No description available.

No description available.

This thesis considers the problem of efficiently identifying and locating instances of classes of three dimensional objects by matching them with a single generic model that represents that entire class of object. Since a member of an object class will normally differ from the prototype that represents the class, the approach used here is to allow the model to stretch, or deform, to fit the object. The input image data may contain one or more objects of unknown identity and location, each of which is assumed to belong to a class of object for which there is a corresponding model. Both objects and models are represented by three dimensional surface data with the object data pre-segmented into surfaces of uniform curvature. The process of deforming and matching the models to the object data is achieved in two stages. In the first stage combinations of object surfaces are formed and a search made for suitable object to model correspondences. Simple constraints are developed to reduce the search space to an acceptable size. When a correspondence is achieved, an initial estimate of the stretch required in the model is made and the model that contains those surfaces is selected for further matching. Because only those models for which there is evidence in the image are selected for further matching, the search space is further reduced. The second stage of the process involves taking those models selected in the first stage and performing a rigorous geometric search for any remaining model to object correspondences. As part of this process, the locations of the objects in the image are predicted and the deformation parameters refined as new correspondences are found. The location and deformation parameters provide further constraints for the geometric search, reducing the search space still more. Recognition is demonstrated with a variety of objects, both synthetic and real, and the results discussed. The use of deformable models in object recognition was found to be a good means by which to represent and match objects from classes showing three types of deformation - scale, stretch and small variation. The model deformation as formulated enabled the identity of the corresponding objects and their parameters of deformation to be determined with accuracy and efficiency.

Many attempts have been made to overcome problems involved in character recognition which have resulted in the manufacture of character reading machines. An investigation into a new approach to character recognition is described. Features for recognition are Fourier coefficients. These are generated optically by convolving characters with periodic gratings. The development of hardware to enable automatic measurement of contrast and position of periodic shadows produced by the convolution is described. Fourier coefficients of character sets were measured, many of which are tabulated. Their analysis revealed that a few low frequency sampling points could be selected to recognise sets of numerals. Limited treatment is given to show the effect of type face variations on the values of coefficients which culminated in the location of six sampling frequencies used as features to recognise numerals in two type fonts. Finally, the construction of two character recognition machines is compared and contrasted. The first is a pilot plant based on a test bed optical Fourier analyser, while the second is a more streamlined machine d(3signed for high speed reading. Reasons to indicate that the latter machine would be the most suitable to adapt for industrial and commercial applications are discussed.

This thesis presents a numerical and experimental investigation on applications of ultralong Raman fibre lasers in optical communications, supercontinuum generation and soliton transmission. The research work is divided in four main sections. The first involves the numerical investigation of URFL intra-cavity power and the relative intensity noise transfer evolution along the transmission span. The performance of the URFL is compared with amplification systems of similar complexity. In the case of intracavity power evolution, URFL is compared with a first order Raman amplification system. For the RIN transfer investigation, URFL is compared with a bi-directional dual wavelength pumping system. The RIN transfer function is investigated for several cavity design parameters such as span length, pump distribution and FBG reflectivity. The following section deals with experimental results of URFL cavities. The enhancement of the available spectral bandwidth in the C-band and its spectral flatness are investigated for single and multi-FBGs cavity system. Further work regarding extended URFL cavity in combination with Rayleigh scattering as random distributed feedback produced a laser cavity with dual wavelength outputs independent to each other. The last two sections relate to URFL application in supercontinuum (SC) generation and soliton transmission. URFL becomes an enhancement structure for SC generation. This thesis shows successful experimental results of SC generation using conventional single mode optical fibre and pumped with a continuous wave source. The last section is dedicated to soliton transmission and the study of soliton propagation dynamics. The experimental results of exact soliton transmission over multiple soliton periods using conventional single mode fibre are shown in this thesis. The effect of the input signal, pump distribution, span length and FBGs reflectivity on the soliton propagation dynamics is investigated experimentally and numerically.

Photonic approaches to quantum information science and technology promise new scientific discoveries and new applications. Linear optics underpins all of these protocols, and the advent of integrated quantum photonics has has brought with it a step change in complexity and control over quantum photonic systems. As systems scale up, near-term nonclassical computational possibilities emerge, as does the prospect of practical quantum technologies. In this work we present a series of linear optical processors comprising the union of a multi-photon source, active and passive waveguide devices, and a single photon detection system; and use these systems to explore a multitude of quantum information processing protocols. VYe propose and implement efficient and robust techniques for device level characterisation of linear optical circuitry, where full scale tomographic techniques become intractable. We demonstrate machine level verification protocols for systems whose complexity renders them formally unverifiable, demonstrating the efficacy of our protocols on systems of up to five photons in 21 waveguides, generating Hilbert spaces of 50,000 dimensions. Finally we present a fully reconfigurable universal linear optical processor with sufficient control to implement all possible linear optical protocols up to the size of the circuit. We programme this device to implement heralded quantum logic gates and entangling operations, simultaneous bosons sampling and verification protocols and six dimensional complex Hadamard operations. These results and techniques will find application as large scale universal linear optical processors begin to replace existing and future prototype systems.

Multilayer reflectors consisting of 'stacks' of alternating layers of guanine crystals with cytoplasm gaps are present in many different animals including fish and spiders. Common to these reflectors is high birefringence in the guanine crystals and disorder in thickness of the layers in the reflector. The central theme of this thesis is the development of a theoretical framework for understanding the physics of these structures based upon the optics of anisotropic stratified media, and the Anderson localization of light in one dimension. The thesis begins with an investigation of the physical origin of broadband, polarization-insensitive reflectivity from three species of 'silvery' teleost fish: Clupea harengus, Sardina pilchardus and Sprattus sprattus. Reflectance spectrophotometry is used to characterise the spectral, angular and polarization dependence of the reflectivity, and digital holographic microscopy is used to investigate the dielectric properties of the birefringent guanine crystals. It is discovered that there are two 'populations' of guanine crystal present in the fish stratum argenteum that have different alignments of their optic axes in the multilayer structure. An anisotropic model of birefringent-guanine isotropic-cytoplasm reflectors is developed using the 4x4 matrix method and is fitted to the two-crystal system. The model is then used to investigate numerically the production of a non-polarizing reflection, and the proposed camouflage function of the fish reflectors. An explanation for the broadband, polarization-insensitive reflectivity of the two-crystal system in fish reflectors is then developed from the perspective of Anderson localization and the structural property of the localization length. This is achieved by constructing an analytical model of a bio-inspired isotropic-birefringent random stack system. The model system is shown to exhibit the polarization-insensitive localization of light, which is in contrast to isotropic dielectric random stack systems that all have strong polarization-dependence to the localization length. Finally, the application of the theoretical framework of Anderson localization to random stack models of animal multilayer reflector is explored. This includes: a review of the thickness disorder for reflectors in fish, spiders and cephalopods; and investigations of the relationship between thickness disorder, localization length spectra and reflectivity spectra at normal and oblique incidence.

Laser drilling is a technologically important process, used in a variety of material processing applications. In many cases, the precise size and shape of individual holes, as well as their location, are of considerable importance. The current work has concentrated on the details of the physical phenomena which take place during laser drilling and on formulating models for quantitative prediction of the effects of material properties and laser drilling conditions on the size and shape of the hole and of the surrounding heat-affected zone. Drilling various materials, including metals, composites and coated specimens, has been undertaken, using Nd:YAG and CO<SUB>2</SUB> lasers with power densities ranging from 10<SUP>9</SUP> to 10<SUP>11</SUP> W m<SUP>-2</SUP>. Experimental results obtained recently by other workers have also been examined in some detail. One of the issues of interest is the relative significance of the removal of material during drilling by vaporisation and by melt ejection. Melt ejection is more energy-efficient, but it tends to lead to irregularly-shaped holes and also causes more contamination of the surroundings. When drilling metallic materials, the level of melt ejection has been found to depend, not only on the thermophysical properties of the specimen, but also on the laser beam parameters. For a given pulse power, a shorter pulse width produces more concentrated heat fluxes, which promotes vaporisation. Improvements in hole quality are observed when multiple pulses are used. Multiple-pulse operation, in the form of a series of short intense pulses, allows vapour to escape before the pressure can build up and promote melt ejection. Little or no melt ejection occurs with carbon fibre reinforced polymer composite specimens, presumably because the fibres do not melt and the viscosity of the molten matrix is very high. Carbon fibres in the heat-affected zone around the hole were observed to have become swollen by up to about 50% in diameter. This is attributed to irreversible changes in the arrangements of the basal planes in the turbostratic structure, caused by rapid thermal expansion. This effect may have been accentuated by the pressurisation of fine pores within the structure of the fibres.

No description available.

No description available.