Digital Watermarking is the act of hiding a message related to an image within the image itself. Watermarking has many desirable properties like effectiveness, image fidelity and robustness.

Multilayer artificial neural network is used to achieve image compression. The input pixels are used as target values and the hidden layer output is the compressed image.

Our proposed technique uses an image as the vessel data and embeds secret information in the noise-like region of the bit-planes and watermark in the informative region of the bit-planes of the vessel image without deteriorating image quality. The secret image is first compressed by artificial neural network and then embedded in the vessel image. This technique makes use of the characteristics of the human vision system where a human eye cannot perceive any change in the information in a very complicated binary pattern