Data Hiding in Binary Image for Authentication and Annotation
Abstract—This paper proposes a new method to embed data in
binary images, including scanned text, figures, and signatures. The
method manipulates “flippable” pixels to enforce specific blockbased
relationship in order to embed a significant amount of data
without causing noticeable artifacts. Shuffling is applied before embedding
to equalize the uneven embedding capacity from region to
region. The hidden data can be extracted without using the original
image, and can also be accurately extracted after high quality
printing and scanning with the help of a few registration marks.
The proposed data embedding method can be used to detect unauthorized
use of a digitized signature, and annotate or authenticate
binary documents. The paper also presents analysis and discussions
on robustness and security issues.
Index Terms—Annotation, authentication, binary image, data
hiding, digital watermarking.
I. INTRODUCTION
DIGITAL watermarking and data hiding techniques have
been proposed for a variety of digital media applications,
including ownership protection, copy control, annotation, and
authentication. Most prior works on image data hiding are for
color and grayscale images in which the pixels take on a wide
range of values [2]–[9]. For most pixels, changing the pixel
values by a small amount may not be noticeable under normal
viewing conditions. This property of human visual system plays
a key role in watermarking of perceptual data [5], [6]. For images
in which the pixels take value from only a few possibilities,
hiding data without causing visible artifacts becomes more difficult.
In particular, flipping white or black pixels that are not
on the boundary is likely to introduce visible artifacts in binary
images.
Hiding data in binary image, though difficult, is getting
higher demands from our everyday life. An increasingly large
number of digital binary images have been used in business.
Handwritten signatures captured by electronic signing pads are
digitally stored and used as records for credit card payment
by many department stores and for parcel delivery by major
courier services such as the United Parcel Service (UPS). Word
Manuscript received April 1, 2001; revised March 28, 2002. This work was
supported in part by a New Jersey State R&D Excellence Award, NSF Grants
MIP-9408462 and CCR-0133704, and an Intel Technology for Education 2000
Grant. Part of this work was presented at the IEEE International Conference
on Multimedia & Expo (ICME’00) [1]. The associate editor coordinating the
review of this manuscript and approving it for publication was Dr. Sankar Basu.
M.Wu was with the Department of Electrical Engineering, Princeton University,
Princeton, NJ 08544 USA. She is now with the Department of Electrical
and Computer Engineering, University of Maryland, College Park, MD 20742
USA (e-mail: minwu@eng.umd.edu).
B. Liu is with the Department of Electrical Engineering, Princeton University,
Princeton, NJ 08544 USA (e-mail: liu@princeton.edu).
Digital Object Identifier 10.1109/TMM.2004.830814
processing software like Microsoft Word allows a user to store
his/her signature in a binary image file for inclusion at specified
locations of an electronic document. The documents signed
in such a way can be sent directly to a fax machine or be distributed
across a network. The unauthorized use of a signature,
such as copying it onto an unauthorized payment, is becoming
a serious concern. In addition, a variety of important documents,
such as social security records, insurance information,
and financial documents, have also been digitized and stored.
Because of the ease to copy and edit digital images, annotation
and authentication of binary images as well as detection of
tampering are very important.
This paper discusses data hiding techniques for these authentication
and annotation purposes, possibly as an alternative to or
in conjunction with the cryptographic authentication approach.
Such targeted applications calls for fragile or semifragile embedding
of many bits. It should be stressed that while it is desirable
for the embedded data to have some robustness against
minor distortion and preferably to withstand printing and scanning,
the robustness of embedded data against intentional removal
or other obliteration is not a primary concern. This is because
an adversary in authentication applications would have
much more incentive to counterfeit valid embedded data than to
remove them, and there is no obvious threat of removing embedded
data in many annotation applications.
Several methods for hiding data in specific types of binary
images have been proposed in literature. Matsui et al. [10]
embedded information in dithered images by manipulating
the dithering patterns and in fax images by manipulating the
run-lengths. Maxemchuk et al. [11] changed line spacing and
character spacing to embed information in textual images for
bulk electronic publications. These approaches cannot be easily
extended to other binary images and the amount of data that
can be hidden is limited. In [12], Koch and Zhao proposed a
data hiding algorithm which enforces the ratio of black versus
white pixels in a block to be larger or smaller than 1. Although
the algorithm aims at robustly hiding information in binary
image, it is vulnerable to many distortions/attacks, and it is
not secure enough to be directly applied for authentication or
other fragile use. The number of bits that can be embedded is
limited because the particular enforcing approach has difficulty
in dealing with blocks that have low or high percentage of
black pixels. However, the idea of enforcing properties of a
group of pixels via the local manipulation of a small number
of pixels can be extended as a general framework of data
embedding. Another approach of marking a binary document is
proposed in [13] by treating a binary image as a grayscale one
and manipulating the luminance of dark pixels slightly so that
No comments :
Post a Comment