238
Maria Calagna and Luigi V. Mancini
performed. Similarly, ODRL [24] is based on the following features: assets, identifying any physical or digital content; rights, identifying permissions on the assets;
and parties, including end-users and rights holders. Both the approaches may be
integrated as shown in [8].
In the following, we describe in detail two information hiding techniques, watermarking and steganography, that may be applied to the GIS field.
11.2.1 Watermarking
Given a cover that represents some valuable data (a song, a video, a map image) it is
possible to embed a digital watermark inside them [16, 17, 19]. After embedding,
the watermark and the cover are inseparable and they are used to generate the
watermarked content. Often, the watermark is invisible, so it is imperceptible in
the watermarked content. According to applications, watermarking systems can be
classified into two categories: fragile and robust. In a fragile watermarking scheme,
the watermark is designed to be fragile so as to detect and localize possible
modifications made to the watermarked content. Typical applications of fragile watermarking systems include content authentication (the watermark is used to detect modifications applied to the original content), copy control (the watermark
indicates whether the content can be copied or not) and transaction tracking (the
watermark records transaction history of the content, typically identifying first authorized user). Robust watermarking schemes are used for copyright protection: the watermark is designed to be robust against attacks so as to prove ownership of the image
(the watermark is used to prove ownership in a court of law). In addition, according
to the detection phase, watermarking systems may be classified into two categories:
non-blind or blind, based on the use of the original cover to extract the watermark
by the detector. The term oblivious is also used as a synonym for blind. Blind watermarking systems are preferable since they can easily be used in a distributed environment where the original cover is not required and multiple watermarked copies
may be delivered to the users.
11.2.2 Steganography
Steganography (literally, covered writing) is an ancient art to hide traditional writing.
Since Greek and Roman times, it has been used to convey a secret without raising any
suspicion [16]. The main feature of classical steganography consists of the encoding
system secrecy. Once it is discovered, it makes all future communications useless.
Modern steganography is based on the secrecy of a shared secret-key that the entities
use to communicate with each other. As for watermarking systems, steganography
systems embed some secret information inside a cover object, producing a stegoobject. Based on the underlying cryptographic techniques, steganography systems
are classified into secret-key and public-key, based on the use of a shared secretkey or a pair of secret-key and public-key for secure communication. Secret-key
stegosystems use a shared secret-key in embedding and extraction phases. A typical
Maria Calagna and Luigi V. Mancini
performed. Similarly, ODRL [24] is based on the following features: assets, identifying any physical or digital content; rights, identifying permissions on the assets;
and parties, including end-users and rights holders. Both the approaches may be
integrated as shown in [8].
In the following, we describe in detail two information hiding techniques, watermarking and steganography, that may be applied to the GIS field.
11.2.1 Watermarking
Given a cover that represents some valuable data (a song, a video, a map image) it is
possible to embed a digital watermark inside them [16, 17, 19]. After embedding,
the watermark and the cover are inseparable and they are used to generate the
watermarked content. Often, the watermark is invisible, so it is imperceptible in
the watermarked content. According to applications, watermarking systems can be
classified into two categories: fragile and robust. In a fragile watermarking scheme,
the watermark is designed to be fragile so as to detect and localize possible
modifications made to the watermarked content. Typical applications of fragile watermarking systems include content authentication (the watermark is used to detect modifications applied to the original content), copy control (the watermark
indicates whether the content can be copied or not) and transaction tracking (the
watermark records transaction history of the content, typically identifying first authorized user). Robust watermarking schemes are used for copyright protection: the watermark is designed to be robust against attacks so as to prove ownership of the image
(the watermark is used to prove ownership in a court of law). In addition, according
to the detection phase, watermarking systems may be classified into two categories:
non-blind or blind, based on the use of the original cover to extract the watermark
by the detector. The term oblivious is also used as a synonym for blind. Blind watermarking systems are preferable since they can easily be used in a distributed environment where the original cover is not required and multiple watermarked copies
may be delivered to the users.
11.2.2 Steganography
Steganography (literally, covered writing) is an ancient art to hide traditional writing.
Since Greek and Roman times, it has been used to convey a secret without raising any
suspicion [16]. The main feature of classical steganography consists of the encoding
system secrecy. Once it is discovered, it makes all future communications useless.
Modern steganography is based on the secrecy of a shared secret-key that the entities
use to communicate with each other. As for watermarking systems, steganography
systems embed some secret information inside a cover object, producing a stegoobject. Based on the underlying cryptographic techniques, steganography systems
are classified into secret-key and public-key, based on the use of a shared secretkey or a pair of secret-key and public-key for secure communication. Secret-key
stegosystems use a shared secret-key in embedding and extraction phases. A typical
