High-capacity separable data hiding in encrypted image based on compressive sensing

A novel scheme for high-capacity separable data hiding in an encrypted image based on compressive sensing (CS) is presented. First, the original image is converted to the DC coefficient matrix, AC coefficient matrix and information embedding matrix by block discrete cosine transform. In the encryption phase, different encryption algorithms are applied on the DC matrix and the AC coefficient matrix, respectively. In the data hiding phase, the secret data is embedded in the reserved location of the encrypted image. After compressive sensing, the encrypted image with embedded data is generated. For the receiver, the way to obtain image content or/and additional data is separable according to the keys he owns. Experimental results show that the proposed method has the merits of high-capacity, anti-packet loss and cipher text compressibility.