Stegnographic Method for Biometric Authentication

In wireless communications sensitive data is frequently changed, requiring remote authentication. Remote authentication involves the submission of encrypted data, along with visual and audio cues (facial images/videos, human voice etc.). Nonetheless, malicious program and different attacks will cause serious issues, particularly in cases of remote examinations or interviewing. This paper proposes a sturdy authentication mechanism supported semantic segmentation, chaotic cryptography and knowledge concealment. Assuming that user X needs to be remotely documented, initially X�s video object (VO) is mechanically segmental, employing a head and-body detector. Next, one amongst X�s biometric signals is encrypted by a chaotic cipher. Subsequently the encrypted signal is inserted to the most vital riffle coefficients of the VO, victimization its Qualified Significant riffle Trees (QSWTs). QSWTs give invisibility and vital resistance against loss transmission and compression, conditions that area unit typical in wireless networks. Finally, the Inverse distinct riffle rework (IDWT) is applied to supply the stego-object (SO). Experimental results, regarding: (a) security deserves of the planned cryptography theme, (b) strength to steganalytic attacks, to numerous transmission losses and JPEG compression ratios and (c) information measure potency measures indicate the promising performance of the planned biometrics-based authentication theme.