Abstract: A method of similarity measurement between cloud models was proposed based on mutual membership degree to solve the problems existing in current approaches,such as high consumption,unstable conclusion and limited applicability.Firstly,the membership degree of cloud models was defined with the ratio of common cloud area to itself cloud area according to expectation curve.Then,the similarity between cloud models was calculated with the mutual membership degree.Finally,the calculation formula and properties of similarity were produced,considering the calculation of the area with different intersection points and half cloud.Simulation results show that the proposed method can achieve rational similarity measurement between cloud models with high accuracy and low consumption in different situations.And the effectiveness of the method was further verified based on cloud model with an evaluation example.