Noise removal using Cohen-Grossberg neural network for improving the quality of the decrypted image in color encryption
Yanling Liu, Jianxiong Zhang, Wansheng Tang · 2011
In this paper, a color image encryption method is proposed with the removal of noise generated during the transmission based on Cohen-Grossberg neural networks, where the color image is expressed in terms of the standard red-green-blue (RGB) space, and the corresponding pixel matrix is hidden by Arnold transform (AT). The Cohen-Grossberg neural network is added to store the hidden message as the stable equilibria, which achieves the noise removal. The hidden message without noise is recovered by performing AT with accurate iteration numbers. Experimental results show that the proposed method achieves effective resistance against transmission noise.