of 
the 
the 
ely 
Lh 
the 
een 
ith 
ea, 
are 
M 
two 
eed 
one 
ave 
ern 
We 
ion 
ion 
in 
As it shown in table 3 the fragments 
numb.1, 3, 5, 6 3 decision function 
value are maximum for H hypothesis. 
BA 
Decision function value for fragment 
numb.4 which havn't movi pattern is 
Fragment 
maximum for Hoc hypothesis. 
numb.2 one is maximum for Ha hypothesis. 
This is consequence of partial pattern 
shading. 
Thus propoused suboptimal procedure makes 
absolute true result. 
REFERENCES 
Barmea D.I.,. Silverman H.P., .1979. A 
Class of Algorithms for Fast Image 
Registration. IEEE Trans. Comput., 2, pp. 
179-186. 
BaiftureiH JI.A., 3y6akoB  —B.I., 1960. 
BurereHte  curHaJuoB Ha doue cAJyuaituux 
HoMex. Cor.pamuo, Mockea. 
Duda Re, Hart -P., 1973. Pattern 
Classification and Scene Analysis. John 
Wiley, New York, pp. 40-42. 
Holben R.D., 1980. A NTI (Moving Target 
Identification) Algorithm to Passive 
Sensors. In: NABCON'8BO, May 20-22, 1, 
Dpp.114-119. 
433 
Jess B.P., 1989. TeopeTWHueocKHe  OCHOBH 
CTATHOTHYECKOÏ  pamsoTexHuKM. COB.pamto, 
MockBa. 
Jeruvx B.J., 1978. O6pacoTka mumbopmairm 5 
OIITMue cKHXx CHMCTEMAX NENCHTAUM. 
MaummocTpoeHue, Mockea, c. 47-52. 
Lo T., Gerson G., 1979. Guidance System 
Position Update by Multiple  Subarea 
Correlation. SPIE, 186. 
Koskol J., Allebach J., 1986. Estimating 
the Motion of a Maneuvering Target with 
Time-sequentially Sampled Imagery. 
Journal Optic Society, 3(9). 
Pratt W. 1978. Digital Image Processing. 
John Wiley, New York, pp.318-355. 
Stuller J., Krishmamurthy  G., 1983, 
Kalman Filter Formulation of Low-lewel 
Television Image Motion Estimation, - 
Computer Vision, Graphics and Image 
Processing, 21:169-204. 
lkpmax H.7., Mamkoc B.H., 1981. Teopus u 
TeXHMKA OOPDACOTKH PAIHOJOKILIHOHHOÏ 
mH$dopMauum Ha HOoHe-TOMex. Parmo H CRA3E, 
Mockea, c. 7-41.