banctec


 

Biometric techniques for assuring the integrity, authenticity and confidentiality of binary document images

Robert Sabourin, Ph.D. and Éric Granger, Ph.D.

Introduction

The accessibility of digitized document images allows for the dissemination of information through a wide range of computerized resources. However, there are currently three main challenges with managing this type of information – assuring the (1) integrity, (2) the authenticity, (3) confidentiality of document images. Two types of techniques have been deployed to overcome these challenges. The first is watermarking, which allows assuring the integrity and authenticity of digital images, and the second is cryptography (i.e., digital signatures, encryption), which enforces data privacy (authenticity and confidentiality). The research project described in this proposal seeks to apply one of these approaches, based on the scenarios that will be defined in the early stages (first months) of the project. Both of these approaches may exploit a user’s biometric features for increased reliability.

Watermarking

To assure integrity and authenticity, watermarking allows attaching check-codes of local areas inside the image itself, allowing localizing altered regions (Deguillaume et al., 2003). There exist two types of watermarking methods – fragile and robust watermarking. Fragile watermarking allows authenticating the ownership of a digital image even in the presence of noise or after the document has been changed. In contrast, robust watermarking allows detecting if a document image had been altered, and locating where these modifications have occurred.

In a hybrid approach (Muharemagic, 2004), both robust and fragile watermarking applied at the same time to enforce the authenticity and integrity of a binary (black and white) document image. Moreover, this approach allows for a randomization stage, where key pixels can be exchanged in order to increase the capacity of encoding the watermark into the image. The randomization key can be either static or dynamic. At first, we will consider use a static key, which can later be replaced by another key created using bio-cryptographic techniques. This hybrid approach may be essential to watermark bank cheque images and will have a key importance in the proposed project.

Biometric cryptography

Cryptographic or digital signatures may be used to detect modifications and assure the authenticity of document images. However, this approach is unable either to highlight which areas have been modified, or to assess the severity of the alteration (Deguillaume et al., 2003).

The use of biometric information to generate cryptography keys for secure transmission of digital documents is an emerging research area (Jain 2006). Several biometric signals like fingerprint, iris, etc, have already been considered as a way of exchanging digital documents securely. However, to our knowledge, the idea of using handwritten signatures to create cryptography keys has only been exploited in (Freire-Santos et al., 2006) using of the temporal signals present in signatures and (Freire-Santos et al., 2007) using handwritten signature images.

References on cryptography

Manuel Freire-Santos, J. Fierrez-Aguilar, M. Martinez-Diaz, and J. Ortega-Garcia, On the applicability of off-line signatures to the fuzzy vault construction, to appear, Proc. ICDAR2007, Curitiba, Brazil, September 2007.
M. Freire-Santos, J. Fierrez-Aguilar, and J. Ortega-Garcia, Cryptographic key generation using handwritten signature, Proc. SPIE, vol. 6202, pp 225–231, 2006.
A.K. Jain et al, Biometrics: A Tool for Information Security, IEEE Trans. on Information Forensics and Security, vol 1, no 2, pp 125-143, 2006.


References on watermarking

Edin Muharemagic, Adaptive Two-level Watermarking for Binary Document Images, PhD thesis, Florida Atlantic University, 2004.
F. Deguillaume, S. Voloshynovskiy and T. Pun , Secure Hybrid Robust Watermarking Resistant Against Tampering and Copy Attack, Signal Processing vol. 83, pp. 2133 – 2170, 2003.
Min Wu and Bede Liu, Data Hiding in Binary Image for Authentication and Annotation, IEEE Trans. on Multimedia, vol. 6, no. 4, pp 528-538, 2004.
Min Wu and Bede Liu, Data Hiding in Image and Video: Part I – Fundamental Issues and Solutions, IEEE Trans. on Image Processing, vol. 12, no. 6, pp 685-695, 2003.
Min Wu and Bede Liu, Data Hiding in Image and Video: Part II – Designs and Applications, IEEE Trans. on Image Processing, vol. 12, no. 6, pp 696-705, 2003.
Min Wu, Multimedia Data Hiding, PhD thesis, Princeton University, 2001

Acknowledgements

This project is finantially supported by Banctec and NSERC-CRSNG.

More:

An evolutive approach for the quick bio-watermarking of digital documents

Bio-crypto Systems Based on the Offline Signature Images

Intelligent Systems for Grayscale Rapid Watermarking

Biowatermarking of grayscale images