630 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 10, NO. 4, JUNE 2000
A Source and Channel-Coding Framework for
Vector-Based Data Hiding in Video
Debargha Mukherjee, Member, IEEE, Jong Jin Chae, Sanjit K. Mitra, Fellow, IEEE, and
B. S. Manjunath, Member, IEEE
Abstract—Digital data hiding is a technology being developed
for multimedia services, where significant amounts of secure data
is invisibly hidden inside a host data source by the owner, for re-
trieval only by those authorized. The hidden data should be re-
coverable even after the host has undergone standard transforma-
tions, such as compression. In this paper, we present a source and
channel coding framework for data hiding, allowing any tradeoff
between the visibility of distortions introduced, the amount of data
embedded, and the degree of robustness to noise. The secure data
is source coded by vector quantization, and the indices obtained in
the process are embedded in the host video using orthogonal trans-
form domain vector perturbations. Transform coefficients of the
host are grouped into vectors and perturbed using noise-resilient
channel codes derived from multidimensional lattices. The pertur-
bations are constrained by a maximum allowable mean-squared
error that can be introduced in the host. Channel-optimized
can be used for increased robustness to noise. The generic approach
is readily adapted to make retrieval possible for applications where
the original host is not available to the retriever. The secure data
in our implementations are low spatial and temporal resolution
video, and sampled speech, while the host data is QCIF video. The
host video with the embedded data is H.263 compressed, before
attempting retrieval of the hidden video and speech from the re-
constructed video. The quality of the extracted video and speech is
shown for varying compression ratios of the host video.
Index Terms—Channel coding, data hiding, lattice VQ, source
coding, watermarking, wavelet transform.
I. INTRODUCTION
W
ITH THE RAPID growth in the mass of multimedia data
freely available through the Internet, and the associated
investment in standardization of hardware and software for open
Manuscript received March 26, 1998; revised August 27, 1999. This work
was supported by a University of California MICRO grant, with matching funds
from Lucent Technologies, Raytheon Missile Systems, Textronix Corporation,
and Xerox Corporation, and by the National Science Foundation under Grant
IRI9704785. This paper was recommended by Associate Editor H. Gharavi.
D. Mukherjee was with the Department of Electrical and Computer Engi-
neering, University of California, Santa Barbara, CA 93106 USA. He is now
with Hewlett-Packard Laboratories, Palo Alto, CA 94304 USA (e-mail: de-
bargha@hpl.hp.com).
J. J. Chae was with the Department of Electrical and Computer Engi-
neering, University of California, Santa Barbara, CA 93106 USA. He is now
with the Institute for Defense Information Systems, Seoul, Korea (e-mail:
chaejj@yahoo.com).
S. K. Mitra and B. S. Manjunath are with the Department of Electrical and
Computer Engineering, University of California, Santa Barbara, CA 93106 USA
(e-mail: mitra@iplab.ece.ucsb.edu; manj@iplab.ece.ucsb.edu).
Publisher Item Identifier S 1051-8215(00)04894-1.
transmission of such data, a mechanism for hidden data trans-
mission over the established infrastructure will provide an eco-
nomical alternative to expensive dedicated secure channels and
specialized terminals. The emerging technology of data hiding
[1]–[4] therefore presents an overwhelming urge in the world
today. Digital watermarking [5]–[12] is a closely related tech-
nology for copyright protection that is receiving much atten-
tion lately. Here, a small amount of a specific signature in-
formation, called the watermark, is invisibly hidden inside a
host data source, typically an image or a video sequence, by
the owner before distributing freely. The challenge is to enable
the owner to retrieve his original signature from the distributed
image or video to check authenticity, even after it has under-
gone significant transformations such as compression. While
most of the early work in this area assumes availability of the
original host to the retriever, the current trend is toward devel-
oping algorithms that allow retrieval even without knowledge
of the original host. In data hiding, the focus is on hiding larger
amounts of data in a host, for a wider range of applications
than just copyright protection. Only those authorized with the
knowledge of “how to” can retrieve the hidden data, even after
standard transformations like compression as required by the
transmission system, or media transformations as required by
the storage and distribution system, have been applied to the
host. Although it is possible for some applications to have the
original host data available during retrieval, the real strength
of a data-hiding scheme is the ability to make authorized re-
trieval possible even without the availability of the original host.
Data hiding has several defense-type applications, such as in-
conspicuous transmission of secret information over an insecure
but readily available medium such as the Internet. It can also
be used for transmitting various kinds of information securely
over the existing infrastructure dedicated for transmitting some-
thing else, such as transmitting hidden nonstandard format video
or hidden speech, using terminals specialized for transmitting
H.263 coded video, as in this work. Since a substantial amount
has already been invested in the development of the infrastruc-
ture for standard-based data transmission, it makes monetary
sense to try to use the same infrastructure for transmission of
nonstandard data. Another application is in secure transmission
of control information along with data in a commercial delivery
system. In general, data hiding makes possible invisible mixing
of different kinds of secure data along with standardized and
1051–8215/00$10.00 © 2000 IEEE