Comparison and Implementation of Biometric Inspired
Rupa Maan, Lovneesh Bansal · 2012
The recent progress in the digital multimedia technologies has offered many facilities in the transmission, reproduction and manipulation of data. However, this advancement has also brought the challenge such as copyright protection for content providers. In the present work, two techniques namely LSB and Biometric pattern (Skin tone) based steganography have been compared and result were evaluated on the basis of comparison of quality of image after embedding of data and change in PSNR value of the image. This project is developed for hiding and store information in any image file by using biometric pattern. Biometric Steganography is presented that uses skin region of images in DWT domain for embedding secret data. This work also showed the results for skin detection which performed pretty well in this case for identifying the skin regions from the given image. By embedding data in only certain region (here skin region) and not in whole image security is enhanced. To evaluate quality of stego image after embedding the secret message, Peak Signal to Noise Ratio (PSNR) has been used. The performance in terms of capacity and PSNR (in dB) is demonstrated for this method. From above results it is observed that this preserves histogram of DWT coefficients after embedding also. That is the main reason why there is not much visual difference between the cover image and embedded image. Performing biometric steganography offers respectable level of security. By adopting embedding algorithm that preserves histogram of DWT coefficients after embedding, prevents histogram based attacks.