A Novel Color Image Encryption Scheme based on Hybrid Chaotic Maps

Abstract

A new adaptive color image encryption algorithm based on hybrid chaotic map is proposed. The algorithm is divided into three steps: key sequence generation, position scrambling and RGB component cross-diffusion. In the key sequence generation stage, improved one-dimensional chaotic map can be used to generate a key stream; in the location scrambling stage, the adaptive strategy is used to select the key stream, so that the generated scrambling sequence is related to the image data content; in the third stage, the RGB components are cross-diffused by adaptively selecting the encryption method to modify the pixel gray value to destroy the strong correlation between adjacent pixels. The use of adaptive strategies in the scrambling and diffusion correlates the key stream with the image data, making the encryption algorithm robust against known/selected plaintext attacks. Histogram analysis, correlation analysis between adjacent pixels, information entropy analysis and differential attack analysis show that the algorithm is safe and effective.