Charles Osborne

Algebraic construction of a new class of quasi-orthogonal arrays for steganography

Watermark recovery is often based on cross-correlating images with pseudo-noise sequences, as access to un-watermarked originals is not required. Successful recovery of these watermarks is determined by the (periodic or aperiodic) sequence auto- and cross-correlation properties. This paper presents several methods of extending the dimensionality of 1D sequences in order to utilize the advantages that this offers. A new type of 2D array construction is described, which meets the above requirements. They are constructed from 1D sequences that have good auto-correlation properties by appending rows of cyclic shifts of the original sequence. The sequence values, formed from the roots of unity, offer additional diversity and security over binary arrays. A family of such arrays is described which have low cross-correlation and can be folded and unfolded, rendering them robust to cryptographic attack. Row and column products of 1D Legendre sequences can also produce equally useful 2D arrays (with interesting properties resulting from the Fourier invariance of Legendre sequences). A metric to characterize all these 2D correlation based watermarks is proposed.

Image watermarking-a spread spectrum application

This paper discusses the feasibility of coding a robust, undetectable, digital watermark on a standard 512*512 intensity image with an 24 bit RGB format. The watermark is capable of carrying such information as authentication or authorisation codes, or a legend essential for image interpretation. This capability is envisaged to find application in image tagging, copyright enforcement, counterfeit protection, and controlled access. The method chosen is based on linear addition of the watermark to the image data. The patterns adopted to carry the watermark are adaptations of m-sequences in one and two dimensions. The recovery process is based on correlation, just as in standard spread spectrum receivers. The technique is quite successful for one dimensional encoding with binary patterns, as shown for a variety of gray scale test images. A discussion of extensions of the method to two dimensions, RGB format and non-binary alphabets is presented. A critical review of other watermarking techniques is included.

Local rules: their application in a kanban system

A theory of local rules was developed using Kauffman and Holland’s work on fitness landscapes. Local rules are used by individuals to increase their chances of survival on a fitness landscape. This paper reports on results from a simulation model of the local rules used by managers in the operation of a kanban system. In this case, local rules were used to optimise chances of survival by deflecting senior management criticism of potential stockouts in the system. The local rules used by the managers were modelled and their success and unintended consequences were reported. Some tentative conclusions about the effectiveness of local rules were advanced.

Collusion resistant fingerprinting of digital audio

Digital fingerprinting is a technique for tracing the distribution of multimedia content, and protecting it from unauthorized manipulation. Unique identification information is embedded into each distributed copy of the signal. In a collusion attack, fingerprints are combined to remove or distort the fingerprints. Audio signals are good candidates for fingerprinting, because of the forgiving nature of the human auditory system to cross-talk between channels. We use principal components of the audio signal to construct an abstract vector space. The fingerprints are ordered rotations in that space. The rotations are determined by arrays with good correlation properties. These arrays are embedded in real audio, and are imperceptible, according to a panel of experts. These fingerprints are resistant to an averaging collusion attack by hundreds or thousands of colluders, and can withstand a worst case RandNeg attack by up to 30 colluders.

Local Rules and Fitness Landscapes: A Catastrophe Model

This paper examines the imposition of “local rules” in five mail delivery centres of Australia Post. Local rules are patterns of behaviour used by subunits of an organization to optimise their payoff. These local rules may ultimately benefit the total organization. Fitness landscapes are used to examine the emergence of local rules in this workplace. This research examined the relationship between the time taken to sort the mail and the volume of this mail. Work rates show a clear catastrophe shift; that is, work rates suddenly drop when the volume of mail exceeds a certain level, as Postal Delivery Officers apply local rules to maximise gains inherent in the pay structures. Such behaviour is close to that predicted by Kauffman (1995) in computer simulations of lattices and may be indicative of the application of local rules in organizations. The implications of the use of local rules are that behaviour in social systems may be dictated by systemic and emergent processes which are outside immediate management control. A further implication is that organizations may be structured to a significant extent by such local rules.

The Euler characteristic on the face-centred cubic lattice

Abstract The Euler characteristic is an important topological invariant of an object, and it has been used in two and three dimensions to provide information about objects in binary digital pictures. Generally it is calculated by counting certain local patterns appearing in an object, and as such it has been defined for the cartesian lattices Z 2 and Z 3 , and the hexagonal lattice. In this paper we extend these results to the face-centred cubic lattice; this lattice being the natural extension of the hexagonal lattice to three dimensions. We show how the Euler characteristic may be defined, and that the definition is consistent with classical results. We also investigate means of implementing our definition.

Algebraic methods for multidimensional digital topology

We show how algebraic methods can be used to provide a mathematical framework suitable for the definition of multidimensional hypersurfaces in digital space, and for proofs of separation theorems. Our work is motivated by the need for a mathematical basis to provide a strong foundation for the creation of image processing algorithms in multidimensions; multidimensional images have been shown to arise naturally in areas as diverse as medical diagnosis and agricultural imaging. Whereas previous work in the area has been either combinatorial or has used the tools of point-set topology, we show how homology and cohomology groups can be defined in digital space. Our definitions are of a broad nature encompassing many of the standard adjacencies used to define digital objects. Given that in Euclidean space these groups satisfy conditions which provide for very neat proofs of separation theorems, we conjecture that an analogous theorem is true in digital space. We further show that the concept of orientability can be given a meaning in digital space more closely analogous to its classical meaning than definitions given previously in the image processing literature.

Local Rules: Emergence on Organizational Landscapes

This paper proposes that the theory of local rules provides a model for explaining organizational behavior as an emergent property of a fitness landscape. While local rule theory has its genesis in evolutionary biology, this paper links it to work in computational mathematical organizational theory. It further proposes that there are conditions, characterized by coadaptation, under which rules will survive in relatively stable forms, and other conditions, characterized by competition, under which local rules will change. The paper then discusses how catastrophe analysis can provide insights into changing patterns of organizational interactions. A discussion of methodology outline shows developments in agent-based simulation modeling can contribute to the development of local rule theory.

Metrics on the face-centered cubic lattice

The face-centered cubic lattice and its analogues in dimensions 4 and 5 are known to solve the sphere-packing problem in those dimensions. This property makes these lattices important for image processing in situations where high angular resolution is required, or if a good distance approximation is required. Given that the approximation to Euclidean distance is better for these lattices than for the standard Cartesian lattice, it is necessary to provide a distance measure for the lattices. We show how to construct two different metrics for the face-centered cubic lattice, and how this metric can be extended to higher dimensions.

Isotopy of 6-connected digital knots

Investigation of the topological nature of digitized knot structures is necessary for laying the foundations of a theory for dealing with the processing of such images. Images containing knots arise, for example, in the electron microscopy of DNA and RNA. The classical theory of knots is well developed, and one important result is that two knots are equivalent if their planar representations can be deformed into each other by a sequence of knot moves. We show here that digital knots, suitably defined, can be considered equal if they can be deformed into each other by a sequence of digital moves.