Haar transformation for compressed speech hiding

Abstract EN

Steganography science is one of the most popular field in security direction.

In this paper an algorithm will be adopted to embed a compressed speech inside a gray image using discrete wavelet(Haar transformation).

in the beginning the speech was compressed up to its half original size by applying(Daubechies) then convert the speech data from decimal code to binary code and embed it inside Haar coefficients of the cover _image using the four sub bands (cA : low low,ch : high low,cv : low high,cd: high high) which got by applying the wavelet on the cover_ image.

measuring peak signal to noise ratio (psnr) to determine the accuracy of the stego_image with respect to the original image, mse and the correlation factors were checked show that the proposed algorithm has positive effect in field of speech hiding.the proposed technique in this research turned out to be able to hide speech data(audio) in the cover image and then extract the hidden data with storage rate (1) bits per pixel.

hiding capacity can be achieved using this method proportionally depends on cover_image size.

high frequency coefficients have also been shown to be better for data hiding in terms of perceptibility and intruders' cannot be able to recognize the cover medium (stego_image) which included secret data.