be coherent. This 
nable elements de- 
jor part of the ship 
t, some part within 
lt to be detected. 
  
Figure 6: Integrated result of tracking and 
optical flow 
6. Conclusion 
  
Experiments to extract and track ships in the spa- 
tiotemporal images are performed. Two computer vi- 
it sion techniques are applied. The first is the template 
matching method and the second is optical low method. 
These methods are applied complimentary and extracted 
information are integrated. The results show that flow 
vectors of the ship part can be extracted. Future works 
directed experiments in more realistic situations. 
  
References 
, 
[1] Y.Hayashi, et al. ” An Image Ranging System at Sea 
IEEE Proc. of PLANS'94(1994). 
[2] M. Inaishi, et al." Basic study on neural system for 
supporting watching in navigation" The journal of 
Japan Institute of Navigation, 91, pp.203-211(1994). (in 
Japanese) 
| [3] W. Ito, et al." Template matching method adaptive for 
apparent scale change of object” Technical report of the 
Institute of Electronics, Information and Communica- 
tion Engineers, PRMU96-43(1996-06). (in Japanese) 
[4] Berthold K.P.Horn and Brian G.Schunck 
" Determining Optical Flow " Artificial Intelligence, 17, 
pp.185-203(1981). 
[5] Michael J.Black and Allan D.Jepson 
" Estimating Optical Flow in Segmented Images Using 
Variable-OrderParametric Mod els With Local Defor- 
mations " 
IEEE Trans. on Pattern Analysis and Machine Intelli- 
gence, Vol.18, No.10, pp.972-986(1996). 
  
791