AN APPROACH FOR HIDING VOICE NOTES IN IMAGE COVER MEDIA USING DISCRETE WAVELET TRANSFORM

Abstract

The secret communications of information can be performed through diverse approaches including the spatial domain and transform domain. There is still the problem of insecurity of undercover communications due to vulnerable channels such as high imperceptibility, large embedding size, and low robustness. Image steganography has significantly been researched in the two decades as means for embedding secret messages of all kinds of media (text, sound, and image) into image carriers. Least Significant Bits (LSBs) and Most Significant Bits (MSBs) substitutions embedding approaches enable secret message bits superimpositions into the bits of image carriers. This paper proposes DWT decomposition of voice notes bits into two components; then, each component is hidden in either LSB or MSB position. The outcomes revealed that the mean, PSNR, and CR for MSB substitution at 7.45, 39.06 dB, and 11.12 outperformed the LSB substitution at 9.81, 38.25 dB, and 10.55 respectively. This implies the suitability of LSB and MSB for protecting secret communication through image cover medium. However, future works can consider the audio and image formats to measure the efficacy of the proposed image steganography.