En 
  
cxssa 
  
ce x sein 
wave theory 
The FFT analysis of radar image can detect the macro 
structure of the wave field. And FFT estimates the wave 
direction and frequency as uniform wave field. So it is 
difficult to explain the error of radar observation. For 
estimate the micro structure of wave field, we used the 
optical flow model as follows. 
Fig.9 shows the result of time series wave direction and 
wave speed by SLO in micro scale. And Fig.10 shows the 
9-D field of wave direction and wave speed by SLO 
method. In Fig.10 the average speed using analytical 
model, solitary wave model, is used for calibration the 
result from the result of radar observation. 
From Fig.9 SLO using radar image can estimate the wave 
speed well. But the wave direction is not so good. For 
explanation of this reason, Fig.11 is used. Fig 11 shows 
time series of the wave direction and wave height. The 
circle means the almost wave breaking 
point by Goda's wave breaking inception. The average 
error of the average of wave direction after neglecting the 
breaking point decreased from 37.6 deg to 35.8 deg. 
Fig 9 (c) and (d) show the wave direction field in almost 
breaking. These image indicate the randomness of the 
wave surface. This means the breaking wave is effective 
for estimating the micro structure of wave surface. 
6. CONCLUSIONS 
The marine radar is effective for observing wave direction 
and the accuracy is good for operational use. Another 
wave parameter must be correlated by the corrected 
radar image considering the shadowing effect. 
Wave direction and frequency by FFT is not effective to 
detect the macro structure of surface wave in nearshore 
region. 
Wave direction and wave speed by optical flow model, 
SLO, can detect the micro structure of wave parameter, 
such as wave speed and wave direction in nearshore 
region. And this model can offer the complements of the 
FFT image analysis. 
REFERENCES 
1) F.Ziemer, and W.Rosenthal : On the Transfer Function 
of Shipbone Radar for Imaging Ocean Waves, 
Proc.IGARSS'87, Ann Arbor, Michigan, pp.1559-1564, 
1987. 
2) F.Ziemer: Directional Spectra from Shipborn Navigation 
Radar during LEWEL, DIRECTIONAL OCEAN WAVE 
SPECTRA, pp.80-84,1989. 
3) M.D.Henschel, R.A.Paul, and B. M. Eid : Use of 
Satellite Synthetic Aperture Radar for ^ Operational 
Measurement of Ocean Wave Spectra, Proc. ERIM 
Second Themantic Conference: Remote Sensing for 
Marine and Coastal Environments, New Orleans, LA, 
Vol.1, PP.1-269 - |-273,1994. 
4) Shintaro Goto, and Kiyonori lisawa: Study on the 
Monitoring of 2-D Wave Parameter Using Simulated 
Image of Marine Radar, Journal of Photogrammetry and 
Remote Sensing, Vol.34, No.2,pp.36-44,1995. (in 
Japanese) 
5) B.K.P.Horn and B.G.Schunk: Determining Optical Flow, 
Artificial Intelligence, 17, pp.185-203,1981. 
6) J.K.Kearney,W.B.Thompson and D.L.Boley: Optical 
Flow Estimation: An Error Analysis of Gradient-Based 
Methods with Local Optimization, IEEE Trans. Pattern 
Anal. Machine Intell.,PAM-9,ppp229-244, 1984. 
  
  
  
  
  
N 12 
5 E 
5 S t 
© = 
$5 CH oe 
S. elf ids MM, d 
= | S 
A A 
wv) © u^ © — ' Q 
987850917858 24% 7858 5° 
gang rS SOR Sq RETRAIT PR SNS | 
Observation time (9,Jan,1996) 
—— Wave direction by Radar 
: Wave direction by Observational data 
77977 Wave height by Observational data 
Avarage error 37.6 deg (including Braking wave) © Almost braking point 
Avaräge error 35.8 deg (not including Braking wave) 
Braking wave inception (Goda) 
H nA l-exp| 
| T, 
-1.52—(1-1 stad ^p] [Jn 
Lo 
Fig 11 Time series of the wave direction, wave height and wave breaking inception 
703