ION, 
aring in 
thod for 
mulated 
0%. 
s Sea of 
inter of 
> length. 
is lower 
eory on 
anslation 
vance for 
by using 
lication 
on. 
jn near 
arameter 
analysis 
1 the sea 
y as the 
orrection 
rence of 
> radar to 
from the 
on polar 
nsformed 
ysis. The 
ed by 
  
Data correcyion of 
distribution 
| 
| Coordinate transformation | 
| Pre-Processing by data Window | 
  
  
  
  
  
  
| Two-dimensional FFT | 
| 
| Filtering Wave namber spectrum | 
  
  
  
Calculation of the wave direction 
and wave length 
  
  
  
END 
Fig.1 Flow diagram showing the analysis 
procedures. 
F(kx,ky)= 
Qc 
Em)” + Jf noy) * expCilleoxekyy)dxdy 
Oc 
N 1) 
T(X,y) is the radar image defined on x-y 
coordinate system. And Wave number spectrum 
is defined by 
S(kx,ky)= | F(kx,ky)* | eerie, 2) 
The relationship between wave number spectrum 
and 2-D wave parameter is shown in Fig.3. 
We use the discrete Fourier component, eq.2) is 
Written as follows. 
S(m,n)= | F(m,n)” | (MAX\(NAy) … 3) 
Where -(M/2-1) mx (M/2-1), and -(N/2-1) €n 
=(N/2-1). Wave number k, wave length L, 
wave frequency T, and wave direction ais 
defined as Fig.3. In this study we use small 
amplitude wave theory in definition of T vs. 
wavenumber (m,n). 
L=@2 mya Y one. 4) 
where g is the gravity acceleration, and h is the 
averaged depth of target area. 
Paemze | ® 
FILLE hh ks Tl kid 
  
  
Wave number : 
k= | Akmn | =2xz/L 
Wave length : 
L=LO// (m2+n2) 
Frequency : 
T=+ 2 x L/(g tanh@2 x h/L) 
Wave direction : 
& = x /2-tan! n/m 
  
  
nAk 
  
  
  
  
  
  
lw 2 m-1 m+1 i 
m Length of analyzi : 
À k=2 x /LO 8 analyzıng area ; LO 
Depth : h 
yo 
Fig.3 The relationships between wave number 
spectrum and wave parameter 
3. 2-D WAVE PARAMETER FROM 
SIMULATED IMAGE OF 
MARINE RADAR 
3.1 Simulation of Marine Radar Image 
Image of marine radar is simulated by FFT. 
Firstly 2-D Gaussian distribution is made as 
Fig.4. where A is the peak value of Gaussian 
253 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996