The remote real time identification of road surfaces is an increasingly important task in the automotive world. The development of automotive active safety system requires a remote sensing technology that alerts drivers to potential hazards such as slippery surfaces caused by water, mud, ice, snow etc. This will improve the safety of driving and reduce the road accidents all over the world. This thesis is dedicated to the experimental study of the feasibility of an affordable short-range ultrasonic and radar system for road surface recognition ahead of a vehicle. It introduces a developed novel system which can recognize the surfaces for all terrains (both on-road and off-road) based on the analysis of backscattered signals. Fundamental theoretical analysis, extensive modelling and practical experiments demonstrated that the use of pattern recognition techniques allows for reliable discrimination of the surfaces of interest. The overall classification system is described, including features extraction and their number reduction, as well as optimization of the algorithms. The performance of 4 classification algorithms was assessed and evaluated to confirm the effectiveness of the system. Several aspects like the complexity of the classification algorithms and the priori knowledge of the environment were investigated to explore the potential of this research and the possibility of introducing the surface classification system into the automotive market in the nearest future.

High dynamic range (HDR) imaging technology has been widely implemented in digital microscopes for taking still images of high-contrast specimens. However, capturing HDR microscopic video is much more challenging. In this dissertation, an HDR microscopic video system based on GPU accelerated computing is presented. By combining CPU and GPU computing, it is possible to build a stable HDR video system using a single off-the-shelf camera. The computing efficiency analysis shows that capturing multiple frames of different exposure intervals, aligning consecutive neighbouring frames, constructing HDR radiance map and tone mapping the radiance map for display, can all be realised by using GPU computing to accelerate the processing speed. The experimental results were presented to show the effectiveness of the system and how HDR video can reveal much more detail than conventional videos. The idea of employing HDR imaging technology in 3D surface construction has been proposed as a solution to the Shape From Focus limitation. Shape From Focus (SFF) is the most effective technique for recovering 3D object shape in optical microscopic scenes. Although numerous methods have recently been proposed, less attention has been paid to the quality of source images, which directly affects the accuracy of 3D shape recovery. One of the critical factors impacting source image quality is the high dynamic range issue, which is caused by the gap between the high dynamic ranges of the real world scenes and the low dynamic range images that the cameras capture. To overcome this issue, a novel microscopic 3D shape recovery system based on high dynamic range (HDR) imaging technique is proposed. By combining SFF and HDR, it is possible to build a robust 3D system using a single off-the-shelf camera and a traditional optical microscope. Experiments on constructing 3D shapes of difficult-to-image materials have been conducted, in terms of metal and shining plastic surfaces where conventional imaging techniques will have difficulty capturing detail, and will thus result in poor 3D reconstruction. The experimental results show the proposed HDR-based SFF 3D method yields more accurate and robust results than traditional non-HDR techniques for a variety materials. After the analysis of HDR and Shape From Focus techniques, another project about microscopy was presented, which is tuberculosis bacteria detection. Tuberculosis (TB) is an infectious disease in low- and middle-income countries. There are many tools behind physical examinations for TB detection, but the most effective method is visual examination using microscopes, in terms of fluorescent microscopy and bright field microscopy. However, the former method is on average 10% more sensitive than the latter. This project not only aims to detect tuberculosis automatically to help technicians, but also aims at the construction of a subsequent autofocus system based on the detection of tuberculosis. The focus analysis, which is the initial step of shape from the focus technique, acted on the region of tuberculosis exists, regardless of the other areas. In this case, a new TB detection method based on Random Forest using fluorescent microscopic images was presented. Experiments on three types of classifiers, in terms of Random Forest (RF), linear SVM (LinSVM), Cross-Validation SVM (CVSVM), were conducted. The experimental results indicate that the RF-based learning method for TB bacteria classification using fluorescent images achieved higher performance than the other two machine learning methods.

Non-contact manipulation technologies present a useful and powerful means of handling particles or cells. Such techniques are of interest in regenerative medicine applications, and in particular the scalability of these techniques is an area of active research. Optical trapping is a precise and dextrous method of manipulating particles with the forces exerted by a laser beam, while acoustic trapping is a scalable technique capable of exerting a force on particles through standing wave resonance. These complimentary modalities can be utilised in a hybrid system to give a resultant technique that borrows from the strengths of each individual method. In this thesis, methods of force balancing, using optics and acoustics, are explored, both independently and in combination with each other. A technique for 3D acoustic trapping in glass capillaries is shown, utilising the two pairs of opposing channel walls and the air-water interfaces of two air bubbles as acoustic reflectors. Standing waves set up between these surfaces show discrete acoustic trapping sites for varying lengths of fluid cavity. A method of optical radiation force balancing is observed in a 3D potential energy landscape, using similar principles as seen in particle trapping with counter-propagating beams. Tuning of the radiation force balance in this system allows particles to, instead of being pinned to the surface by the radiation force from the optical pattern, become localised at discrete planes of trapping sites throughout the fluid volume. A hybrid force balance separation method using the optical and acoustic forces is devised using a single laser beam as the primary deflection mechanism with acoustic trapping providing both localisation and a force balance with the optics. Separation of different sized particles is observed, with larger scale optical deflection mechanisms and their resultant thermal effects demonstrated.

The main goal of this research is to support concept designers’ search for inspirational and meaningful images in developing mood boards. Finding the right images has become a well-known challenge as the amount of images stored and shared on the Internet and elsewhere keeps increasing steadily and rapidly. The development of image retrieval technologies, which collect, store and pre-process image information to return relevant images instantly in response to users’ needs, have achieved great progress in the last decade. However, the keyword-based content description and query processing techniques for Image Retrieval (IR) currently used have their limitations. Most of these techniques are adapted from the Information Retrieval research, and therefore provide limited capabilities to grasp and exploit conceptualisations due to their inability to handle ambiguity, synonymy, and semantic constraints. Conceptual search (i.e. searching by meaning rather than literal strings) aims to solve the limitations of the keyword-based models. Starting from this point, this thesis investigates the existing IR models, which are oriented to the exploitation of domain knowledge in support of semantic search capabilities, with a focus on the use of lexical ontologies to improve the semantic perspective. It introduces a technique for extracting semantic DNA (SDNA) from textual image annotations and constructing semantic image signatures. The semantic signatures are called semantic chromosomes; they contain semantic information related to the images. Central to the method of constructing semantic signatures is the concept disambiguation technique developed, which identifies the most relevant SDNA by measuring the semantic importance of each word/phrase in the image annotation. In addition, a conceptual model of an ontology-based system for generating visual mood boards is proposed. The proposed model, which is adapted from the Vector Space Model, exploits the use of semantic chromosomes in semantic indexing and assessing the semantic similarity of images within a collection.

This work concerns the interaction of intense, ultrashort laser pulses with fullerenes and carbon nanotubes. This includes the excitation and ionisation dynamics of gas phase fullerenes and the response of carbon nanotubes to intense ultrashort laser pulses. When ionising C60 with laser pulses of duration between 50 fs up to a few hundred fs, the ionisation mechanism has been proposed to be thermal in nature, with the electronic subsystem ‘hot’ and the vibrational system ‘cold’ at the time of ionisation. Recent results show an anisotropy in the photoelectron angular distribution which may suggest more direct mechanisms at work. Velocity-Map Imaging photoelectron spectroscopy results are presented for the ionisation of C60 and C70 at various wavelengths, pulse durations and intensities and the results are compared to theoretical models. The results are described well by a thermal ionisation mechanism in which a significant number of electrons are emitted during the laser pulse. Electrons may gain a momentum ‘kick’ from the electric field of the laser which results in an anisotropy in the photoelectron angular distributions. Peaks are observed, superimposed on the thermal background, in the photoelectron kinetic energy spectra of fullerenes ionised by ultrashort laser pulses which were previously assigned as Rydberg peaks. Photoelectron angular distributions of these peaks are presented for C60 and C70 ionised with laser pulses of various wavelengths. The binding energies and anisotropy parameters fitted to the peaks suggest that they are due to the population and one-photon ionisation of superatom molecular orbitals (SAMOs). The results rule out a direct multiphoton population mechanism for these states and show many similarities with Rydberg fingerprint spectroscopy. The fusion of carbon nanotubes has been observed under high energy electron beams and fullerene molecules have been shown to fuse together after irradiation with ultrashort laser pulses. Results are presented for experiments where fusion of carbon nanotubes with ultrashort laser pulses was attempted. Thin carbon nanotube films are analysed via Raman spectroscopy after irradiation by single laser pulses. A number of low frequency radial breathing mode peaks were observed which suggest that fusion may have taken place at certain areas of the sample.

W. Eric L. Grimson ; with contributions from Tomá³ Lozano-Pé²¥z, Daniel P. Huttenlocher. / Includes bibliographical references (p. 350-504) and indexes.

Graphene and layered two-dimensional (2D) materials have set a new paradigm in modern solid-state physics and technology. In particular their exceptional optical and electronic properties have shown great promise for novel applications in light detection. However, several challenges remain to fully exploit such properties in commercial devices. Such challenges include the limited linear dynamic range (LDR) of graphene-based photodetectors (PDs), the efficient extraction of photoexcited charges and ultimately the environmental stability of such atomically-thin materials. In order to overcome the aforementioned limits, novel approaches to tune the properties of graphene and semiconducting \ce{HfS2} are explored in this work, using chemical functionalisation and laser-irradiation. Intercalation of graphene with \ce{FeCl3} is shown to lead to a highly tunable material, with unprecedented stability in ambient conditions. This material is used to define photo-active junctions with an unprecedented LDR via laser-irradiation. Intercalation with \ce{FeCl3} is also used to demonstrate the first all-graphene position-sensitive photodetector (PSD) promising for novel sensing applications. Finally, laser-irradiation is employed, to perform controlled oxidation of ultra-thin \ce{HfS2}, which leads to induced strain in the material and a consequent spatially-varying bandgap. Such structure is used to demonstrate, for the first time, efficient extraction of photogenerated carriers trough the so-called ``charge-funnel'' effect, paving the way to the development of ultra-thin straintronic devices.

Handling large displacement optical flow is a remarkably arduous task. For instance, standard coarse-to-fine techniques often struggle to adequately deal with moving objects whose motion exceeds their size. Here we propose a learning approach to the estimation of large displacement between two non-consecutive images in a sequence on the basis of a learning set of optical flows estimated a priori between different consecutive images in the same sequence. Our method refines an initial estimate of the flow field by replacing each displacement vector by a linear combination of displacement vectors at the center of similar patches taken from a code-book built from the learning set. The key idea is to use the accurate flows estimated a priori between consecutive images to help improve the potentially less accurate flows estimated online between images further apart. Experimental results suggest the ability of a purely data-driven learning approach to handle fine scale structures with large displacements.

Highly stable optical frequency combs (OFCs), particularly those generated by modelocked lasers, have become important tools for frequency and time metrology, and spectroscopy. This is due to their ability to span wide bandwidths, to act as highly accurate frequency references, and to provide a direct link between the optical and radio frequencies. However, the narrow comb spacing of most mode-locked OFCs makes it difficult to access their individual modes for a wide range of other potential applications. This thesis investigates comb mode extraction from a 250 MHz spaced OFC by phase locking semiconductor lasers (slave lasers) to individual comb modes. This was achieved using optical injection locking in combination with a low bandwidth electronic feedback loop. The locking process forced a slave laser to emit at the same frequency as the comb mode it was locked to, but at its natural output power. Hence a locked slave laser effectively behaved as a ultra-narrowband filter with active gain. The locking process was characterised in terms of its long-term frequency stability over a period of 8 hours (minimum Allan deviation of less than 10-18) and its short term phase noise across a bandwidth from 100 Hz to 500 MHz (minimum integrated phase noise of 0.02 rad2). Amplification of the residual comb modes was measured and found to have a dependence on the master-slave frequency detuning. The results from numerical modelling found that this was due to phase modulation induced in the slave laser by the injected OFC and could always be suppressed by controlling the frequency detuning. Fourier synthesis of high repetition rate waveforms was explored as one of the potential applications of this phase locking technique. Multiple lasers were made coherent with one another by locking them to different modes of a common OFC. This enabled them to behave as different frequency components of a Fourier series to generate various waveforms. This was achieved by independently controlling the relative amplitude and phase of each slave laser, and combining them together. The generation of stable waveforms with at-top, triangular, parabolic, and sawtooth intensity profiles was demonstrated at a repetition rate of 100 GHz.

In this thesis we explore the wide parameter space of thulium doped silica fibre lasers,looking at various techniques for the control of operating wavelength and bandwidth as well as transverse beam profile. We demonstrate the extremely broad tunability of thulium fibre sources. Through various device architectures we show wavelength coverage over a 450nm wavelength range from 1660nm to 2115nm. With a simple external cavity containing an electronically controllable acousto-optic tunable filter we construct a wavelength agile tunable fibre source. This source demonstrates rapid wavelength selection and coverage within the thulium gain band including multiple simultaneous wavelength selection. Under fixed wavelength operation we show the generation of high power and efficiency 1726nm light in a monolithic thulium fibre source with output powers of up to 12.6W and internal laser slope efficiencies of 69%. Exploiting the fast cavity dynamics of pulsed fibre sources we demonstrate the generation of a bandwidth and wavelength flexible ASE source. Under pulsed operation we generate peak powers of greater than 2kW with tunable bandwidth from 24nm to 0.28nm corresponding to a change in coherence length from ~5mm down to ~60μm. Applying this source to nonlinear frequency conversion, we show the ASE seeded supercontinuum generation at output powers of up to 1.5W, covering a wavelength range of more than one octave from 1000nm to greater than 2400nm. Looking further ahead with this ASE source, we investigate the effects of propagation and amplification within a multimode waveguide, highlighting the potentially detrimental effects of modal interference and through spectral bandwidth tailoring, successfully supresses these effects. Showing stable high peak power amplification within a multimode thulium fibre amplifier. Finally, we introduce a novel mode selection technique utilising multimode fibre Bragg gratings. Allowing the selection of individual modes within a multimode fibre oscillator. We experimentally demonstrate this technique within a multimode thulium fibre source showing the electronically controllable selection and arbitrary switching between the fundamental and next higher order mode at switching speeds of up to 20kHz and output powers of greater than 5W. Extending this technique, we show the generation of near arbitrary superpositions of the first two guided modes with electronically tunable beam ‘flatness’ and other parameters. The prospects for further core area scaling are also discussed.