Large-capacity information hiding scheme based on minimum pixel modification

Abstract

Information hiding is an important technique for security-related applications, and it can be used wildly in many scenarios. Secret data embedding methods are proposed in spatial, frequency, and other domains by researchers in recent years. After investigating the existing information hiding schemes, we find that there is still room to improve the data embedding performance. Inspired by some works of literature, we propose a new scheme based on the modulus function with Minimum Pixel Modification (MPM22) for hiding secret data into carriers. The main idea of the scheme is to embed confidential bits into cover-image pixels while minimizing modification to original pixels. The correctness and high performance of the scheme are proved in mathematics and comprehensive experiments. Experimental result shows MPM22 has the largest embedding capacity (3 < bpp (bit per pixel) less than 7) compared with other schemes while maintaining the best image quality (peak signal-to-noise ratio (PSNR) >36.1 dB@k=3, PSNR>30.5 dB @k=4)。 The experimental result also confirms that our proposed scheme has higher performance than other schemes.