An effective video steganography method for biometric identification

Authors:
Yingqi Lu;Cheng Lu;Miao Qi
Affiliations:
School of Computer Science and Technology, Jilin University, China;School of Computer Science and Technology, Jilin University, China;School of Computer Science and Information Technology, Northeast Normal University, China
Venue:
AST/UCMA/ISA/ACN'10 Proceedings of the 2010 international conference on Advances in computer science and information technology
Year:
2010

Citing 6
Cited 0

Techniques for data hiding

IBM Systems Journal
Secure data hiding in wavelet compressed fingerprint images

MULTIMEDIA '00 Proceedings of the 2000 ACM workshops on Multimedia
Hiding a Face in a Fingerprint Image

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 3 - Volume 3
Hiding Biometric Data

IEEE Transactions on Pattern Analysis and Machine Intelligence
Locally linear discriminant embedding: An efficient method for face recognition

Pattern Recognition
Feature based RDWT watermarking for multimodal biometric system

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an effective video steganography method to protect the transmitted biometric data for secure personal identification. Unlike the most of existing biometric data hiding methods, the hiding content in this work is an image set for guaranteeing the valid identification, but not a single image or feature vector. On the basis of human visual system (HVS) model, both inter-frame and intra-frame motion information are considered to make the hiding method more invisible and robust. Instead of embedding the data redundantly for resisting attacks, the biometric image set is embedded into the video sequence discretely using discrete wavelet transform (DWT). Specially, the sequence number of each frame is embedded into the corresponding frame as a watermark for detecting the integrity of stego-video and guaranteeing the exactly extraction of the biometric image set. Finally, the extracted image set is identified by the biometrics system. Extensive experimental results demonstrate that the proposed method can achieve perfect imperceptibility and good robustness, and guarantee the secure biometrics.