Application of Bayesian theory in multiband SAR images fusion for classification
Yue Jin · Journal of the Graduate School of the Chinese Academy of Sciences · 2008
In this paper,Bayesian theory is applied to classification of multiband SAR images.The familiar fusion methods like product,averaging and median rules are discussed,and on the base of Bayesian averaging method,three modified rules are proposed according to the relationship between classification accuracy and distance factor of SAR image.Experimental results show that multiband fusion can combine the preponderance and complementary information of each band and achieve classification results which couldn't be got in single band.The modified methods decrease the effect of the wrong classification information through weighting and accordingly improve the classification accuracy.