Efficient color image encryption by color-grayscale conversion based on steganography

Traditional image encryption algorithms usually aim to encrypt the RGB channels of color plain images separately. In this paper, a novel dimension reduction method is proposed for color image encryption, where a color plain image is first converted into a grayscale image by using steganography technique followed by an encryption operation on the resultant grayscale image thereby reducing the dimension of the encryption. And the color information can be restored back without losing any details. More specifically, a color plain image is first converted into the YCbCr channels, where the Y channel is used as the steganography carrier and to hide the other two channels that are down-sampled by a bicubic interpolation. Our algorithm can compress and encrypt the color plain image to generate a grayscale cipher image so as to reduce the repeatability of encryption. Furthermore, only the used pixel values are encrypted, where a bridge is built between binary image encryption and traditional grayscale image encryption. The security and performance of the proposed image encryption algorithm are analyzed, and the unification between binary image encryption and grayscale image encryption is achieved.

Automated summary

In this paper, a novel method for color image encryption is proposed, which enhances the security and performance of encryption by reducing the dimension and compressing the color image. The unification between binary image encryption and grayscale image encryption is achieved.