In order to denoise image more effectively, an improved Contourlet transform denoising algorithm based on gray relational degree was proposed. On one hand, considering the gray relational degree and inter-scale from the high-frequency sub-band and low frequency sub-band by Contourlet transform, the Bayes threshold was improved; On the other hand, in order to achieve the purpose of adaptive denoising, the characteristics of Contourlet coefficients were used to improve the compromising threshold function. The experimental results show that the proposed algorithm can denoise image effectively, get higher PSNR and better visual quality, and has a good practicability.

To improve the computational efficiency in the threshold signature scheme, the authors proposed a new threshold signature scheme based on bilinear pairing via combining Gennaro's (GENNARO R, JAREAKI S, KRAWCZYK H, et al. Secure distributed key generation for discrete-log based cryptosystem. Journal of Cryptology, 2007, 20(1): 51-83) distributed secret key generation solution and Gu's (GU K, JIA W J, JIANG C L. Efficient and secure identity-based signature scheme. Journal of Software, 2011, 22(6): 1350-1360) signature scheme in the standard model. There was no trusted dealer for secret key share distribution and each party could verify the validity of some important information, which guaranteed the proposed scheme can avoid the malicious private key generator attack and public key share replacing attack. The comparison results with two previous threshold signature schemes show that the proposed scheme needs less pairing computation and raises the computational efficiency.