Embedding of sound clips as a watermark in still images using discrete wavelet transform

Embedding of smaller images in larger images using the approach of watermarking is being effectively used for image scrutiny. With the advent of digital image processing, secure addition of watermarks in digitized images using various techniques has evolved. The use of wavelet transform for the said purpose has proved very useful. This paper presents a preliminary research carried out to embed audio clips in still images. The technique uses audio properties and final distortion threshold in the target image as parameters for decision making for various aspects of the implemented scheme. Some of these decisions are selection of either gray scale or color images, decomposition level for the wavelet transform, channel selection, sound sample and synthesis of the sound sample into mini samples. The research is being extended for embedding of audio samples in image sequences for video transmissions for secure audio communication applications.