060310391
0560565
1
2016/9/26 13:00-14:45
@1 - 4
•
•
•
•
→
2
h0p://www.new-fukushima.jp/archives/4992.html
•
“On Growth and Form” D’Arcy Thompson
…
→
–
–
–
8…
•
→
9
10
y(t) = c
ncos 2nπt
T + d
nsin
2nπt
T
#
$ % &
' (
i= 0
∑
N= →
t y
11
∑
∞
=
+
+
=
1
0
( cos sin )
) 2
(
n
n
n
nx b nx
a a
x
f
∫
∫
=
=
π π
π
π
2 0
2 0
sin ) 1 (
cos ) 1 (
nxdx x f b
nxdx x
f a
n n
12
) ( ) 2 (
) 2 ( 1
) 0 ( ) 1 (
x f x f
x x x
f
= +
⎩⎨
⎧
≤ <
−
≤ <
= π
π π
π
0 2 4 6 8 10
-1.00.01.0
x
f(x)
0 2 4 6 8 10
-1.00.01.0
x
f1(x)
x x f1( ) 4sin
=π
0 2 4 6 8 10
-1.00.01.0
x
f3(x)
x x
f sin5
5 ) 4
3( = π
0 2 4 6 8 10
-1.00.01.0
x
f2(x)
x x
f sin3
3 ) 4
2(
= π
0 2 4 6 8 10
-1.00.01.0
x
f4(x)
x x
f sin7
7 ) 4
4( = π
0 2 4 6 8 10
-1.00.01.0
x
f1(x)+f2(x)
0 2 4 6 8 10
-1.00.01.0
x
f1(x)+f2(x)+f3(x)+f4(x)
0 2 4 6 8 10
-1.00.01.0
x
f1(x)+f2(x)+f3(x)
f (x) =4 πsin x +
4 3πsin 3x +
4 5πsin5x +
4
7πsin 7x + ... = 4
(2k−1)πsin(2k−1)π
k =1
∞
∑
●
●
y(t) = c1cos2πt T + d1sin
2πt T
y(t) = c1cos2πt T+ d1sin
2πt T +....+ c5cos10πt
T + d5sin 10πt
T
y(t) = c1cos2πt T+ d1sin
2πt T +....+ c10cos20πt
T + d10sin 20πt
T
y(t) = c1cos2πt T+ c1sin
2πt T +....+ c40cos80πt
T + d40sin 80πt
T
2 =7 2
( 2 ( 2
2 8
N = 1 N = 5 N = 10
N = 20 N = 40 N = 80
●
●
15
3 a1=1, b1=0, c116 =0
a
nb
nc
n
(
17
an
bn cn
77
PC1 PC2
PC1=1.2 PC2=0.8
18
PC1=1.2 PC2=0.8 PC2
PC1
19
Pictured by Dr. Satoshi Niikura
20
6×6
∑
=
⎟⎠
⎜ ⎞
⎝
⎛ +
= N
i n
n T
nt b T nt a t x
0 sin2 cos2 )
( π π
∑=⎜⎝⎛ + ⎟⎠⎞
= N
i n
n T
d nt T c nt t y
0 sin2 cos2 )
( π π
SHAPE
h0p://lbm.ab.a.u-tokyo.ac.jp/~iwata/shape/
21
RGB 256 2
22
∑
=
= Δ
p
i i
p
t
t
1
t
KT =
→
∑
∞=
⎟ ⎠
⎜ ⎞
⎝
⎛ +
=
1
sin 2
cos 2
)
(
n
n
n
T
t
d n
T
t
c n
t
y π π
2 )
2 cos
2
1(cos
1 2
2
∑
=
−
−Δ
Δ
=
K
p
p p
p p
n
T
t
n
T
t
n
t
y
n
c T π π
π
2 )
2 sin
2
1(sin
1 2
2
∑
=
−
−Δ
Δ
=
K
p
p p
p p
n
T
t
n
T
t
n
t
y
n
d T π π
π
Kuhl and Giardina (1982)
... 23
1
⎥ ⎦
⎢ ⎤
⎣
⎡
−
+ −
= +
21 2 1 2 1 2 1
1 1 1 1 1
)
(
arctan 2
2
1
d
b
c
a
d
c
b
θ a
⎥ ⎦
⎢ ⎤
⎣
⎥ ⎡
⎦
⎢ ⎤
⎣
⎡
= −
⎥ ⎦
⎢ ⎤
⎣
⎡
1 1
1 1
1 1
1 1
* 1
* 1
* 1
* 1
cos
sin
sin
cos
d b
c a d
b c a
θ
θ
θ
θ
* 1
* 1 1
arctan
a
= c
ψ
1
1
E*
2
* 1 2
*
* a
1c
E = +
⎥ ⎦
⎢ ⎤
⎣
⎡
⎥ −
⎦
⎢ ⎤
⎣
⎥ ⎡
⎦
⎢ ⎤
⎣
⎡
= −
⎥ ⎦
⎢ ⎤
⎣
⎡
1 1
1 1
1 1
1 1
*
*
*
*
*
*
*
*
cos
sin
sin
cos
cos
sin
sin
cos
*
1
θ
θ
θ
θ
ψ
ψ
ψ
ψ
n
n
n
n
d
b
c
a
d E
b
c
a
n n
n n
n n
n n
... 24
h0p://lbm.ab.a.u-tokyo.ac.jp/~iwata/shape/ 25
73.9%
14.2%
3.9%
1.9%
78.9%
10.3%
5.6%
3.8%
h0p://lbm.ab.a.u-tokyo.ac.jp/~iwata/shape/ 26
27
28
↓ ↓
0
2
2 2 0 ) 2 2 29
30
… T ... C ... T ... A ... G ... C ... T ...
… A ... G ... C ... A ... G ... T ... T ...
… T ... C ... C ... A ... A ... C ... G ...
… A ... C ... C ... A ... G ... C ... G ...
… T ... C ... T ... C ... A ... C ... G ...
… T ... C ... T ... C ... A ... T ... T ...
… A ... C ... C ... C ... G ... C ... T ...
… A ... G ... T ... C ... A ... C ... T ...
• DNA
8
•
DNA
31
−0.20 −0.15 −0.10 −0.05 0.00 0.05 0.10 0.15
−0.02−0.010.000.010.02
PC1
PC2
2
=7
( 2 2 32
1
DNA
DNA
33
x
y = f (x)
\\\x
y
273 142
304
423 173 373 234 138
203133 223
y = f (x)
GS
PLS
x 1
x 2
x 3
x 4
y 1
t 1
t 2
y 2
y 3
y 4
s 1
s 2
77
DNA
36,901
PLS
φ
1x y
1t1
t2
y
2y
3y
4s1
s2
φ
2x
φ
3x
φ
4x
x
PLS
input space)
x
feature space)
φ
xy
i= x
iTw + e
iy
i= φ (x
i)
Tw + e
i∞ 77
•
•
HP
(Yoshioka 2004)
(Goto 2005)
38
(Goto et al. 2005)
→
39
8.3 m tower
40
• 3 4 AP3 AP5
3 BP3
• 70% 1 2
AP1 AP2 AP3, AP5,
BP3 9.5%
•
Table 4. Estimation of parameters in regresssion models.
Variables Estimate SE t P(Prob>|t|)
AP3 -21.214 5.549 -3.820 0.0003
AP5 -41.990 14.138 -2.970 0.0040
BP3 -38.009 18.649 -2.040 0.0452
Area 0.013 0.002 5.250 <0.0001
Weight -16.181 5.998 -2.700 0.0087
41
42
43
CG
PC1 (42.09%) PC2 (37.81%)
CG
-2 s.d. +2 s.d.
PC1:
PC :
44
0
100
0
100
P = 0.31, n = 24
P = 0.57, n = 28
0
100
0
100
P < 0.02, n = 45
P = 0.33, n = 26 PC1: mean-2s.d. vs. mean+2s.d. PC2: mean-2s.d. vs. mean+2s.d.
Training Training
±2 95
PC1 PC2
Results of two-tailed binomial tests
45
0
100 P = 1, n = 28
P = 0.52, n = 22
0
100
P < 0.02, n = 38
P < 0.04, n = 25
0
100
0
100
PC1: mean-4s.d. vs. mean+4s.d. PC2: mean-4s.d. vs. mean+4s.d.
Training Training
±4
PC1
PC2
Results of two-tailed binomial tests
46
Contour density /
Dafni and Neal, 1997; Dafni et al., 1997
Contour density
Contour Density P < 0.02, n = 38
0
100
P < 0.04, n = 25
0
100 PC1: mean-4s.d. vs. mean+4s.d.
Training Training
Contour density 7.468 1.876
47
•
―
• ( )
• (2003/06)
• ISBN-10: 4320017382
• ISBN-13: 978-4320017382
www.amazon.co.jp
48