Indian Journal of Science and Technology

Abstract

Steganography plays a vital role in the field of covert communication. This paper proposes a novel embedding scheme which embeds secret data adaptively. The proposed scheme employs integer to integer Lifting Wavelet Transforms (LWT) and pixel strength to specify suitable locations for data embedding. The integer LWT eliminates distortions, which is produced because of rounding process in the convolutional wavelet transforms. The performance of the proposed scheme is addressed using image quality metrics based on Peak Signal To Noise Ratio (PSNR) versus embedding ratio. Moreover, Bit Error Rate (BER) is utilized to assess robustness of the scheme. The experimental results show that the proposed is superior over several embedding techniques such as embedding based convolutional wavelet transforms and Lest Significant Bits (LSBs). The range of PSNR is gained between 51.52 and 22.15dB for range of embedding rate between 5% and 35%. Furthermore, the technique results error free for retrieving of secret data. In addition to the above features, the proposed technique increases security level of embedded data due to adaptive embedding process that increases randomly of embedding locations. The proposed scheme can be used in several important applications such as bank tag for authentication, military communications. Also, it can be implemented easily as a hardware chip and combine with covert imaging systems.

Keywords: Adaptive Embedding, Covert Communication, Lifting Wavelet Transform, Steganography, Security