Video-Object Steganography Mechanism to Overcome Compression Loss in Biometric Based Authentication

Abstract

When sensitive information is exchanged through wireless communication, it requires remote authentication. The Authentication is obtained by the biometric signal, which is difficult to forge copy and share. The method of steganography is employed in providing the authentication. The biometric signals is encrypted with a video-object and sent. The remote authentication is based on the semantic segmentation, chaotic encryption and data hiding. To authenticate user X remotely, X video-object is extracted then the biometric signal is encrypted by chaotic encryption. The encrypted biometric signal is inserted to the most significant wavelet coefficient of video-object by using QSWT (Qualified Significant Wavelet Tree). QSWT provide invisibility and resistant against lossy transition and compression. Now video-object along with biometric signal is extracted as stego-object which is compressed and sent to receiver. While decryption, IDWT (Inverse Discrete Wavelet Transform) is used to retrieve the biometric signal from videoobject. One of the challenging problem in decrypting is the biometric signal is not clear for authentication. This loss occurs due to compression of stegoobject. To overcome the above issue, Image Coder is used to detect the compression loss and Integrate Region Matching is used to reduce the compression loss. This method provides security in encryption and robustness against steganalytic attacks to various transmission losses and JPEG compression issues. This proposed method of biometric based authentication uses bandwidth in an efficient manner.