Abstract:Based on Bayesian hypothesis test, the method of making use of the prior information to improve the sensitivity and separability of deformation models was studied. The formulae for computing the sensitivity and separability with regard to prior information have been derived, and the relationship between the sensitivity and separability was also studied. The measurement of the sensitivity and the separability, which is very useful for the design of the deformation network, has been established for the situation with regard to prior information. The results show that the sensitivity and the separability of deformation model will be largely improved if reasonable prior information is used. Finally an example is given to show the improvement.

Key words: sensitivity separability prior information Bayesian hypothesis test