Hiding Coded Speech in Color Images using Zaslavsky-Map and AMBTC Technique

Abstract

The idea of hiding information is to shield a powerful message in public records. The present application-based hidden information is sensitive information disguised in many testing sectors, such as watermarking, fingerprinting, and steganography, which is the art of covered or hidden writing, which is covert contact with coded communications to hide the presence of a document on a communication channel from hostile attackers. This paper presents hiding speech in a coded image, depending on the quantization and discrete wavelet transformation (DWT). The proposed method used the discrete wavelet transform on a three-color band of covered images. Moreover, the term AMBTC which refers to “Absolute Moment Block Truncation Coding” is used for hiding information. The secret information is a short speech (representing a password for example) embedded based on the quantization level modification technique, which will be hidden in a codded image. The three color-bands of the image are transformed using discreet wavelet transform all decomposed regions are used for hiding except the Low-Low band is kept without any change to get high quality of the stego-image with respect to original image. In the bitmap, the secret bits of a speech data are replaced. The experimental results produced an average of PSNR for tested images is about 35.0. Secret speech is compressed using discrete cosine transform (DCT) about (35%-40%), and the result is converted to a binary form that will be embedded in a color image.