A robust and imperceptible steganography technique for SD and HD videos

Abstract

In this era of internet transfer of information is in digital form using multimedia files such as image, video, audio, etc. which relies on secure communication techniques to convey information safely. Due to the frequent transfer of videos over the internet nowadays they have become a good cover media for secure and covert communication in the form of video steganography. For efficient video steganography, it must fulfill its basic requirements such as capacity, imperceptibility, and robustness. In order to make a balance between imperceptibility and robustness, an efficient video steganography scheme is proposed for Standard Definition (SD) and High Definition (HD) videos. This scheme employs DWT (discrete wavelet transforms) for embedding the secret message inside the video frames utilizing only luminance (Y) component, and the security of the proposed scheme is strengthen by pre-processing the secret message with encryption before embedding. The embedding process is done by utilizing the middle-frequency sub-bands after applying second level 2-D DWT to the video frames to decompose it into 16 sub-bands. The performance of the proposed scheme is tested on different videos with quality metrics including peak signal to noise ratio (PSNR), structural similarity (SSIM) index, bit error rate (BER) and also by applying Gaussian and salt & pepper noise attacks. Moreover, the scheme is tested for the different level of compression on stego-video and also compared with U and V components used while embedding. Experimental results show that for both types of videos (HD and SD) the proposed scheme is able to achieve high imperceptibility. Further, it also provides robustness against different types of noise attacks and different compression levels which makes the proposed scheme evident for secure data transmission.