The American alligator, Alligator mississippiensis, is a large extant archosaur and member of the Order Crocodilia. Crocodilian ontogeny has been studied in great detail, the skull being of particular interest. One aspect of the skull left unstudied is how individual cranial elements change through ontogeny independent of one another. This study observed morphological change in a growth series of 34 specimens of A. mississippiensis from ETSU Vertebrate Paleontology Lab collections. The premaxilla, maxilla, nasal, jugal, frontal, and parietal were analyzed using landmark morphometrics. The frontal, jugal, and parietal showed more allometric growth with the orbits reducing in size posteriorly. The premaxilla, maxilla, and nasal showed more isometric growth. This suggests the common observation that the snout elongates with age is mistaken. The cranium showed allometric growth in very early in life but more isometric growth after that. Unique to this study, the premaxilla showed almost no shape change throughout ontogeny.

Traditional morphometric method measures the linear distance between two points on the body surface. Statistical techniques, mainly covariance analysis and principle component analysis, are respectively used in single- and multi- variable statistical inferences. Their purpose is to overcome the bias due to allometric growth. Geomorphometric methods (mainly superimposition method) studied the size and shape of organisms was developed at recent decade. These methods not only measure and analyze body shape and size directly, but are immune to the disturbance of allometric growth. They also enable scientists to study organic structure on a three dimensional space. In this study, coordinates of landmarks on beaks surface were recorded to give information of shape as well as caliper distances. Superimposition method (Procrustes residuals) was applied to examine the difference between sexes, local populations and among different species of octopus. And the phenograms based on caliper distance and superimposition method were compared with the hypothesis of Norman and Sweeney on Octopus phylogeny. Samples of all 11 species of octopus, belonging to genus Cistopus and three species-group of genus Octopus were collected around the water of Taiwan. No significant differences on landmark shape between sexes of O. aegina and of O. marginatus were found through principal component analysis. Samples of O. marginatus from Tungkang and Dahsi could be discriminated through the first principal component. The result does not match with that from DNA sequence analysis already reported. Such conflicts were considered the result of environmental effects. Canonical discrimination method was used for two types of data, namely Procrustes residuals and caliper distances. All species were significantly different from each other. A discriminate function based on Procrustes residuals data reclassified 92.7% of the specimens correctly, incomparing to the 86.1% based on caliper distances. Phenogram constructed from a matrix of Mahalanobis distance (D2) also showed different result. It was concluded that: 1) Result of geomorphometric analysis based on landmark data is not compatible with the hypothesis of Norman and Sweeny. 2) Beak characters based on caliper distance are suitable for discrimination between genera or species-groups, and support the separation of O. luteus from O. minor and O. sp. TW35. Our study suggests that, either traditional morphometric method or the new geomorphometric method is a better tool for showing environmental effects than for phylognetic studies.

This dissertation is a theoretical and historical account of urban sculpture in the U.S. following World War II. The title refers to an example set by corporations during the 1940s and 1950s for commissioning modernist office towers and abstract sculpture that fundamentally shaped the early history of a modern public art in the U.S. This corporate model was taken up by American cities during the 1960s in the construction of new civic centers that combined large-scale, abstract sculpture with glass and steel city office buildings. Federal funding further encouraged new sculpture commissions, which proliferated across the U.S. Emerging theories about visual communication impacted both urban planning and the corporate image during this period, as urban renewal reshaped cities for maximum legibility and corporations commissioned designers to create new trademarks. I argue that these twin aims conditioned the planning, production, and distribution of urban sculpture, whose status oscillated between the landmark within urban planning and the trademark of corporate America, between a concrete city element and an abstract symbol. I tell the history of post-war urban sculpture through three case studies. In the first case study, I examine three significant sculpture commissions for urban building lobbies realized by the architects Skidmore, Owings & Merrill during the 1950s: Harry Bertoia’s screen (1954) at the Manufacturers Trust Company Bank on New York’s Fifth Avenue; Richard Lippold’s Radiant “I” (1958) at the Inland Steel Company Headquarters Building in Chicago; and Alexander Calder’s mobile (1959) for the Chase Manhattan Bank branch at 410 Park Avenue. In the second case study, I trace the parallel trajectories of urban renewal in downtown Grand Rapids, Michigan and Alexander Calder’s fountains and stabiles made for World’s Fairs and international expositions, which intersected in La Grande Vitesse (1969), the National Endowment for the Arts’ first sculpture commission for its Art in Public Places program. In the third case study, I look at three sculptures produced by the fabricator Lippincott Inc., either as a series or in multiple editions, during its first five years of operation: Tony Rosenthal’s cubes (1967-68), Barnett Newman’s Broken Obelisk (1963-67), and Claes Oldenburg’s Geometric Mouse (1969-71). / text

This thesis describes a method for the recognition of objects in an unconstrained environment with a widely ranging illumination, imaged from unknown view points and complicated background. The general problem is simplified by placing specially designed patterns on the object that allows us to solve the pose determination problem easily. There are several key components involved in the proposed recognition approach. They include pattern detection, pose estimation, model acquisition and matching, searching and indexing the model database. Other crucial issues pertaining to the individual components of the recognition system such as the choice of the pattern, the reliability and accuracy of the pattern detector, pose estimator and matching and the speed of the overall system are addressed. After establishing the methodological framework, experiments are carried out on a wide range of both synthetic and real data to illustrate the validity and usefulness of the proposed methods. The principal contribution of this research is a methodology for Tagged Object Recognition (TOR) in unconstrained conditions. A robust pattern (calibration chart) detector is developed for off-the-shelf use. To empirically assess the effectiveness of the pattern detector and the pose estimator under various scenarios, simulated data generated using a graphics rendering process is used. This simulated data provides ground truth which is difficult to obtain in projected images. Using the ground truth, the detection error, which is usually ignored, can be analysed. For model matching, the Chamfer matching algorithm is modified to get a more reliable matching score. The technique facilitates reliable Tagged Object Recognition (TOR). Finally, the results of extensive quantitative and qualitative tests are presented that show the plausibility of practical use of Tagged Object Recognition (TOR). The features characterising the enabling technology developed are the ability to a) recognise an object which is tagged with the calibration chart, b) establish camera position with respect to a landmark and c) test any camera calibration and 3D pose estimation routines, thus facilitating future research and applications in mobile robots navigations, 3D reconstruction and stereo vision.

Over the past years, we have made countless explorations to promote urbanisation. The whole project begins with my interest in possibility of non-urbanized area. And desert as the largest account of untouched area on earth, it is an interest topic to explore how human activities will take place in a natural context rather than urban context. The whole project begins with my interest in possibility of non-urbanizsed area. And desert, as the largest account of untouched area on earth, is an interest topic to explore how human activities will take place in a natural context rather than urban context. I started with Burningman Festival in Black Rock city. It is temporary city of 7 days located in desert. It is often reported that upon leaving Burningman, participants feel possessed of a desire to make their year-round world feel like life during the festival. So I decided to set up something to make it a longer festival site. A hiking routine with several rest station heading to the camp site came into being. In desert scale, it is a simple line pointing at the camp centre. In people scale, it hybrids the nature pattern abstracted from desert Band the strong geometric pattern. The destination steps a little higher than other points, where people can enjoy the spectacular view

Localization is a crucial topic in navigation, especially in autonomous vehicles navigation. It is usually done by using a global positioning system (GPS) sensor. Even though there have been many studies of vehicle localization in recent years, most of them combine GPS sensor with other sensors to get a more accurate result [1]. In this thesis, we propose a novel image-based vehicle localization by utilizing vision sensor and computer vision techniques to extract vehicle surrounding text landmarks and to locate the vehicle position. Firstly, we explore the feasibility of image-based vehicle localization by using text landmark of a position to locate vehicle position. A text landmark model, a location matching algorithm and a basic localization model are proposed, which allow a vehicle to find the best matching location in the database by cross-checking the text landmarks from query image and reference location images. Secondly, we propose two more robust localization models by applying vehicle moving distance and heading direction data as part of inputs, which significantly improve the localization accuracy. Finally, we simulate an experiment to evaluate our three different localization models and further prove the robustness of our model through experimental results. / Master of Science / In modern days, global positioning system (GPS) is the major approach to locate positions. However, GPS is not as reliable as we thought. Under some environmental situations, GPS cannot provide continuous navigation information. Besides, GPS signals can be jammed or spoofed by malicious attackers. In this thesis, we aim to explore how to locate the vehicle’s position without using GPS sensor. Here, we propose a novel image-based vehicle localization by utilizing vision sensor and computer vision techniques to extract vehicle surrounding text landmarks and to locate the vehicle position. Various tools and techniques are explored in the process of the research. With the explored result, we propose several localization models and simulate an experiment to prove the robustness of these models.

Master of Landscape Architecture / Department of Landscape Architecture/Regional and Community Planning / Brent Chamberlain / Landmarks are universal components of human urbanization. We are a species driven to mark the land with symbolic structures and craft meaning in our built environments. From ancient wonders such as Stonehenge to modern icons like the St. Louis Arch, we have been designing landmarks since the dawn of civilization. Cities, towns, and neighborhoods incorporate landmarks as elements of cultural expression and tools for navigation. Individuals use landmarks as reference points to create an internal cognitive map, permitting more efficient navigation throughout a city and contributing to a heightened sense of place. To aid in research regarding the role of landmarks on cognitive maps and place-identity, we have designed a novel testing paradigm in which subjects wear a virtual reality (VR) head-mounted display (HMD) and traverse a hypothetical urban environment using a gaming controller. The virtual environment (VE) features a gridded street network measuring 5x5 blocks and guides subjects along a fixed route through residential, park, commercial and industrial districts. Along this fixed route, subjects are exposed to ten distinct landmarks. After navigating the VE, subjects are tasked with delineating their perceived route, landmark locations, and district boundaries through map drawing tasks on grid paper as well as a scene recognition task. The most significant finding revealed landmark configuration accuracy to be highly correlated with performance on the route recall and moderately correlated with performance on the scene recognition task. This suggests that, regardless of the landmark type, individuals who more precisely recalled landmark locations also navigated the route and identified scenes more accurately. Landscape and urban planners can leverage these findings to advocate for the strategic inclusion of landmarks throughout an urban fabric, which we term Landmark Configuration Plans (LCP).

Architektūros vietoženklių statyba – vienas ryškiausių visuotinių miestokūros reiškinių, kuris Vakaruose stebimas ir nagrinėjamas jau daugiau nei dešimtmetį. Tačiau į urbanistinėje aplinkoje išsiskiriančius architektūros objektus dar dažnai žvelgiama pro pusės amžiaus teorijų prizmę. Pirmojoje tokio pobūdžio disertacijoje pasiūlytas miesto įvaizdį formuojančių šiuolaikinės architektūros vietoženklių tyrimo modelis, kuris leidžia išsiaiškinti miesto įvaizdį formuojančių šiuolaikinės architektūros vietoženklių reikšmę. Teoriniai aspektai čia yra suderinti su praktiniais. Metodologiniu pagrindu čia tapo R.Barthes išplėtoti teksto analizės principai, akcentuojantys prasminį – giluminį – kūrinio lygmenį. 1990–2010 m. pastatyti Vilniaus, Kauno ir Klaipėdos įvaizdį formuojantys šiuolaikinės architektūros vietoženkliai čia traktuojami kaip vizualus tekstas. Remiantis sukurtuoju tyrimo modeliu buvo išsiaiškintos urbanistinėje aplinkoje vietoženklius išskiriančios fizinės, estetinės ir kitos funkcijos bei išskirtinės dislokacijos ypatybės. Architektų bei eilinių Vilniaus, Kauno ir Klaipėdos miestiečių sociologinė apklausa (buvo apklausti 363 skirtingo amžiaus ir išsilavinimo asmenys) leido sukurti miestų įvaizdžiui svarbių šiuolaikinės architektūros vietoženklių reikšmės atvaizdus, kuriuos sudaro 9 būdingos vertės. Matematinės statistikos priemonėmis atlikta šių verčių dažnio koreliacija su vietoženklius urbanistinėje aplinkoje išskiriančiomis ypatybėmis padėjo surasti patikimus... [toliau žr. visą tekstą] / In the global urban development, the construction of new architectural landmarks is one of the most prominent phenomena, which has been widely observed and analyzed in Western Europe for more than a decade. However, in urban areas, distinctive architectural objects are still often viewed through the prism of the half-century theories. The key innovation proposed in the thesis is a construction of the research model that helps to explore the meaning of the contemporary architectural landmarks important for the city image. The theoretical aspects in the dissertation are in line with the practical ones. Methodological basis there were the text analysis principles developed by R. Barthes, emphasizing meaningful level of an object. From 1990 to 2010, built in Vilnius, Kaunas and Klaipėda image forming modern architecture landmarks here treated as a visual text. The research model reveals aesthetic, physical and some other features of contemporary architectural landmarks (exceptional location, distinctive function), distinguishing them in the urban context. In total, 363 respondents of citizens and architects groups were interviewed in the sociological survey and it led to the creation of The images of meaning, which consist of nine key values. In order to process the data of the survey and to find the correlation between the contemporary architectural landmarks’ values frequency and aesthetic, physical, and the distinctive function, the exceptional location characteristics... [to full text]

Kernel methods have recently become popular in bioinformatics machine learning. Kernel methods allow linear algorithms to be applied to non-linear learning situations. By using kernels, non-linear learning problems can benefit from the statistical and runtime stability traditionally enjoyed by linear learning problems. However, traditional kernel learning frameworks use implicit feature spaces whose mathematical properties were hard to characterize. In order to address this problem, recent research has proposed a vector learning framework that uses landmark vectors which are unlabeled vectors belonging to the same distribution and the same input space as the training vectors. This thesis introduces an extension to the landmark vector learning framework that allows it to utilize two new classes of landmark vectors in the input space. This augmented learning framework is named the power landmark vector learning framework. A theoretical description of the power landmark vector learning framework is given along with proofs of new theoretical results. Experimental results show that the performance of the power landmark vector learning framework is comparable to traditional kernel learning frameworks.

Kernel methods have recently become popular in bioinformatics machine learning. Kernel methods allow linear algorithms to be applied to non-linear learning situations. By using kernels, non-linear learning problems can benefit from the statistical and runtime stability traditionally enjoyed by linear learning problems. However, traditional kernel learning frameworks use implicit feature spaces whose mathematical properties were hard to characterize. In order to address this problem, recent research has proposed a vector learning framework that uses landmark vectors which are unlabeled vectors belonging to the same distribution and the same input space as the training vectors. This thesis introduces an extension to the landmark vector learning framework that allows it to utilize two new classes of landmark vectors in the input space. This augmented learning framework is named the power landmark vector learning framework. A theoretical description of the power landmark vector learning framework is given along with proofs of new theoretical results. Experimental results show that the performance of the power landmark vector learning framework is comparable to traditional kernel learning frameworks.