Unsupervised alignment of thousands of images

The task addressed in this thesis is the automatic alignment of an ensemble of misaligned images in an unsupervised manner. This application is especially useful in computer vision applications where annotations of the shape of an object of interest present in a collection of images is required. Performing this task manually is a slow, tedious, expensive and error prone process which hinders the progress of research laboratories and businesses. Most recently, the unsupervised removal of geometric variation present in a collection of images has been referred to as congealing based on the seminal work of Learned-Miller [21]. The only assumption made in congealing is that the parametric nature of the misalignment is known a priori (e.g. translation, similarity, a�ne, etc) and that the object of interest is guaranteed to be present in each image. The capability to congeal an ensemble of misaligned images stemming from the same object class has numerous applications in object recognition, detection and tracking. This thesis concerns itself with the construction of a congealing algorithm titled, least-squares congealing, which is inspired by the well known image to image alignment algorithm developed by Lucas and Kanade [24]. The algorithm is shown to have superior performance characteristics when compared to previously established methods: canonical congealing by Learned-Miller [21] and stochastic congealing by Z�ollei [39].

Wide-baseline keypoint detection and matching with wide-angle images for vision based localisation

This thesis addresses the problem of detecting and describing the same scene points in different wide-angle images taken by the same camera at different viewpoints. This is a core competency of many vision-based localisation tasks including visual odometry and visual place recognition. Wide-angle cameras have a large field of view that can exceed a full hemisphere, and the images they produce contain severe radial distortion. When compared to traditional narrow field of view perspective cameras, more accurate estimates of camera egomotion can be found using the images obtained with wide-angle cameras. The ability to accurately estimate camera egomotion is a fundamental primitive of visual odometry, and this is one of the reasons for the increased popularity in the use of wide-angle cameras for this task. Their large field of view also enables them to capture images of the same regions in a scene taken at very different viewpoints, and this makes them suited for visual place recognition. However, the ability to estimate the camera egomotion and recognise the same scene in two different images is dependent on the ability to reliably detect and describe the same scene points, or ‘keypoints’, in the images. Most algorithms used for this purpose are designed almost exclusively for perspective images. Applying algorithms designed for perspective images directly to wide-angle images is problematic as no account is made for the image distortion. The primary contribution of this thesis is the development of two novel keypoint detectors, and a method of keypoint description, designed for wide-angle images. Both reformulate the Scale- Invariant Feature Transform (SIFT) as an image processing operation on the sphere. As the image captured by any central projection wide-angle camera can be mapped to the sphere, applying these variants to an image on the sphere enables keypoints to be detected in a manner that is invariant to image distortion. Each of the variants is required to find the scale-space representation of an image on the sphere, and they differ in the approaches they used to do this. Extensive experiments using real and synthetically generated wide-angle images are used to validate the two new keypoint detectors and the method of keypoint description. The best of these two new keypoint detectors is applied to vision based localisation tasks including visual odometry and visual place recognition using outdoor wide-angle image sequences. As part of this work, the effect of keypoint coordinate selection on the accuracy of egomotion estimates using the Direct Linear Transform (DLT) is investigated, and a simple weighting scheme is proposed which attempts to account for the uncertainty of keypoint positions during detection. A word reliability metric is also developed for use within a visual ‘bag of words’ approach to place recognition.

Contrast enhancement of EPID images via difference imaging : a feasibility study

In this study, the feasibility of difference imaging for improving the contrast of electronic portal imaging device (EPID) images is investigated. The difference imaging technique consists of the acquisition of two EPID images (with and without the placement of an additional layer of attenuating medium on the surface of the EPID)and the subtraction of one of these images from the other. The resulting difference image shows improved contrast, compared to a standard EPID image, since it is generated by lower-energy photons. Results of this study show that, ¯rstly, this method can produce images exhibiting greater contrast than is seen in standard megavoltage EPID images and that, secondly, the optimal thickness of attenuating material for producing a maximum contrast enhancement may vary with phantom thickness and composition. Further studies of the possibilities and limitations of the di®erence imaging technique, and the physics behind it, are therefore recommended.

Digital processing of diffusion-tensor images of avascular tissues

Diffusion is the process that leads to the mixing of substances as a result of spontaneous and random thermal motion of individual atoms and molecules. It was first detected by the English botanist Robert Brown in 1827, and the phenomenon became known as ‘Brownian motion’. More specifically, the motion observed by Brown was translational diffusion – thermal motion resulting in random variations of the position of a molecule. This type of motion was given a correct theoretical interpretation in 1905 by Albert Einstein, who derived the relationship between temperature, the viscosity of the medium, the size of the diffusing molecule, and its diffusion coefficient. It is translational diffusion that is indirectly observed in MR diffusion-tensor imaging (DTI). The relationship obtained by Einstein provides the physical basis for using translational diffusion to probe the microscopic environment surrounding the molecule.

Applying incremental EM to Bayesian classifiers in the learning of hyperspectral remote sensing data

In this paper, we apply the incremental EM method to Bayesian Network Classifiers to learn and interpret hyperspectral sensor data in robotic planetary missions. Hyperspectral image spectroscopy is an emerging technique for geological investigations from airborne or orbital sensors. Many spacecraft carry spectroscopic equipment as wavelengths outside the visible light in the electromagnetic spectrum give much greater information about an object. The algorithm used is an extension to the standard Expectation Maximisation (EM). The incremental method allows us to learn and interpret the data as they become available. Two Bayesian network classifiers were tested: the Naive Bayes, and the Tree-Augmented-Naive Bayes structures. Our preliminary experiments show that incremental learning with unlabelled data can improve the accuracy of the classifier.

Toward a more robust facial expression recognition in occluded images using randomly sampled Gabor based templates

Occlusion is a big challenge for facial expression recognition (FER) in real-world situations. Previous FER efforts to address occlusion suffer from loss of appearance features and are largely limited to a few occlusion types and single testing strategy. This paper presents a robust approach for FER in occluded images and addresses these issues. A set of Gabor based templates is extracted from images in the gallery using a Monte Carlo algorithm. These templates are converted into distance features using template matching. The resulting feature vectors are robust to occlusion. Occluded eyes and mouth regions and randomly places occlusion patches are used for testing. Two testing strategies analyze the effects of these occlusions on the overall recognition performance as well as each facial expression. Experimental results on the Cohn-Kanade database confirm the high robustness of our approach and provide useful insights about the effects of occlusion on FER. Performance is also compared with previous approaches.

Unpacking Japan's 21st century "National Conversation" : images of the future beyond the iron cage of the "Catch Up" model

How does the image of the future operate upon history, and upon national and individual identities? To what extent are possible futures colonized by the image? What are the un-said futurecratic discourses that underlie the image of the future? Such questions inspired the examination of Japan’s futures images in this thesis. The theoretical point of departure for this examination is Polak’s (1973) seminal research into the theory of the ‘image of the future’ and seven contemporary Japanese texts which offer various alternative images for Japan’s futures, selected as representative of a ‘national conversation’ about the futures of that nation. These seven images of the future are: 1. Report of the Prime Minister’s Commission on Japan’s Goals in the 21st Century—The Frontier Within: Individual Empowerment and Better Governance in the New Millennium, compiled by a committee headed by Japan’s preeminent Jungian psychologist Kawai Hayao (1928-2007); 2. Slow Is Beautiful—a publication by Tsuji Shinichi, in which he re-images Japan as a culture represented by the metaphor of the sloth, concerned with slow and quality-oriented livingry as a preferred image of the future to Japan’s current post-bubble cult of speed and economic efficiency; 3. MuRatopia is an image of the future in the form of a microcosmic prototype community and on-going project based on the historically significant island of Awaji, and established by Japanese economist and futures thinker Yamaguchi Kaoru; 4. F.U.C.K, I Love Japan, by author Tanja Yujiro provides this seven text image of the future line-up with a youth oriented sub-culture perspective on that nation’s futures; 5. IMAGINATION / CREATION—a compilation of round table discussions about Japan’s futures seen from the point of view of Japan’s creative vanguard; 6. Visionary People in a Visionless Country: 21 Earth Connecting Human Stories is a collection of twenty one essays compiled by Denmark born Tokyo resident Peter David Pedersen; and, 7. EXODUS to the Land of Hope, authored by Murakami Ryu, one of Japan’s most prolific and influential writers, this novel suggests a future scenario portraying a massive exodus of Japan’s youth, who, literate with state-of-the-art information and communication technologies (ICTs) move en masse to Japan’s northern island of Hokkaido to launch a cyber-revolution from the peripheries. The thesis employs a Futures Triangle Analysis (FTA) as the macro organizing framework and as such examines both pushes of the present and weights from the past before moving to focus on the pulls to the future represented by the seven texts mentioned above. Inayatullah’s (1999) Causal Layered Analysis (CLA) is the analytical framework used in examining the texts. Poststructuralist concepts derived primarily from the work of Michel Foucault are a particular (but not exclusive) reference point for the analytical approach it encompasses. The research questions which reflect the triangulated analytic matrix are: 1. What are the pushes—in terms of current trends—that are affecting Japan’s futures? 2. What are the historical and cultural weights that influence Japan’s futures? 3. What are the emerging transformative Japanese images of the future discourses, as embodied in actual texts, and what potential do they offer for transformative change in Japan? Research questions one and two are discussed in Chapter five and research question three is discussed in Chapter six. The first two research questions should be considered preliminary. The weights outlined in Chapter five indicate that the forces working against change in Japan are formidable, structurally deep-rooted, wide-spread, and under-recognized as change-adverse. Findings and analyses of the push dimension reveal strong forces towards a potentially very different type of Japan. However it is the seven contemporary Japanese images of the future, from which there is hope for transformative potential, which form the analytical heart of the thesis. In analyzing these texts the thesis establishes the richness of Japan’s images of the future and, as such, demonstrates the robustness of Japan’s stance vis-à-vis the problem of a perceived map-less and model-less future for Japan. Frontier is a useful image of the future, whose hybrid textuality, consisting of government, business, academia, and creative minority perspectives, demonstrates the earnestness of Japan’s leaders in favour of the creation of innovative futures for that nation. Slow is powerful in its aim to reconceptualize Japan’s philosophies of temporality, and build a new kind of nation founded on the principles of a human-oriented and expanded vision of economy based around the core metaphor of slowness culture. However its viability in Japan, with its post-Meiji historical pushes to an increasingly speed-obsessed social construction of reality, could render it impotent. MuRatopia is compelling in its creative hybridity indicative of an advanced IT society, set in a modern day utopian space based upon principles of a high communicative social paradigm, and sustainability. IMAGINATION / CREATION is less the plan than the platform for a new discussion on Japan’s transformation from an econo-centric social framework to a new Creative Age. It accords with emerging discourses from the Creative Industries, which would re-conceive of Japan as a leading maker of meaning, rather than as the so-called guzu, a term referred to in the book meaning ‘laggard’. In total, Love Japan is still the most idiosyncratic of all the images of the future discussed. Its communication style, which appeals to Japan’s youth cohort, establishes it as a potentially formidable change agent in a competitive market of futures images. Visionary People is a compelling image for its revolutionary and subversive stance against Japan’s vision-less political leadership, showing that it is the people, not the futures-making elite or aristocracy who must take the lead and create a new vanguard for the nation. Finally, Murakami’s Exodus cannot be ruled out as a compelling image of the future. Sharing the appeal of Tanja’s Love Japan to an increasingly disenfranchised youth, Exodus portrays a near-term future that is achievable in the here and now, by Japan’s teenagers, using information and communications technologies (ICTs) to subvert leadership, and create utopianist communities based on alternative social principles. The principal contribution from this investigation in terms of theory belongs to that of developing the Japanese image of the future. In this respect, the literature reviews represent a significant compilation, specifically about Japanese futures thinking, the Japanese image of the future, and the Japanese utopia. Though not exhaustive, this compilation will hopefully serve as a useful starting point for future research, not only for the Japanese image of the future, but also for all image of the future research. Many of the sources are in Japanese and their English summations are an added reason to respect this achievement. Secondly, the seven images of the future analysed in Chapter six represent the first time that Japanese image of the future texts have been systematically organized and analysed. Their translation from Japanese to English can be claimed as a significant secondary contribution. What is more, they have been analysed according to current futures methodologies that reveal a layeredness, depth, and overall richness existing in Japanese futures images. Revealing this image-richness has been one of the most significant findings of this investigation, suggesting that there is fertile research to be found from this still under-explored field, whose implications go beyond domestic Japanese concerns, and may offer fertile material for futures thinkers and researchers, Japanologists, social planners, and policy makers.

Automated object identification using optical video cameras on construction sites

Visual recording devices such as video cameras, CCTVs, or webcams have been broadly used to facilitate work progress or safety monitoring on construction sites. Without human intervention, however, both real-time reasoning about captured scenes and interpretation of recorded images are challenging tasks. This article presents an exploratory method for automated object identification using standard video cameras on construction sites. The proposed method supports real-time detection and classification of mobile heavy equipment and workers. The background subtraction algorithm extracts motion pixels from an image sequence, the pixels are then grouped into regions to represent moving objects, and finally the regions are identified as a certain object using classifiers. For evaluating the method, the formulated computer-aided process was implemented on actual construction sites, and promising results were obtained. This article is expected to contribute to future applications of automated monitoring systems of work zone safety or productivity.

Comparison of standard image segmentation methods for segmentation of brain tumors from 2D MR images

In the analysis of medical images for computer-aided diagnosis and therapy, segmentation is often required as a preliminary step. Medical image segmentation is a complex and challenging task due to the complex nature of the images. The brain has a particularly complicated structure and its precise segmentation is very important for detecting tumors, edema, and necrotic tissues in order to prescribe appropriate therapy. Magnetic Resonance Imaging is an important diagnostic imaging technique utilized for early detection of abnormal changes in tissues and organs. It possesses good contrast resolution for different tissues and is, thus, preferred over Computerized Tomography for brain study. Therefore, the majority of research in medical image segmentation concerns MR images. As the core juncture of this research a set of MR images have been segmented using standard image segmentation techniques to isolate a brain tumor from the other regions of the brain. Subsequently the resultant images from the different segmentation techniques were compared with each other and analyzed by professional radiologists to find the segmentation technique which is the most accurate. Experimental results show that the Otsu’s thresholding method is the most suitable image segmentation method to segment a brain tumor from a Magnetic Resonance Image.

Shanghai : images of modernity

This chapter profiles China's biggest city and economic powerhouse, Shanghai. The authors examine the city’s use of culture to position itself as a global city and how a particular narrative of the city has informed western commentators and Shanghai policy makers. They also analyze the development of an arts and cultural infrastructure and the parallel separation of art and entertainment, with contemporary art as an unexpected challenge, but one the city successfully negotiated. They looks at the marketisation of culture and the context in which this takes place, tracing the connections between market reforms in culture and those in the wider economy. The authors are convinced that the half-formed or distorted use of western concepts like creative industries or creative clusters, rather than indicating a duplicity or an incomplete modernity actually highlight some of the complicities of canonical cultural policy.

Pattern recognition using invariants defined from higher order spectra : 2-D image inputs

A new algorithm for extracting features from images for object recognition is described. The algorithm uses higher order spectra to provide desirable invariance properties, to provide noise immunity, and to incorporate nonlinearity into the feature extraction procedure thereby allowing the use of simple classifiers. An image can be reduced to a set of 1D functions via the Radon transform, or alternatively, the Fourier transform of each 1D projection can be obtained from a radial slice of the 2D Fourier transform of the image according to the Fourier slice theorem. A triple product of Fourier coefficients, referred to as the deterministic bispectrum, is computed for each 1D function and is integrated along radial lines in bifrequency space. Phases of the integrated bispectra are shown to be translation- and scale-invariant. Rotation invariance is achieved by a regrouping of these invariants at a constant radius followed by a second stage of invariant extraction. Rotation invariance is thus converted to translation invariance in the second step. Results using synthetic and actual images show that isolated, compact clusters are formed in feature space. These clusters are linearly separable, indicating that the nonlinearity required in the mapping from the input space to the classification space is incorporated well into the feature extraction stage. The use of higher order spectra results in good noise immunity, as verified with synthetic and real images. Classification of images using the higher order spectra-based algorithm compares favorably to classification using the method of moment invariants

Position, rotation, and scale invariant recognition of images using higher-order spectra

A new approach to recognition of images using invariant features based on higher-order spectra is presented. Higher-order spectra are translation invariant because translation produces linear phase shifts which cancel. Scale and amplification invariance are satisfied by the phase of the integral of a higher-order spectrum along a radial line in higher-order frequency space because the contour of integration maps onto itself and both the real and imaginary parts are affected equally by the transformation. Rotation invariance is introduced by deriving invariants from the Radon transform of the image and using the cyclic-shift invariance property of the discrete Fourier transform magnitude. Results on synthetic and actual images show isolated, compact clusters in feature space and high classification accuracies