum numbers of the independent
invariants that can be constructed from moments of specific orders. There are the whole
numbers there, e.g. the value 9 from orders 2,3,4 means the whole number of the invariants
including homogeneous ones, simultaneous invariants from two orders and simultaneous
ones from three orders.
Table 11: The maximum numbers of the independent invariants up to the 5th order by
the rule of thumb.
orders
2 3 2,3 4 2,4 3,4
numbers 1 1 4 2 5
6
2,3,4 5 2,5
9
3 6
The proof of independency and completeness of the chosen set is difficult and needs
another mathematical tools than the previous sections. Its main idea: if we have another
complete and independent set of affine invariants and we can compute unambiguously
values of our invariants from them and them from our invariants, then our set is complete
and independent too. This alternative set can be set of normalized moments. We can
normalize geometric moments, but we obtain simpler equations, when we use complex
moments. The complex moment cpq of order (p + q) is defined as
cpq =
(85)
where i denotes imaginary unit. Each complex moment can be expressed in terms of
geometric moments mpq as
cpq =
p X
q
X
k=0 j=0
p
k
q
(1)qj ip+qkj mk+j,p+qkj .
j
(86)
If we use central moments pq instead of the mpq in (86), we obtain complex moments
normalized to translation.
To express geometric moments in terms of complex moments, we can substitute new
variables for (x + iy) and (x iy) into (85) and the definition of the geometric moments
40
mpq =
2p+q iq
k=0 j=0
p
k
q
(1)qj ck+j,p+qkj .
j
(87)
(88)
It follows immediately from (88) that cpq = cqp (the asterisk denotes complex conjugate). After the rotation by an angle the complex moment becomes
cpq = ei(pq) cpq .
(89)
n
Y
ckpiiqi
(90)
i=1
is invariant to rotation, if
n
X
i=1
ki (pi qi ) = 0,
u=x
v = y + .
(92)
u = x
v = y.
(93)
u = x cos y sin
v = x sin + y cos .
(94)
u = x
v = 1 y.
(95)
u = x cos y sin
v = x sin + y cos .
(96)
Scaling
First rotation
Stretching:
Second rotation:
41
The moments are successively normalized to these elementary transforms. The normalization to the translation is done by translation the origin of the coordinate system
to the centroid of the image (using central moments in (86)). The normalization to the
scaling is similar to that of AMIs, we use ratios
p+q+2
cpq /c00 2 .
(97)
The complex moment c20 can be used for normalization to the first rotation. The
straight line passing through the centroid on the angle equaling a half of the phase of c20
is called principle axis. If an image is rotated by angle so the principal axis agrees with
the axis x, then the c20 is real and positive. It holds
sin = s
cos =
q2
c
20 +c02
),
2 c20 c02
c
+c
+ 2 20c20 c0202 )
(1
1
(1
2
02
s = sign( c20 c
)
i
(98)
211
(c20 )
c20 c02
=
=
,
20 02
(c20 )
i(c20 + c02 )
(99)
where (c20 ) and (c20 ) are real and imaginary parts of the c20 .
We can normalize the image to the stretching by scaling by the coefficient
v
u
u c c20 c02
4t 11
=
(100)
(101)
c11 +
c20 c02
horizontally and by the coefficient 1/ vertically. Now, the image is normalized to the
affine transform up to the second rotation. During this normalization, the c10 , c01 , c20 as
well as c02 become zero, the c00 becomes one and rotation invariants from (90) become
affine invariants. We can either continue with the normalization to the second rotation
by the c21 or to use . It emerged that the latter possibility leads to simpler equations,
because we can choose the most appropriate set of s.
There is a question why to use affine moment invariants I computed from geometric
moments and not these normalized rotational invariants . One reason could be better
numerical behavior in some situations, another one easier computation; if we unite all
expressions for normalization to one formula, it is more complicated than that for AMIs.
We can compute values of the affine moment invariants directly from the complex
moments by means of the following theorem.
Theorem 2: Let us denote the value of the invariant computed from the geometric
moments I(pq ) and I(cpq ) the value obtained by substitution the complex moments cpq
instead of pq . Then there is relation between them
(102)
Its Jacobian J = 2i. According to (85) the value of the moment after the transform
pq =
(103)
(the coordinates of the centroid remains unchanged). From Theorem 1 we have for the
values of the invariants without normalization by 00
I(pq ) = J w |J|k I(pq )
(104)
(105)
=
=
=
=
c211
c230 c203 + 6c30 c21 c12 c03 4c30 c312 4c321 c03 + 3c221 c212
c11 c30 c03 + c11 c21 c12
2c311 c30 c03 + 6c311 c21 c12 .
(106)
= 2 I1
= 1I1 (2I3 II41 )
= 1I1 (6I3 + II41 )
c11
c21 c12
c30 c03
I32
I1
I42
I13
(107)
.
The set c11 , c21 c12 , c30 c03 and (c30 c312 ) is used as the other set of invariants. This system
of rotation invariants is independent and complete except the sign of (c30 c312 ). It relates with fact that the set I1 , I2 , I3 , I4 cannot distinguish two objects differing by mirror
reflection. The general affine transform include the mirror reflection, therefore the set
I1 , I2 , I3 , I4 is complete from this point of view, obversely, we should use absolute values
of the invariants with odd weights. If we insert I5 there, we can compute
4I5
(c30 c312 ) = 3 ,
( I1 )
(108)
(109)
I8
I1
I82
I12
c22
c31 c13
c40 c04
= 16I6 +
(c40 c213 ) =
(c31 c212 ) =
4 II19
I82
I12
32I7 + 8 I6I1I8
1 (2 I3 I8
I1
I1
I9
I1
(111)
I83
I13
12 II8 I29 +
1
I4 I8
2 II221 ) .
I2
1
So the set c22 , c31 c13 , c40 c04 , (c40 c213 ) and (c31 c212 ) together with the invariants of the
second and third orders is used as the other set of invariants. Absolute values of the
vanishing imaginary parts can be computed
|(c40 c213 )| =
|(c31 c212 )| =
(112)
(113)
8I10
(c40 c213 ) = 3 ,
( I1 )
(114)
From it
but we know the absolute value of it from (112). It means the I10 can have only one of
two values given by other invariants, it is dependent, while the phase of the c31 c212 cannot
be computed.
We may want to use I16 instead of I22 because of less weight, then the last equation
from (110) becomes
(2i)6 I16 = 2c11 c30 c12 c13 2c11 c21 c03 c31 + 2c11 c30 c03 c22 2c11 c21 c12 c22
+2c11 c221 c13 + 2c11 c212 c31
(115)
I16
c3 c03
c30 c3
= c31 c212 ( 221 2 1) + c13 c221 ( 2 12
1) c30 c03 c22 + c21 c12 c22
c11
c21 c12
c21 c212
44
(116)
or
32
c30 c3
c30 c312
I16
2
= 2(c31 c212 )( 2 12
)(
1)
+
2(c
c
1) c30 c03 c22 + c21 c12 c22 . (117)
31
12
c11
c21 c212
c221 c212
c c3
c c3
30 12
I16
2
2 (c31 c212 )(2 ( c230 c212 1) + 2 ( c230 c12
+ c30 c03 c22
2 1)) (c31 c12 )( c2 c2 1)(32 c
11
21 12
21 12
c c3
21 12
(118)
I16
12
( c30
+ c30 c03 c22 c21 c12 c22 )
2 c2 1)(32 c
11
21 12
(c31 c212 ) =
c30 c312
c221 c212
v
u
u (32 I16 + c30 c03 c22 c21 c12 c22 )2
u
c11
3
1)t
2 2
2 c30 c12
2
21 c12
c c312
2(2 ( c30
2 c2
21 12
1) +
c c3
c c3
2 ( c230 c12
2
21 12
21 12
1))
. (119)
You can see that a real solution always exists, i.e. the set set I1 , I2 , I3 , I4 , I6 , I7 , I8 , I9 , I16
is independent, but also increasing complexity of the equations. If we use I11 instead
of I16 (e.g. it has moments of two orders only), the final equation would be quartic. It
leads to effort for simplification of the equations. If we need not have general formulas,
but only dependency test, the solution in some points might be satisfactory. We obtain
simple equations, if we choose the values of the affine invariants so the values of rotation
invariants would be 1.
It relates with the fact that the sum of the coefficients of an invariant is zero. The
invariant can be written as a weighted sum of products of moments
I=
t
X
j=1
cj
s
Y
pj ,qj .
(120)
=1
The t is the number of terms and the s is the degree of the invariant. If we substitute it
into Cayley - Aronhold differential equation (14), we obtain
n
X
i=1
pi pi 1,qi +1
t
X
cj e(pi , qi , j)
j=1
1
pi ,qi
s
Y
pj ,qj = 0 ,
(121)
=1
where e(pi , qi , j) is the number of occurrences of the pi ,qi in the j-th term. The order of
the summation can be exchanged
n
t X
X
j=1 i=1
But
pi pi 1,qi +1 cj e(pi , qi , j)
n
X
1
pi ,qi
pi e(pi , qi , j)
i=1
45
s
Y
pj ,qj = 0 ,
(122)
=1
(123)
equals the weight w of the invariant. If we substitute the same value, let us say , for all
moments, we obtain
s w
t
X
cj = 0
(124)
j=1
or the sum of the coefficients of an invariant is zero. It is an easy test, if some polynomial
can be affine invariant, and it also means that if all moments have the same values, then
all invariants are zero, and vice versa. In our case, we must compensate the fact that
11 = 0 because of the normalization by non-zero values of some invariants.
In our case, if we choose
1
1
I1 = , I2 = 0, I3 = 0, I4 = ,
4
8
then we obtain
c11 = 1, c21 c12 = 1, c30 c03 = 1, c30 c312 = 1.
Similarly, if
1
I6 = 0, I7 = 0, I8 = , I9 = 0,
4
then
c22 = 1, c31 c13 = 1, c40 c04 = 1, c40 c213 = 1.
Now, if we use I10 , then the last equation becomes
(c40 c213 ) = 64I10 ,
but from the previous equations we have
(c40 c213 ) = 0 ,
therefore we cannot choose I10 freely, it is dependent. If we use I22 instead of I10 , then
the last equation becomes
(c31 c212 ) = 1 16I22 .
We can choose I22 freely, if I22 = 0, then c31 c212 = 1. The I22 is independent. If we use
I16 , then the last equation becomes I16 = 0. It looks like the I16 would be dependent, but
if we change e.g. I2 = 41 , then c30 c312 = 1 and we obtain
(c31 c212 ) = 8I16 .
The I16 can be chosen freely, it is independent. If we look at the general formula (119), we
found that the values of the affine invariants must be chosen so the denominator would
be non-zero. This is the biggest problem with limited solution in one point. There are
singular points in the space of invariants and if we choose such a point, the dependency
test fails in spite of the invariants are independent.
46
The I11 is similar to I16 in this sense. For I2 = 0 we obtain last equation I11 = 0 like
it would be dependent, but for I2 = 14 it becomes
256I11 = 82 (c31 c212 ),
I11 can be chosen freely and it means it is independent. In reality, it must be from 0 to
1
, but it is whole interval, not only finite number of values, therefore it is satisfactory.
32
The following analysis of higher orders will not be so detailed, we will describe one
choice of invariants with one test of independency only (except 8th order). We will
suppose choice I2 = 41 and I16 = 18 , i.e. c30 c312 = 1, c31 c212 = 1 and from before
c11 = 1, c21 c12 = 1, c30 c03 = 1, c22 = 1, c31 c13 = 1, c40 c04 = 1 and c40 c213 = 1.
In the 5th order, we choose I33 , I34 , I35 , I36 , I42 , I43 . We obtain equations
(2i)10 I33 = c250 c205 + 10c50 c05 c41 c14 4c50 c05 c32 c23 16c50 c32 c214 16c241 c23 c05
9c241 c214 + 12c50 c223 c14 + 12c41 c232 c05 + 76c41 c14 c32 c23 48c41 c323
48c332 c14 + 32c232 c223
(2i)6 I34 = c11 c50 c05 + 3c11 c41 c14 2c11 c32 c23
(2i)8 I35 = 4c311 c41 c14 + 4c311 c32 c23
(2i)10 I36 = 2c511 c50 c05 + 10c511 c41 c14 + 20c511 c32 c23
(2i)5 I42 = 2c11 c03 c41 + 2c11 c30 c14 + 6c11 c12 c32 6c11 c21 c23
(2i)6 I43 = 2c211 c03 c41 + 2c211 c30 c14 2c211 c12 c32 2c211 c21 c23 .
(125)
1
From I34 = 0, I35 = 0, I36 = 32 we have
c50 c05 = 1, c41 c14 = 1, c32 c23 = 1
and from I42 = 0, I43 = 0 we obtain two solutions
a)
c41 c03 = 1,
b) c41 c03 = 1,
c32 c12 = 1
c32 c12 = 1
47
In the 6th order, we choose I112 , I113 , I114 , I115 , I116 , I117 , I124 . We obtain equations
(2i)6 I112
(2i)6 I113
(2i)8 I114
(2i)10 I115
I117 = 12I
q 115
I124 = 21 I115 (1 + 32I115 )
1
and we can choose I115 freely in the interval < 32
, 0 >, the set is independent. We will
use the values I115 = 0, I117 = 0, I124 = 0 and c51 c224 = 1 for future computations of the
other set of invariants c33 , c42 c24 , c51 c15 , c60 c06 , (c42 c212 ), (c60 c324 ) and (c51 c224 ).
48
In the 7th order, we choose I205 , I206 , I210 , I211 , I212 , I213 , I214 , I215 . We obtain equations
(2i)8 I205
(2i)10 I206
(2i)7 I210
(2i)8 I211
(2i)9 I212
=
=
=
=
=
c11 c70 c07 + 5c11 c61 c16 9c11 c52 c25 + 5c11 c43 c34
4c311 c61 c16 + 12c311 c52 c25 8c311 c43 c34
4c211 c30 c25 + 12c211 c21 c34 12c211 c12 c43 + 4c211 c03 c52
4c311 c30 c25 + 4c311 c21 c34 + 4c311 c12 c43 4c311 c03 c52
2c11 c230 c21 c07 2c11 c230 c12 c16 12c11 c30 c221 c16 + 24c11 c30 c21 c12 c25
4c11 c30 c21 c03 c34 12c11 c30 c212 c34 + 4c11 c30 c12 c03 c43 + 18c11 c321 c25
54c11 c221 c12 c34 + 12c11 c221 c03 c43 + 54c11 c21 c212 c43 24c11 c21 c12 c03 c52
+2c11 c21 c203 c61 18c11 c312 c52 + 12c11 c212 c03 c61 2c11 c12 c203 c70
9
(2i) I213 = 2c11 c230 c12 c16 2c11 c230 c03 c25 2c11 c30 c221 c16 4c11 c30 c21 c12 c25
+8c11 c30 c21 c03 c34 + 4c11 c30 c212 c34 8c11 c30 c12 c03 c43 + 2c11 c30 c203 c52
+6c11 c321 c25 12c11 c221 c12 c34 4c11 c221 c03 c43 + 12c11 c21 c212 c43
+4c11 c21 c12 c03 c52 2c11 c21 c203 c61 6c11 c312 c52 + 2c11 c212 c03 c61
(2i)9 I214 = 16c411 c21 c34 + 16c411 c12 c43
(2i)10 I215 = 2c211 c230 c21 c07 2c211 c230 c12 c16 8c211 c30 c221 c16 + 12c211 c30 c21 c12 c25
4c211 c30 c212 c34 + 6c211 c321 c25 6c211 c221 c12 c34 4c211 c221 c03 c43
6c211 c21 c212 c43 + 12c211 c21 c12 c03 c52 2c211 c21 c203 c61 + 6c211 c312 c52
8c211 c212 c03 c61 + 2c211 c12 c203 c70
(127)
From I214 = 0, I210 = 0, I211 = 0, I206 = 0 and I205 = 0 we have
(c43 c12 ) = 0, c52 c03 = c43 c12 , i.e. c52 c25 = c43 c34 , c61 c16 = c43 c34 , c70 c07 = c43 c34 .
1
.
128
256I215 = (c70 c12 c203 ) 3(c61 c212 c03 ) + 3(c52 c03 ) (c43 c12 )
or
q
2
65536I215
+1
=
.
512I215
1
We can choose I215 freely, the set is independent. We will use the values I215 = 256
,
2
2
c43 c12 = 1, c52 c03 = 1, c61 c12 c03 = 1 and c70 c12 c03 = i in the future of the other set of
invariants c43 c12 , c52 c03 , c61 c212 c03 and c70 c12 c203 (It is 8 real values).
The 8th order is an example, where the first choice is dependent, in spite of it satisfies
the rule of thumb. If we choose the set I280 , I281 , I282 , I283 , I284 , I288 , I289 , I290 , I291 , then
49
the equation for I284 does not bring any new information, the value of the I284 cannot be
chosen, while the equation for relation between c62 c26 and c53 c35 vanish. If we choose the
set I280 , I281 , I282 , I283 , I288 , I289 , I290 , I291 , I294 instead, we obtain equations
(2i)8 I280
(2i)8 I281
(2i)10 I282
(2i)10 I283
=
=
=
=
(2i)8 I288
(2i)9 I289
(2i)10 I290 =
(2i)10 I291 =
(2i)9 I294
(128)
1
we have c44 = 1, from I282 = 0 we have c71 c17 = 6c62 c26 15c53 c35 + 10,
From I281 = 16
1
and from I280 = 0 we have c80 c08 = 20c62 c26 64c53 c35 + 45. From I291 = 32
we obtain
2
(c53 c12 ) = 1, from I290 = 0 we obtain (c62 c12 c03 ) = 1, from I288 = 0 we obtain
1
we obtain (c80 c212 c203 ) = 1. The last two equations
(c71 c203 ) = 1 and from I283 = 512
50
(2i)10 I328 = c11 c90 c09 + 7c11 c81 c18 20c11 c72 c27 + 28c11 c63 c36 14c11 c54 c45
(2i)9 I329 = c330 c09 + 9c230 c21 c18 9c230 c12 c27 + 3c230 c03 c36 27c30 c221 c27
+54c30 c21 c12 c36 18c30 c21 c03 c45 27c30 c212 c45 + 18c30 c12 c03 c54
3c30 c203 c63 + 27c321 c36 81c221 c12 c45 + 27c221 c03 c54 + 81c21 c212 c54
54c21 c12 c03 c63 + 9c21 c203 c72 27c312 c63 + 27c212 c03 c72 9c12 c203 c81
+c303 c90
(2i)9 I330 = 8c311 c30 c36 24c311 c21 c45 + 24c311 c12 c54 8c311 c03 c63
(2i)10 I331 = c11 c330 c09 + 7c11 c230 c21 c18 5c11 c230 c12 c27 + c11 c230 c03 c36
15c11 c30 c221 c27 + 18c11 c30 c21 c12 c36 2c11 c30 c21 c03 c45 3c11 c30 c212 c45
2c11 c30 c12 c03 c54 + c11 c30 c203 c63 + 9c11 c321 c36 9c11 c221 c12 c45
3c11 c221 c03 c54 9c11 c21 c212 c54 + 18c11 c21 c12 c03 c63 5c11 c21 c203 c72
+9c11 c312 c63 15c11 c212 c03 c72 + 7c11 c12 c203 c81 c11 c303 c90
10
(2i) I332 = 8c411 c30 c36 8c411 c21 c45 8c411 c12 c54 + 8c411 c03 c63
(2i)9 I334 = 4c211 c50 c27 + 20c211 c41 c36 40c211 c32 c45 + 40c211 c23 c54
20c211 c14 c63 + 4c211 c05 c72
10
(2i) I335 = 4c311 c50 c27 + 12c311 c41 c36 8c311 c32 c45 8c311 c23 c54
+12c311 c14 c63 4c311 c05 c72
9
(2i) I340 = 2c11 c70 c18 14c11 c61 c27 + 42c11 c52 c36 70c11 c43 c45 + 70c11 c34 c54
42c11 c25 c63 + 14c11 c16 c72 2c11 c07 c81
10
(2i) I341 = 2c211 c70 c18 10c211 c61 c27 + 18c211 c52 c36 10c211 c43 c45 10c211 c34 c54
+18c211 c25 c63 10c211 c16 c72 + 2c211 c07 c81
(2i)9 I344 = 2c11 c30 c40 c18 8c11 c30 c31 c27 + 12c11 c30 c22 c36 8c11 c30 c13 c45
+2c11 c30 c04 c54 6c11 c21 c40 c27 + 24c11 c21 c31 c36 36c11 c21 c22 c45
+24c11 c21 c13 c54 6c11 c21 c04 c63 + 6c11 c12 c40 c36 24c11 c12 c31 c45
+36c11 c12 c22 c54 24c11 c12 c13 c63 + 6c11 c12 c04 c72 2c11 c03 c40 c45
+8c11 c03 c31 c54 12c11 c03 c22 c63 + 8c11 c03 c13 c72 2c11 c03 c04 c81
(129)
From I330 = 0 we have (c63 c03 ) = 3(c54 c12 ), from I332 = 0 we have (c63 c03 ) =
(c54 c12 ), from I334 = 0 we have
(c72 c05 ) = 5(c63 c03 ) 10(c54 c12 ) = 5(c54 c12 ) .
From I335 = 0 we have
(c72 c05 ) = 3(c63 c14 ) 2(c54 c23 ) = (c54 c12 ) .
From I340 = 0 we have
(c81 c07 ) = 7(c72 c05 ) 21(c63 c03 ) + 35(c54 c12 ) = 7(c54 c12 ) .
5
From I341 = 256
we have
(c81 c07 ) = 10 + 5(c72 c16 ) 9(c63 c25 ) + 5(c54 c34 ) = 10 9(c54 c12 ) .
From I329 =
9
512
we have
(c90 c303 ) = 9 + 9(c81 c07 ) + 18(c72 c05 ) + 30(c63 c03 ) 72(c54 c12 )
= 81 81(c54 c12 ) + 72(c54 c12 ) .
51
11
we have
From I331 = 512
(c90 c303 ) = 11 7(c81 c07 ) + 10(c72 c05 ) + 10(c63 c03 ) 8(c54 c12 )
= 11 49(c54 c12 ) + 12(c54 c12 ) .
1
From I344 = 128
we have
The set is independent, but not complete, we cannot compute imaginary parts of the
variables unambiguously. The other set of invariants c55 , c64 c212 , (c73 c12 c03 ), (c82 c203 ),
(c91 c08 ) and (c10,0 c205 ) correspondingly is not complete.
Conclusion
The report deals with affine moment invariants. It shows relation between graphs and
affine moment invariants and uses the graph approach to the generation of the invariants.
It is simpler for programming than traditional solution of a system of linear equations
derived from Cayley - Aronhold differential equation and more precise, but little bit slower.
The report also explains difficulties with the rule of thumb (16) in case of the second order
moments.
The possibilities of computation of the number of irreducible and independent invariants by means of Cayley-Sylvester theorem are described. It is perfect tool for tests of
linear independency, but in case of polynomial independency it is unreliable, particularly
if the invariants are non-homogeneous. Nevertheless, if the weight of the invariants is not
higher than 11, as 10 is in our case, the results of it agree with other methods.
The method for elimination of reducible invariants is described. The computational
complexity of the similar method for elimination of irreducible invariants with polynomial
dependency is too high, therefore only partial results are presented. Instead, an analytical
test by means of normalized rotation invariants from complex moments is showed. If we
have not any restrictions on the order of the invariants, then the following set of 56
independent invariants passed through this test
I1 , I2 , I3 , I4 , I6 , I7 , I8 , I9 , I16 , I33 , I34 , I35 , I36 , I42 , I43 , I112 , I113 , I114 , I115 , I116 , I117 , I124 , I205 ,
I206 , I210 , I211 , I212 , I213 , I214 , I215 , I280 , I281 , I282 , I283 , I288 , I289 , I290 , I291 , I294 , I328 , I329 , I330 ,
I331 , I332 , I334 , I335 , I340 , I341 , I344 , I351 , I352 , I353 , I354 , I356 , I358 , I361 .
If there are some restrictions on the invariants, a user can choose his/her own set
with keeping some rules. The irreducible invariants should be chosen only, the whole set
and each its subset should satisfy the rule of thumb (16). An attempt at an analytical
normalized rotation invariant test can be done.
In the CD - ROM enveloped, there are two files with invariants, afinv10irred.txt
includes all 362 irreducible invariants published in this report in electronic form,
afinv10indep.txt includes independent 56 of them. readinv.m, cm.m and cafmi.m
are Matlab m-files for reading the invariants from these files, computing moments from
an image and for computing the values of the invariants. readinv.c includes c-language
function that selects invariants by some criterion. It can be rewritten for some other type
of processing.
We hope this tables of affine moment invariants will serve as useful aid for recognition
of affinely deformed objects.
53
References
[1] M. K. Hu, Visual pattern recognition by moment invariants, IRE Trans. Information Theory, vol. 8, pp. 179187, 1962.
[2] S. Maitra, Moment invariants, Proc. IEEE, vol. 67, pp. 697699, 1979.
[3] T. C. Hsia, A note on invariant moments in image processing, IEEE Trans. Syst.
Man Cybern., vol. 11, pp. 831834, 1981.
[4] M. R. Teague, Image analysis via the general theory of moments, J. Optical Soc.
of America, vol. 70, pp. 920930, 1980.
[5] Y. S. Abu-Mostafa and D. Psaltis, Recognitive aspects of moment invariants, IEEE
Trans. Pattern Analysis and Machine Intelligence, vol. 6, pp. 698706, 1984.
[6] C. H. Teh and R. T. Chin, On image analysis by the method of moments, IEEE
Trans. Pattern Analysis and Machine Intelligence, vol. 10, pp. 496513, 1988.
[7] S. O. Belkasim, M. Shridhar, and M. Ahmadi, Pattern recognition with moment
invariants: a comparative study and new results, Pattern Recognition, vol. 24,
pp. 11171138, 1991.
[8] A. Khotanzad and Y. H. Hong, Invariant image recognition by Zernike moments,
IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 12, pp. 489497, 1990.
[9] J. Flusser, On the independence of rotation moment invariants, Pattern Recognition, vol. 33, pp. 14051410, 2000.
[10] L. Van Gool, T. Moons, E. Pauwels and A. Oosterlinck, Vision and Lies approach to
invariance, invited paper for Image and Vision Computing, vol. 13, No. 4, pp. 259
277, 1995.
[11] T. Suk and J. Flusser, Projective moment invariants, IEEE Trans. Pattern Analysis
and Machine Intelligence, vol. 26, No. 10, pp. 13641367, 2004.
[12] T. H. Reiss, The revised fundamental theorem of moment invariants, IEEE Trans.
Pattern Analysis and Machine Intelligence, vol. 13, pp. 830834, 1991.
[13] J. Flusser and T. Suk, Pattern recognition by affine moment invariants, Pattern
Recognition, vol. 26, no. 1, pp. 167174, 1993.
[14] J. Flusser and T. Suk, Pattern Recognition by Means of Affine Moment Invariants,
Praha, 1991.
Tech. Rep. 1726, UTIA
AV CR,
54
[15] J. J. Sylvester, Tables of the generating functions and groundforms for the binary
quantics of the first ten orders, American Journal of Mathematics, vol. 2, pp. 223
251, 1879.
[16] J. J. Sylvester, Tables of the generating functions and groundforms for simultaneous binary quantics of the first four orders taken two and two together, American
Journal of Mathematics, vol. 2, pp. 293306, 324329, 1879.
[17] I. Schur, Vorlesungen u
ber Invariantentheorie. Berlin: Springer, 1968.
[18] G. B. Gurevich, Foundations of the Theory of Algebraic Invariants. Groningen, The
Netherlands: Nordhoff, 1964.
[19] D. Hilbert, Theory of Algebraic Invariants. Cambridge University Press, 1993.
[20] T. H. Reiss, Recognizing Planar Objects using Invariant Image Features, vol. 676 of
LNCS. Berlin: Springer, 1993.
[21] D. Shen and H. H. S. Ip, Generalized affine invariant image normalization, IEEE
Trans. Pattern Analysis and Machine Intelligence, vol. 19, pp. 431440, 1997.
[22] J. Flusser and T. Suk, Normalization approach to the Affine Moment Invariants,
Advanced Concepts for Intelligent Vision Systems. Proceedings. ( Lecture Notes in
Computer Science. 3708). Berlin: Springer, 2005, pp. 100-107.
[23] J. Flusser and T. Suk, Construction of complete and independent systems of rotation moment invariants, Computer Analysis of Images and Patterns. Proceedings.(
Lecture Notes in Computer Science. 2756). Berlin: Springer, 2003, pp. 41-48.
55
56
I4 = (320 203 + 6220 11 12 03 3220 02 212 620 211 21 03 620 211 212
+1220 11 02 21 12 320 202 221 + 2311 30 03 + 6311 21 12
6211 02 30 12 6211 02 221 + 611 202 30 21 302 230 )/11
00
weight=6
structure: 3,2
Generating graph:
1 1 1 2 3 4
2 3 4 5 5 5
weight=9
structure: 3,4
Generating graph:
1 1 1 2 2 3 3 4 4
2 3 4 5 5 6 6 7 7
57
58
weight=9
structure: 3,0,3
Generating graph:
1 1 1 1 2 2 2 3 3
2 3 3 4 4 5 5 6 6
59
60
structure: 0,4,2
Generating graph:
1 1 1 1 2 2 3 3 4 4
2 3 4 5 5 5 6 6 6 6
weight=6
structure: 1,2,1
Generating graph:
1 1 1 1 2 2
2 3 3 4 4 4
62
63
1 1 1 1 2 2 3
2 3 4 4 5 5 5
64
Generating graph:
1 1 1 1 2 2 3 3
2 3 4 4 5 5 5 5
66
67
weight=9
structure: 2,2,2
Generating graph:
1 1 1 1 2 2 2 3 3
2 3 4 4 5 5 5 6 6
68
weight=9
structure: 1,4,1
Generating graph:
1 1 1 1 2 2 3 3 4
2 3 4 4 5 5 6 6 6
1 1 1 1 2 2 3 3 4
2 3 4 5 5 6 6 6 6
71
72
73
75
76
weight=6
structure: 1,0,0,2
Generating graph:
1 1 1 1 1 2
2 2 2 2 3 3
Generating graph:
1 1 1 1 1 2 2 2
2 2 3 3 4 4 5 5
I36 = (520 205 + 10420 11 14 05 5420 02 214 20320 211 23 05 20320 211 214
+40320 11 02 23 14 10320 202 223 + 20220 311 32 05 + 60220 311 23 14
60220 211 02 32 14 60220 211 02 223 + 60220 11 202 32 23
10220 302 232 1020 411 41 05 4020 411 32 14 3020 411 223
+4020 311 02 41 14 + 12020 311 02 32 23 6020 211 202 41 23
6020 211 202 232 + 4020 11 302 41 32 520 402 241 + 2511 50 05
+10511 41 14 + 20511 32 23 10411 02 50 14 40411 02 41 23
30411 02 232 + 20311 202 50 23 + 60311 202 41 32 20211 302 50 32
20211 302 241 + 1011 402 50 41 502 250 )/17
00
weight=10
structure: 5,0,0,2
Generating graph:
1 1 1 1 1 2 3 4 5 6
2 3 4 5 6 7 7 7 7 7
78
1 1 1 1 1 2 2
2 3 3 3 4 4 4
79
80
81
82
83
84
85
86
weight=9
structure: 2,3,0,1
Generating graph:
1 1 1 1 1 2 2 3 3
2 3 4 4 5 5 5 6 6
87
weight=10
structure: 1,1,0,3
Generating graph:
1 1 1 1 1 2 2 2 2 3
2 3 3 3 3 4 4 4 5 5
88
89
91
structure: 2,2,0,2
Generating graph:
1 1 1 1 1 2 2 2 2 3
2 3 3 4 4 5 5 5 6 6
93
1 1 1 1 1 2 2 2 2 3
2 3 3 4 4 5 5 6 6 6
94
structure: 3,3,0,1
Generating graph:
1 1 1 1 1 2 2 3 3 4
2 3 4 4 5 5 6 6 7 7
96
97
99
100
weight=10
structure: 1,0,2,2
Generating graph:
1 1 1 1 1 2 2 2 2 3
2 3 3 3 4 4 4 4 5 5
101
weight=10
structure: 1,0,2,2
Generating graph:
1 1 1 1 1 2 2 2 2 3
2 3 3 3 4 4 5 5 5 5
102
weight=10
structure: 3,0,1,2
Generating graph:
1 1 1 1 1 2 2 2 2 3
2 3 3 3 4 4 5 5 6 6
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
weight=10
structure: 0,2,1,2
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 4 5 5 5 5 5
118
119
weight=8
structure: 2,1,1,1
Generating graph:
1 1 1 1 1 2 2 2
2 3 3 3 4 4 5 5
121
122
123
weight=9
structure: 1,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3
2 3 3 3 4 4 4 5 5
124
weight=9
structure: 1,1,2,1
Generating graph:
125
1 1 1 1 1 2 2 2 3
2 3 3 3 4 4 5 5 5
I91 = (20 21 40 13 05 + 20 21 40 04 14 + 20 21 31 22 05
+220 21 31 13 14 320 21 31 04 23 320 21 222 14
+320 21 22 13 23 + 220 21 22 04 32 220 21 213 32
+220 12 40 13 14 220 12 40 04 23 220 12 31 22 14
420 12 31 13 23 + 620 12 31 04 32 + 620 12 222 23
620 12 22 13 32 420 12 22 04 41 + 420 12 213 41
20 03 40 13 23 + 20 03 40 04 32 + 20 03 31 22 23
+220 03 31 13 32 320 03 31 04 41 320 03 222 32
+320 03 22 13 41 + 220 03 22 04 50 220 03 213 50
+11 30 40 13 05 11 30 40 04 14 11 30 31 22 05
211 30 31 13 14 + 311 30 31 04 23 + 311 30 222 14
311 30 22 13 23 211 30 22 04 32 + 211 30 213 32
+211 21 40 22 05 511 21 40 13 14 + 311 21 40 04 23
211 21 231 05 + 511 21 31 22 14 + 611 21 31 13 23
711 21 31 04 32 911 21 222 23 + 711 21 22 13 32
+411 21 22 04 41 411 21 213 41 411 12 40 22 14
+711 12 40 13 23 311 12 40 04 32 + 411 12 231 14
711 12 31 22 23 611 12 31 13 32 + 511 12 31 04 41
+911 12 222 32 511 12 22 13 41 211 12 22 04 50
+211 12 213 50 + 211 03 40 22 23 311 03 40 13 32
+11 03 40 04 41 211 03 231 23 + 311 03 31 22 32
+211 03 31 13 41 11 03 31 04 50 311 03 222 41
+11 03 22 13 50 202 30 40 22 05 + 302 30 40 13 14
02 30 40 04 23 + 202 30 231 05 302 30 31 22 14
202 30 31 13 23 + 02 30 31 04 32 + 302 30 222 23
02 30 22 13 32 + 402 21 40 22 14 602 21 40 13 23
+202 21 40 04 32 402 21 231 14 + 602 21 31 22 23
+402 21 31 13 32 202 21 31 04 41 602 21 222 32
+202 21 22 13 41 202 12 40 22 23 + 302 12 40 13 32
02 12 40 04 41 + 202 12 231 23 302 12 31 22 32
202 12 31 13 41 + 02 12 31 04 50 + 302 12 222 41
02 12 22 13 50 )/14
00
weight=9
structure: 1,1,2,1
Generating graph:
126
1 1 1 1 1 2 2 2 3
2 3 3 4 4 4 4 5 5
127
I93 = (20 21 40 13 05 20 21 40 04 14 20 21 31 22 05
220 21 31 13 14 + 320 21 31 04 23 + 320 21 222 14
320 21 22 13 23 220 21 22 04 32 + 220 21 213 32
220 12 40 22 05 + 220 12 40 13 14 + 220 12 231 05
220 12 31 22 14 220 12 31 04 32 + 220 12 22 13 32
+220 12 22 04 41 220 12 213 41 + 220 03 40 22 14
320 03 40 13 23 + 20 03 40 04 32 220 03 231 14
+320 03 31 22 23 + 220 03 31 13 32 20 03 31 04 41
320 03 222 32 + 20 03 22 13 41 11 30 40 13 05
+11 30 40 04 14 + 11 30 31 22 05 + 211 30 31 13 14
311 30 31 04 23 311 30 222 14 + 311 30 22 13 23
+211 30 22 04 32 211 30 213 32 + 211 21 40 22 05
311 21 40 13 14 + 11 21 40 04 23 211 21 231 05
+311 21 31 22 14 + 211 21 31 13 23 11 21 31 04 32
311 21 222 23 + 11 21 22 13 32 + 11 12 40 13 23
11 12 40 04 32 11 12 31 22 23 211 12 31 13 32
+311 12 31 04 41 + 311 12 222 32 311 12 22 13 41
211 12 22 04 50 + 211 12 213 50 211 03 40 22 23
+311 03 40 13 32 11 03 40 04 41 + 211 03 231 23
311 03 31 22 32 211 03 31 13 41 + 11 03 31 04 50
+311 03 222 41 11 03 22 13 50 + 02 30 40 13 14
02 30 40 04 23 02 30 31 22 14 202 30 31 13 23
+302 30 31 04 32 + 302 30 222 23 302 30 22 13 32
202 30 22 04 41 + 202 30 213 41 202 21 40 22 14
+202 21 40 13 23 + 202 21 231 14 202 21 31 22 23
202 21 31 04 41 + 202 21 22 13 41 + 202 21 22 04 50
202 21 213 50 + 202 12 40 22 23 302 12 40 13 32
+02 12 40 04 41 202 12 231 23 + 302 12 31 22 32
+202 12 31 13 41 02 12 31 04 50 302 12 222 41
+02 12 22 13 50 )/14
00
weight=9
structure: 1,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3
2 3 4 4 4 4 5 5 5
129
1 1 1 1 1 2 2 2 3
2 3 4 4 5 5 6 6 6
131
132
133
weight=10
structure: 2,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 3 4 4 4 5 5 6 6
135
+211 02 30 40 04 23 + 211 02 30 31 13 23
211 02 30 31 04 32 + 211 02 30 22 13 32
211 02 30 22 04 41 211 02 30 213 41
+211 02 30 13 04 50 + 611 02 21 40 22 14
611 02 21 40 13 23 611 02 21 31 22 23
+611 02 21 31 13 32 611 02 21 222 32
+1211 02 21 22 13 41 611 02 21 213 50
611 02 12 40 31 14 + 611 02 12 40 22 23
+611 02 12 231 23 611 02 12 31 13 41 611 02 12 222 41
+611 02 12 22 13 50 + 211 02 03 240 14
411 02 03 40 31 23 211 02 03 40 22 32
+211 02 03 40 13 41 + 211 02 03 231 32
+211 02 03 31 22 41 211 02 03 31 13 50
+202 30 40 13 23 202 30 40 04 32 2202 30 31 13 32
+2202 30 31 04 41 + 202 30 22 13 41 202 30 22 04 50
3202 21 40 22 23 + 3202 21 40 13 32 + 6202 21 31 22 32
6202 21 31 13 41 3202 21 222 41 + 3202 21 22 13 50
+3202 12 40 31 23 3202 12 40 22 32 6202 12 231 32
+9202 12 31 22 41 3202 12 222 50 202 03 240 23
+3202 03 40 31 32 202 03 40 22 41 2202 03 231 41
+202 03 31 22 50 )/16
00
weight=10
structure: 2,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 3 4 4 5 5 5 6 6
137
139
weight=10
structure: 2,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 4 4 5 5 6 6
141
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 4 5 5 5 6 6
143
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 4 5 5 6 6 6
145
402 21 212 13 50 + 02 21 12 03 40 32
+202 21 12 03 31 41 302 21 12 03 22 50
202 21 203 40 41 + 202 21 203 31 50 202 312 40 32
202 312 31 41 + 402 312 22 50 + 202 212 03 40 41
202 212 03 31 50 )/16
00
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 5 5 5 6 6 6
147
weight=10
structure: 4,1,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 3
2 3 4 4 5 5 6 6 7 7
149
weight=10
structure: 2,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 3 3 4 4 5 5 6 6
151
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 3 4 4 5 5 5 6 6
153
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 3 4 4 5 5 6 6 6
155
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 3 4 5 5 5 6 6 6
157
weight=10
structure: 2,1,2,1
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 4 5 5 5 5 6 6 6
159
weight=10
structure: 4,1,1,1
Generating graph:
1 1 1 1 1 2 2 2 3 4
2 3 4 5 5 6 6 7 7 7
161
+602 30 21 03 22 32 + 202 30 21 03 13 41
+302 30 212 31 23 + 902 30 212 22 32 + 302 30 212 13 41
402 30 12 03 31 32 402 30 12 03 22 41
+02 30 203 31 41 202 321 31 14 702 321 22 23
702 321 13 32 202 321 04 41 + 02 221 12 40 14
+1102 221 12 31 23 + 2102 221 12 22 32 + 1102 221 12 13 41
+02 221 12 04 50 02 221 03 40 23 502 221 03 31 32
502 221 03 22 41 02 221 03 13 50 302 21 212 40 23
1502 21 212 31 32 1502 21 212 22 41 302 21 212 13 50
+402 21 12 03 40 32 + 1002 21 12 03 31 41
+402 21 12 03 22 50 02 21 203 40 41 02 21 203 31 50
+202 312 40 32 + 502 312 31 41 + 202 312 22 50
302 212 03 40 41 302 212 03 31 50 + 02 12 203 40 50 )/16
00
weight=10
structure: 1,3,1,1
Generating graph:
1 1 1 1 1 2 2 3 3 4
2 3 4 4 5 5 6 6 6 6
163
structure: 1,0,0,0,3
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 2 3 3 3 3 3 3 4 4
1 1 1 1 1 1
2 2 2 3 3 3
1 1 1 1 1 1 2 2 3
2 3 3 4 4 4 5 5 5
weight=8
structure: 0,0,1,0,2
Generating graph:
1 1 1 1 1 1 2 2
2 2 2 2 3 3 3 3
166
167
168
169
170
171
172
weight=10
structure: 1,2,0,0,2
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 2 3 3 3 4 4 4 5 5
173
174
175
176
177
178
179
180
181
182
weight=9
structure: 4,0,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 2
2 3 3 4 4 5 5 6 6
184
185
186
weight=10
structure: 3,0,2,0,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 3 4 4 5 5 6 6
187
weight=8
structure: 0,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2
2 2 3 3 3 4 4 4
189
190
191
192
193
weight=10
structure: 2,0,0,2,1
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 3 3 3 4 4 5 5
195
structure: 0,1,0,1,1
Generating graph:
1 1 1 1 1 1 2
2 2 2 2 3 3 3
197
198
199
200
201
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 3 3 4 4 4 4 4
203
204
205
206
weight=9
structure: 1,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 3
2 2 3 3 4 4 5 5 5
207
208
+3611 02 21 12 31 33 2711 02 21 12 22 42
+411 02 21 03 40 33 1011 02 21 03 31 42
+611 02 21 03 22 51 + 611 02 212 40 33 1511 02 212 31 42
+911 02 212 22 51 511 02 12 03 40 42
+811 02 12 03 31 51 311 02 12 03 22 60
+11 02 203 40 51 11 02 203 31 60 + 202 230 40 06
202 230 31 15 5202 30 21 40 15 + 5202 30 21 31 24
+4202 30 12 40 24 4202 30 12 31 33 202 30 03 40 33
+202 30 03 31 42 + 6202 221 40 24 6202 221 31 33
9202 21 12 40 33 + 9202 21 12 31 42 + 2202 21 03 40 42
2202 21 03 31 51 + 3202 212 40 42 3202 212 31 51
202 12 03 40 51 + 202 12 03 31 60 )/16
00
weight=10
structure: 2,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 4 4 4 5 5 6 6
210
weight=10
structure: 2,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 4 4 5 5 5 6 6
212
weight=10
structure: 2,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 4 4 5 5 6 6 6
214
weight=10
structure: 2,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 4 4 5 5 5 6 6 6
216
weight=10
structure: 2,2,1,0,1
Generating graph:
217
1 1 1 1 1 1 2 2 2 5
2 3 3 4 4 5 5 6 6 6
218
weight=10
structure: 2,2,1,0,1
Generating graph:
1 1 1 1 1 1 2 2 3 3
2 2 3 4 4 4 5 5 6 6
220
Generating graph:
1 1 1 1 1 1 2 2
2 2 2 3 3 4 4 4
222
223
224
225
226
227
228
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 4 4 5 5 5 6 6
weight=10
230
structure: 3,1,0,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 4 4 5 5 6 6 6
231
weight=10
structure: 3,1,0,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 2 3 3 4 4 5 5 6 6
233
234
235
236
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 3 3 4 4 4 5 5
238
weight=10
structure: 1,1,1,1,1
Generating graph:
239
1 1 1 1 1 1 2 2 2 2
2 3 3 3 3 4 4 5 5 5
240
+302 03 22 32 51 + 02 03 22 23 60 + 202 03 13 50 42
202 03 13 32 60 02 03 04 50 51 + 02 03 04 41 60 )/15
00
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 3 4 4 4 4 5 5
242
weight=10
structure: 1,1,1,1,1
Generating graph:
243
1 1 1 1 1 1 2 2 2 2
2 3 3 3 4 4 5 5 5 5
244
I198 = (20 30 40 14 06 20 30 40 05 15 20 30 31 23 06
220 30 31 14 15 + 320 30 31 05 24 + 320 30 22 23 15
320 30 22 05 33 320 30 13 23 24 + 220 30 13 14 33
+20 30 13 05 42 + 20 30 04 23 33 20 30 04 14 42
320 21 40 23 06 + 320 21 40 14 15 + 320 21 31 32 06
+620 21 31 23 15 920 21 31 14 24 920 21 22 32 15
+920 21 22 14 33 + 920 21 13 32 24 620 21 13 23 33
320 21 13 14 42 320 21 04 32 33 + 320 21 04 23 42
+320 12 40 32 06 320 12 40 23 15 320 12 31 41 06
620 12 31 32 15 + 920 12 31 23 24 + 920 12 22 41 15
920 12 22 23 33 920 12 13 41 24 + 620 12 13 32 33
+320 12 13 23 42 + 320 12 04 41 33 320 12 04 32 42
20 03 40 41 06 + 20 03 40 32 15 + 20 03 31 50 06
+220 03 31 41 15 320 03 31 32 24 320 03 22 50 15
+320 03 22 32 33 + 320 03 13 50 24 220 03 13 41 33
20 03 13 32 42 20 03 04 50 33 + 20 03 04 41 42
211 30 40 14 15 + 211 30 40 05 24 + 211 30 31 23 15
+411 30 31 14 24 611 30 31 05 33 611 30 22 23 24
+611 30 22 05 42 + 611 30 13 23 33 411 30 13 14 42
211 30 13 05 51 211 30 04 23 42 + 211 30 04 14 51
+611 21 40 23 15 611 21 40 14 24 611 21 31 32 15
1211 21 31 23 24 + 1811 21 31 14 33 + 1811 21 22 32 24
1811 21 22 14 42 1811 21 13 32 33 + 1211 21 13 23 42
+611 21 13 14 51 + 611 21 04 32 42 611 21 04 23 51
611 12 40 32 15 + 611 12 40 23 24 + 611 12 31 41 15
+1211 12 31 32 24 1811 12 31 23 33 1811 12 22 41 24
+1811 12 22 23 42 + 1811 12 13 41 33 1211 12 13 32 42
611 12 13 23 51 611 12 04 41 42 + 611 12 04 32 51
+211 03 40 41 15 211 03 40 32 24 211 03 31 50 15
411 03 31 41 24 + 611 03 31 32 33 + 611 03 22 50 24
611 03 22 32 42 611 03 13 50 33 + 411 03 13 41 42
+211 03 13 32 51 + 211 03 04 50 42 211 03 04 41 51
+02 30 40 14 24 02 30 40 05 33 02 30 31 23 24
202 30 31 14 33 + 302 30 31 05 42 + 302 30 22 23 33
302 30 22 05 51 302 30 13 23 42 + 202 30 13 14 51
+02 30 13 05 60 + 02 30 04 23 51 02 30 04 14 60
302 21 40 23 24 + 302 21 40 14 33 + 302 21 31 32 24
+602 21 31 23 33 902 21 31 14 42 902 21 22 32 33
+902 21 22 14 51 + 902 21 13 32 42 602 21 13 23 51
302 21 13 14 60 302 21 04 32 51 + 302 21 04 23 60
+302 12 40 32 24 302 12 40 23 33 302 12 31 41 24
602 12 31 32 33 + 902 12 31 23 42 + 902 12 22 41 33
902 12 22 23 51 902 12 13 41 42 + 602 12 13 32 51
+302 12 13 23 60 + 302 12 04 41 51 302 12 04 32 60
02 03 40 41 24 + 02 03 40 32 33 + 02 03 31 50 24
+202 03 31 41 33 302 03 31 32 42 302 03 22 50 33
245
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 2
2 3 3 4 4 4 4 5 5 5
246
202 21 40 41 06 + 302 21 40 32 15 02 21 40 14 33
+602 21 31 41 15 902 21 31 32 24 + 302 21 31 14 42
602 21 22 41 24 + 902 21 22 32 33 302 21 22 14 51
+202 21 13 41 33 302 21 13 32 42 + 02 21 13 14 60
+02 12 40 50 06 302 12 40 32 24 + 202 12 40 23 33
302 12 31 50 15 + 902 12 31 32 33 602 12 31 23 42
+302 12 22 50 24 902 12 22 32 42 + 602 12 22 23 51
02 12 13 50 33 + 302 12 13 32 51 202 12 13 23 60
02 03 40 50 15 + 202 03 40 41 24 02 03 40 32 33
+302 03 31 50 24 602 03 31 41 33 + 302 03 31 32 42
302 03 22 50 33 + 602 03 22 41 42 302 03 22 32 51
+02 03 13 50 42 202 03 13 41 51 + 02 03 13 32 60 )/15
00
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 2 3 3 3 4 4 4 5 5
248
11 03 04 50 42 + 211 03 04 41 51 11 03 04 32 60
+02 30 40 32 06 202 30 40 23 15 + 02 30 40 14 24
202 30 31 41 06 + 202 30 31 32 15 + 202 30 31 23 24
202 30 31 14 33 + 02 30 22 50 06 + 202 30 22 41 15
602 30 22 32 24 + 202 30 22 23 33 + 02 30 22 14 42
202 30 13 50 15 + 202 30 13 41 24 + 202 30 13 32 33
202 30 13 23 42 + 02 30 04 50 24 202 30 04 41 33
+02 30 04 32 42 202 21 40 32 15 + 402 21 40 23 24
202 21 40 14 33 + 402 21 31 41 15 402 21 31 32 24
402 21 31 23 33 + 402 21 31 14 42 202 21 22 50 15
402 21 22 41 24 + 1202 21 22 32 33 402 21 22 23 42
202 21 22 14 51 + 402 21 13 50 24 402 21 13 41 33
402 21 13 32 42 + 402 21 13 23 51 202 21 04 50 33
+402 21 04 41 42 202 21 04 32 51 + 02 12 40 32 24
202 12 40 23 33 + 02 12 40 14 42 202 12 31 41 24
+202 12 31 32 33 + 202 12 31 23 42 202 12 31 14 51
+02 12 22 50 24 + 202 12 22 41 33 602 12 22 32 42
+202 12 22 23 51 + 02 12 22 14 60 202 12 13 50 33
+202 12 13 41 42 + 202 12 13 32 51 202 12 13 23 60
+02 12 04 50 42 202 12 04 41 51 + 02 12 04 32 60 )/15
00
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 2 3 3 4 4 4 4 5 5
250
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 2 3 3 4 4 5 5 5 5
252
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 2 3 4 4 4 5 5 5 5
254
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 3 3 4 4 4 5 5
256
weight=10
structure: 1,1,1,1,1
Generating graph:
1 1 1 1 1 1 2 2 2 3
2 3 3 3 3 4 4 5 5 5
258
259
260
261
structure: 1,3,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 2 2
2 3 3 3 4 4 5 5 5
263
264
265
266
267
270
271
272
273
Generating graph:
1 1 1 1 1 1 1 2 2
2 2 2 3 3 4 4 5 5
275
276
277
279
280
Generating graph:
1 1 1 1 1 1 1 2
2 2 2 3 3 3 4 4
282
283
284
285
1 1 1 1 1 1 1 2 2 2
2 3 3 3 3 4 4 4 5 5
weight=10
structure: 1,1,2,0,0,1
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 3 3 3 3 4 4 5 5 5
288
I247 = (20 30 40 13 07 20 30 40 04 16 20 30 31 22 07
220 30 31 13 16 + 320 30 31 04 25 + 320 30 222 16
320 30 22 13 25 220 30 22 04 34 + 220 30 213 34
320 21 40 13 16 + 320 21 40 04 25 + 320 21 31 22 16
+620 21 31 13 25 920 21 31 04 34 920 21 222 25
+920 21 22 13 34 + 620 21 22 04 43 620 21 213 43
+320 12 40 13 25 320 12 40 04 34 320 12 31 22 25
620 12 31 13 34 + 920 12 31 04 43 + 920 12 222 34
920 12 22 13 43 620 12 22 04 52 + 620 12 213 52
20 03 40 13 34 + 20 03 40 04 43 + 20 03 31 22 34
+220 03 31 13 43 320 03 31 04 52 320 03 222 43
+320 03 22 13 52 + 220 03 22 04 61 220 03 213 61
211 30 40 22 07 + 211 30 40 13 16 + 211 30 231 07
211 30 31 22 16 211 30 31 04 34 + 211 30 22 13 34
+211 30 22 04 43 211 30 213 43 + 611 21 40 22 16
611 21 40 13 25 611 21 231 16 + 611 21 31 22 25
+611 21 31 04 43 611 21 22 13 43 611 21 22 04 52
+611 21 213 52 611 12 40 22 25 + 611 12 40 13 34
+611 12 231 25 611 12 31 22 34 611 12 31 04 52
+611 12 22 13 52 + 611 12 22 04 61 611 12 213 61
+211 03 40 22 34 211 03 40 13 43 211 03 231 34
+211 03 31 22 43 + 211 03 31 04 61 211 03 22 13 61
211 03 22 04 70 + 211 03 213 70 + 202 30 40 22 16
302 30 40 13 25 + 02 30 40 04 34 202 30 231 16
+302 30 31 22 25 + 202 30 31 13 34 02 30 31 04 43
302 30 222 34 + 02 30 22 13 43 602 21 40 22 25
+902 21 40 13 34 302 21 40 04 43 + 602 21 231 25
902 21 31 22 34 602 21 31 13 43 + 302 21 31 04 52
+902 21 222 43 302 21 22 13 52 + 602 12 40 22 34
902 12 40 13 43 + 302 12 40 04 52 602 12 231 34
+902 12 31 22 43 + 602 12 31 13 52 302 12 31 04 61
902 12 222 52 + 302 12 22 13 61 202 03 40 22 43
+302 03 40 13 52 02 03 40 04 61 + 202 03 231 43
302 03 31 22 52 202 03 31 13 61 + 02 03 31 04 70
+302 03 222 61 02 03 22 13 70 )/15
00
weight=10
structure: 1,1,2,0,0,1
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 3 3 3 4 4 4 4 5 5
289
1 1 1 1 1 1 1 2 2 2
2 3 3 3 4 4 5 5 6 6
291
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 3 3 4 4 5 5 5 6 6
weight=10
294
structure: 1,1,2,0,0,1
Generating graph:
1 1 1 1 1 1 1 2 2 3
2 2 3 3 3 4 4 4 5 5
295
weight=10
structure: 1,1,2,0,0,1
Generating graph:
296
1 1 1 1 1 1 1 2 2 3
2 2 3 4 4 4 4 5 5 5
297
structure: 3,1,1,0,0,1
Generating graph:
1 1 1 1 1 1 1 2 2 3
2 2 3 4 4 5 5 6 6 6
299
300
301
302
1 1 1 1 1 1 1 2 2 2
2 2 3 3 3 4 4 5 5 5
303
305
306
1 1 1 1 1 1 1 2 2 3
2 2 2 3 4 4 4 5 5 5
308
weight=9
structure: 1,0,1,1,0,1
Generating graph:
1 1 1 1 1 1 1 2 2
2 2 3 3 3 3 3 4 4
310
311
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 2 3 3 4 4 4 4 5 5
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 2 3 3 4 4 5 5 5 5
314
315
structure: 1,1,0,0,1,1
Generating graph:
1 1 1 1 1 1 1 2 2
2 3 3 3 3 3 3 4 4
1 1 1 1 1 1 1 2 2 2
2 2 2 3 3 3 4 4 5 5
structure: 2,1,0,0,1,1
Generating graph:
1 1 1 1 1 1 1 2 2 2
2 2 2 3 3 4 4 4 5 5
1 1 1 1 1 1 1 2 2 2
2 2 2 3 3 4 4 5 5 5
weight=10
structure: 2,1,0,0,1,1
Generating graph:
1 1 1 1 1 1 1 2 2 3
2 2 2 2 3 4 4 5 5 5
321
322
323
324
325
326
327
328
329
330
331
332
333
structure: 2,0,1,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1
2 2 2 2 3 3 4 4
335
336
337
338
structure: 0,2,1,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1 2
2 2 2 3 3 3 4 4 4
340
341
342
343
1 1 1 1 1 1 1 1 2 2
2 3 3 3 3 3 3 4 4 4
345
1 1 1 1 1 1 1 1 2
2 2 2 2 2 2 3 3 3
346
347
348
349
1 1 1 1 1 1 1 1 2 2
2 3 3 3 3 4 4 4 5 5
weight=10
structure: 1,2,1,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1 2 2
2 3 3 3 3 4 4 5 5 5
350
351
352
353
354
355
356
357
358
359
360
361
362
weight=10
structure: 1,1,0,0,0,1,1
Generating graph:
1 1 1 1 1 1 1 1 2 2
2 3 3 3 3 3 3 3 4 4
365
366
367
1 1 1 1 1 1 1 1 1
2 2 2 2 2 3 3 4 4
369
374
375
376
377
378
379
380
weight=10
structure: 5,0,0,0,0,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1 1 1
2 2 3 3 4 4 5 5 6 6
381
weight=10
structure: 2,2,0,0,0,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1 1 1
2 2 2 3 3 3 4 4 5 5
weight=10
structure: 1,0,2,0,0,0,0,0,1
Generating graph:
1 1 1 1 1 1 1 1 1 1
2 2 2 2 3 3 3 3 4 4
383
386
387
Dependencies
weight=18, dependency no.1
structure: 6,8
4I13 I22 + 12I12 I2 I32 12I1 I34 I2 I42 + 4I33 I4 I52 = 0
weight=18, dependency no.2
structure: 6,0,6
16I13 I72 8I12 I6 I7 I8 I1 I62 I82 + 4I1 I6 I92 + 12I1 I7 I8 I9 + I6 I82 I9
2
I7 I83 4I93 I10
=0
2
I1 I2 I6 4I1 I13 3I2 I9 + 3I32 I6 + 6I3 I20 3I8 I11 + 3I16
=0
2
2I1 I3 I19 + 4I1 I3 I20 I1 I15
2I1 I15 I16 + I3 I8 I15 + I4 I19 2I4 I20
2
+I15 I22 I17 + I17 I18 = 0
2
2
4I1 I2 I9 + 4I1 I16
+ I2 I82 + 4I32 I9 4I3 I8 I16 + I18
=0
388
2I1 I32 I6 8I1 I3 I19 + 2I1 I3 I20 6I1 I15 I16 + 4I32 I9 I3 I4 I6
+3I3 I8 I15 4I3 I31 3I4 I20 + 3I16 I22 + 3I17 I18 = 0
4I1 I3 I12 2I3 I6 I17 + 2I3 I6 I18 + 2I4 I12 3I15 I23 I15 I24 + 4I17 I19
2I17 I20 4I18 I19 + 2I18 I20 = 0
2I3 I6 I17 + I15 I23 + I15 I24 + 2I15 I27 2I17 I20 2I17 I21 + 2I18 I19 4I18 I20
=0
weight=15, dependency no.12
structure: 5,4,2
2I1 I3 I24 + I1 I15 I17 I4 I24 I15 I28 I17 I22 + I18 I22 = 0
weight=15, dependency no.13
structure: 4,6,1
2I1 I2 I17 2I1 I2 I18 4I1 I3 I25 + 2I1 I3 I26 2I32 I17 + 2I32 I18 + 2I4 I25
I4 I26 I5 I15 = 0
4I1 I2 I17 + 2I1 I2 I18 2I1 I3 I25 2I1 I3 I26 2I2 I28 + 2I32 I17 2I32 I18
+I4 I25 I4 I26 3I5 I16 = 0
2
2I3 I6 I19 + 4I3 I6 I20 8I3 I7 I15 I6 I15
+ 2I6 I15 I16 + 2I12 I17 I15 I29
2
2
+I15 I30 + 4I19 10I19 I20 + 4I20 = 0
5I1 I3 I6 I15 4I1 I15 I19 + 7I1 I15 I20 4I3 I9 I15 + I4 I6 I15 + I15 I31
+3I17 I24 + 3I19 I22 6I20 I22 = 0
I1 I3 I6 I15 + 2I1 I3 I30 + I1 I15 I20 2I3 I9 I15 I4 I30 I15 I31
I17 I24 + I18 I24 = 0
389
4I12 I2 I16 + 2I1 I2 I22 + 8I1 I32 I16 + I2 I4 I8 2I33 I8 2I32 I22 2I3 I4 I16
I5 I18 = 0
2
4I1 I3 I13 + I2 I8 I15 I3 I15
2I3 I15 I16 2I4 I13 + I15 I32 I17 I26
+I18 I26 = 0
2
2I1 I3 I30 + I3 I6 I22 I4 I30 I8 I15
2I17 I24 + 2I18 I24 I19 I22 + 3I20 I22
+I21 I22 = 0
2I1 I3 I70 + 2I1 I15 I69 I4 I70 + I15 I85 3I15 I86 + 2I15 I87 3I15 I88 I17 I84
+I18 I84 = 0
weight=14, dependency no.26
structure: 5,3,1,1
390
2I1 I3 I87 + I1 I15 I43 I4 I87 I15 I47 I17 I44 + I18 I44 = 0
weight=14, dependency no.27
structure: 4,2,1,2
8I1 I3 I60 + 4I1 I42 I82 + 2I1 I42 I83 + 2I1 I43 I69 4I3 I8 I34 4I3 I65
2I4 I59 4I15 I35 + 12I16 I35 2I42 I92 + 2I42 I94 + 4I43 I85 2I43 I86 + 2I43 I87 4I43 I88
3I44 I82 I44 I83 + I44 I84 2I47 I69 = 0
weight=14, dependency no.28
structure: 2,2,2,2
8I3 I66 4I3 I67 + 2I15 I59 6I16 I59 2I42 I89 + 2I42 I90 + 2I42 I91 4I42 I93 4I43 I70
2
+6I43 I71 + 4I69 I86 4I69 I87 + 2I69 I88 3I82 I83 + 3I83
+ I83 I84 = 0
weight=14, dependency no.29
structure: 3,3,2,1
2I3 I6 I43 4I3 I8 I69 + 4I3 I97 + 2I3 I99 4I3 I100 + I15 I86 + I16 I86 4I16 I87
+2I16 I88 + I18 I83 2I20 I43 + I21 I43 + I24 I42 I27 I42 = 0
weight=14, dependency no.30
structure: 3,3,2,1
I8 I102 4I9 I45 + 2I9 I46 + 2I16 I85 I18 I82 = 0
weight=14, dependency no.31
structure: 2,5,1,1
I2 I85 I3 I102 2I16 I45 + I16 I46 I18 I37 = 0
weight=14, dependency no.32
structure: 2,5,1,1
4I2 I86 4I2 I87 + 2I2 I88 + 12I3 I101 6I3 I103 12I3 I106 6I11 I43 I15 I46 + 3I16 I46
+3I25 I42 2I26 I42 = 0
weight=14, dependency no.33
structure: 3,3,2,1
16I1 I3 I70 + 12I1 I3 I71 4I1 I15 I69 + 6I3 I6 I43 12I3 I109 + 2I4 I70 2I15 I85
+2I15 I87 + 6I16 I86 6I16 I87 + 12I16 I88 + 6I17 I83 + 2I17 I84 2I18 I84 12I19 I43 + 6I20 I43
+3I23 I42 3I24 I42 = 0
391
4I1 I3 I70 + 4I1 I3 I71 + 4I1 I16 I69 2I3 I8 I69 + 8I3 I97 2I3 I99 + 4I3 I100
8I3 I105 + 2I4 I70 2I4 I71 2I15 I85 + 3I15 I86 + 4I16 I85 3I16 I86 4I16 I87 + 6I16 I88
+2I17 I82 + I18 I82 I18 I84 4I19 I43 + I21 I43 2I22 I69 + I23 I42 I27 I42 = 0
weight=14, dependency no.36
structure: 4,5,0,1
6I1 I3 I46 + 2I2 I47 6I32 I43 6I3 I58 I4 I46 + I5 I42 = 0
weight=14, dependency no.37
structure: 6,2,0,2
2
2
2I12 I3 I34 + 2I12 I42
2I1 I3 I35 3I1 I42 I44 + I1 I43
I3 I36 I4 I35
2
+I43 I47 + I44 = 0
I15 I89 I15 I90 I17 I70 I19 I84 + 2I20 I84 = 0
weight=15, dependency no.41
structure: 4,3,2,1
2I1 I3 I89 + 2I1 I3 I90 I1 I15 I84 2I1 I16 I84 2I1 I17 I69 + I3 I8 I84
+I4 I89 I4 I90 I17 I85 + 3I17 I86 2I17 I87 + 3I17 I88 + I22 I84 = 0
392
I1 I15 I83 I15 I94 I17 I87 I19 I44 + 2I20 I44 = 0
weight=15, dependency no.43
structure: 6,3,1,1
2I12 I3 I83 + 2I1 I3 I94 + I1 I4 I83 I1 I15 I44 2I1 I16 I44 I1 I17 I43
+I3 I8 I44 I4 I94 + I17 I47 + I22 I44 = 0
2I3 I6 I84 2I6 I15 I42 2I15 I89 + 3I15 I91 I15 I93 6I15 I95 + 2I18 I70 + 2I20 I84
+2I21 I84 = 0
weight=15, dependency no.45
structure: 4,3,2,1
4I1 I3 I6 I42 + 4I1 I3 I89 6I1 I3 I91 + 2I1 I3 I93 + 12I1 I3 I95 + 4I1 I16 I84
+4I1 I18 I69 2I3 I8 I84 2I4 I6 I42 2I4 I89 + 3I4 I91 I4 I93 6I4 I95
+2I18 I85 6I18 I86 + 4I18 I87 6I18 I88 2I22 I84 = 0
2I1 I3 I6 I42 2I1 I3 I89 + 2I1 I3 I93 2I1 I16 I84 2I1 I18 I69 + 6I3 I6 I44
3I3 I8 I83 + I3 I8 I84 + I4 I6 I42 + I4 I89 I4 I93 + 3I15 I92 3I15 I96
6I16 I92 6I16 I94 + 3I16 I96 3I17 I86 I18 I85 + I18 I87 3I19 I44 + 6I20 I44 + 3I22 I83
+I22 I84 = 0
weight=15, dependency no.47
structure: 3,5,1,1
2I1 I3 I73 2I1 I3 I74 2I1 I15 I37 + I3 I15 I42 + I4 I73 + I4 I74 + I15 I50
+2I17 I45 I17 I46 2I18 I45 + I18 I46 = 0
I3 I6 I44 I8 I15 I42 + 2I15 I92 I15 I96 + I18 I87 + I20 I44 + I21 I44 = 0
weight=15, dependency no.49
structure: 3,5,1,1
2I1 I2 I82 4I1 I16 I37 I2 I8 I42 2I32 I82 + 2I3 I8 I37 + 2I3 I16 I42
+2I18 I45 I18 I46 = 0
393
4I1 I9 I37 + 2I1 I16 I82 I3 I8 I82 + 2I3 I9 I42 + I82 I37 I8 I16 I42
+I18 I85 = 0
weight=15, dependency no.52
structure: 4,3,2,1
4I1 I3 I6 I42 2I1 I3 I89 + 20I1 I3 I91 + 6I1 I3 I93 30I1 I15 I83 + 8I1 I16 I82
+24I1 I16 I83 + 6I1 I16 I84 + 4I1 I17 I69 + 6I1 I18 I69 + 8I1 I19 I42 + 16I1 I20 I42
+62I3 I6 I44 4I3 I8 I82 36I3 I8 I83 + 5I3 I8 I84 20I3 I9 I42 + 10I4 I6 I42
+I4 I89 2I4 I91 + 5I4 I93 + 13I15 I92 48I16 I92 48I16 I94 43I17 I86 + 30I17 I87
30I17 I88 I18 I85 28I18 I86 + 29I18 I87 19I18 I88 21I19 I44 + 24I20 I44 2I22 I82
+32I22 I83 I22 I84 + 12I23 I43 + 8I24 I43 4I28 I69 = 0
weight=15, dependency no.53
structure: 3,5,1,1
4I1 I2 I82 6I1 I2 I83 I1 I2 I84 + 6I1 I3 I72 14I1 I3 I73 + 2I1 I3 I74
8I1 I11 I42 2I1 I15 I37 + 10I1 I16 I37 2I2 I92 2I2 I94 + 7I2 I96 + 8I32 I82
3I32 I84 5I3 I8 I37 + 4I3 I16 I42 + I4 I72 I4 I73 I4 I74 I5 I69 + 3I11 I44
I15 I50 + I15 I51 6I16 I51 6I17 I45 + I17 I46 3I18 I45 I18 I46 + 2I22 I37 4I25 I43
+5I26 I43 = 0
weight=15, dependency no.54
structure: 2,4,1,2
2I18 I38 + 2I37 I85 I42 I102 2I45 I82 + I46 I82 = 0
weight=15, dependency no.55
structure: 3,2,2,2
I1 I42 I70 + I1 I42 I71 + 2I1 I69 I83 2I17 I59 I18 I59 I42 I97 + I42 I99
2I42 I100 + 2I42 I105 I42 I109 + 2I44 I70 3I44 I71 2I69 I92 2I69 I94 + I69 I96 + 2I83 I85
3I83 I86 + I83 I87 = 0
weight=15, dependency no.56
structure: 5,2,1,2
394
I1 I42 I86 2I1 I42 I87 + 4I1 I42 I88 + I1 I43 I83 4I17 I35 2I18 I35 I42 I104
I42 I110 + 2I43 I92 + 2I43 I94 I43 I96 + 2I44 I87 3I44 I88 + I47 I83 = 0
2I1 I3 I89 2I1 I3 I93 + 6I1 I15 I83 + 2I1 I16 I84 + 2I1 I18 I69 8I1 I19 I42
4I1 I20 I42 6I3 I6 I44 I3 I8 I84 + 4I3 I9 I42 2I4 I6 I42 I4 I89
+I4 I93 + 3I15 I92 + 3I17 I86 6I17 I87 + 6I17 I88 + I18 I85 I18 I87 + 3I18 I88 + 9I19 I44
6I22 I83 I22 I84 4I31 I42 = 0
weight=15, dependency no.58
structure: 3,5,1,1
I1 I2 I82 + I1 I2 I83 2I1 I16 I37 2I2 I92 2I2 I94 + I2 I96 I32 I82
3I32 I83 + I3 I8 I37 + I3 I15 I42 + 3I11 I44 + I17 I46 + I18 I45 I32 I42 = 0
2I1 I3 I192 + 2I1 I15 I159 I4 I192 + I15 I182 3I15 I183 + 2I15 I184 3I15 I185 I17 I181
+I18 I181 = 0
weight=15, dependency no.60
structure: 4,4,1,0,1
2I1 I3 I169 I4 I169 + I15 I127 I15 I128 I17 I125 + I18 I125 = 0
weight=14, dependency no.61
structure: 2,3,1,1,1
6I3 I196 I15 I179 + 3I16 I179 2I42 I168 + 2I42 I169 I42 I172 + 3I43 I149 3I83 I124
+2I86 I117 2I87 I117 + I88 I117 = 0
weight=14, dependency no.62
structure: 4,3,0,1,1
6I1 I3 I179 + 4I1 I42 I124 3I1 I43 I117 6I3 I186 + I4 I179 I42 I127 + I42 I128
+3I43 I125 3I44 I124 I47 I117 = 0
6I1 I3 I149 2I1 I15 I117 + 2I1 I16 I117 I3 I8 I117 + 4I3 I173 4I3 I174 + 4I3 I175
2I3 I178 + I4 I149 + I15 I125 3I16 I125 2I17 I124 I18 I124 + I22 I117 = 0
395
396