11 Information Hiding for Spatial and Geographical Data
239
issue in secret-key steganography concerns the difficulty of exchanging the secretkey between the communicating entities. This problem is overcome by public-key
steganography: the secret is embedded in the public-key of the receiver. Then, the
receivers can extract the secret from the stego-object by applying their secret-key. In
Sect. 11.6, we apply a public key stegosystem to describe a possible application to
geospatial maps.
While the main requirements for a watermarking system are the imperceptibility
and the robustness against possible attacks, the goal of steganography systems is
the undetectability of the secret message. In the following section, some properties
concerning watermarking and steganography systems are illustrated.
11.2.3 Properties of Information Hiding Systems
We refer to an information hiding system as a system that embeds a secret into a
cover-object, producing a stego-object. In the following, we use this terminology in
order to illustrate some properties of these systems.
Based on the redundancy of the coding, some information may be embedded in
a cover-object. The total number of bits that are required for encoding the secret
message is called “data payload.” For images, the data payload is equivalent to the
number of bits required to encode the secret message; for audio files, the data payload is equivalent to the number of bits of the secret message per second and, for
video files, it is equivalent to the number of bits of the secret message that can be
transmitted per second or on a frame-by-frame basis.
The Imperceptibility Requirement
The main requirement for the information hiding systems is imperceptibility, that is,
the stego-object and the cover-object maintain the same level of quality. This means
that embedding the secret is not intrusive, thus it does not degrade the original quality.
A common metric for the quality difference is the peak-to-signal-noise ratio (PSNR)
that is based on the minimum squared error (MSE). For images, the MSE is given by
MS E =
M
i=1
N
j=1
[I 1 (i, j) − I 2 (i, j)]/MN
(11.1)
where I 1 and I 2 represent two images of M × N pixels. Then, the PSNR is given by
PS NR = 20 log(max intensity/RMS E)
(11.2)
with max intensity indicating the maximum intensity value in the images and, RMSE
indicating the squared root of the MSE.
When the PSNR value is higher, the quality difference is worse. This metric
provides an estimate of the quality difference between two different objects (we are
interested in the quality difference between the cover-object and the stego-object).
Then, the relative values are more meaningful than the absolute values. The use of the
PSNR metric helps to analyze different information hiding systems and the distortion
due to the embedded information.
239
issue in secret-key steganography concerns the difficulty of exchanging the secretkey between the communicating entities. This problem is overcome by public-key
steganography: the secret is embedded in the public-key of the receiver. Then, the
receivers can extract the secret from the stego-object by applying their secret-key. In
Sect. 11.6, we apply a public key stegosystem to describe a possible application to
geospatial maps.
While the main requirements for a watermarking system are the imperceptibility
and the robustness against possible attacks, the goal of steganography systems is
the undetectability of the secret message. In the following section, some properties
concerning watermarking and steganography systems are illustrated.
11.2.3 Properties of Information Hiding Systems
We refer to an information hiding system as a system that embeds a secret into a
cover-object, producing a stego-object. In the following, we use this terminology in
order to illustrate some properties of these systems.
Based on the redundancy of the coding, some information may be embedded in
a cover-object. The total number of bits that are required for encoding the secret
message is called “data payload.” For images, the data payload is equivalent to the
number of bits required to encode the secret message; for audio files, the data payload is equivalent to the number of bits of the secret message per second and, for
video files, it is equivalent to the number of bits of the secret message that can be
transmitted per second or on a frame-by-frame basis.
The Imperceptibility Requirement
The main requirement for the information hiding systems is imperceptibility, that is,
the stego-object and the cover-object maintain the same level of quality. This means
that embedding the secret is not intrusive, thus it does not degrade the original quality.
A common metric for the quality difference is the peak-to-signal-noise ratio (PSNR)
that is based on the minimum squared error (MSE). For images, the MSE is given by
MS E =
M
i=1
N
j=1
[I 1 (i, j) − I 2 (i, j)]/MN
(11.1)
where I 1 and I 2 represent two images of M × N pixels. Then, the PSNR is given by
PS NR = 20 log(max intensity/RMS E)
(11.2)
with max intensity indicating the maximum intensity value in the images and, RMSE
indicating the squared root of the MSE.
When the PSNR value is higher, the quality difference is worse. This metric
provides an estimate of the quality difference between two different objects (we are
interested in the quality difference between the cover-object and the stego-object).
Then, the relative values are more meaningful than the absolute values. The use of the
PSNR metric helps to analyze different information hiding systems and the distortion
due to the embedded information.