To see the other types of publications on this topic, follow the link: Objects of a crime classification.
Dissertations / Theses on the topic 'Objects of a crime classification'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Objects of a crime classification.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Kahn, Karl. "Crime and equality, or crime and punishment? : population heterogeneity and fear of crime as determinants of redistribution preferences." Thesis, University of Oxford, 2014. http://ora.ox.ac.uk/objects/uuid:0cc102b1-c86d-4323-bfb0-3753c33baa33.
Full textAbstract:
Despite considerable research efforts, the relationship between inequality and demand for redistribution remains a highly contested topic within comparative political economy. This paper argues that a central yet widely overlooked mechanism linking macro-level income inequality to preferences for redistribution has to with the micro-level implications of certain externalities of inequality. Focusing on fear of crime, as one such externality, I argue that because (i) in- equality and crime are positively related, and (ii) because crime and fear of crime have a negative effect in individual utility, it follows that increasing in- equality should have a positive effect on support for redistribution. Importantly, however, the argument of this paper also recognises that redistribution is but one of several means through which a concern about crime can be addressed, with the most relevant alternatives being increased policing and harsher punitive measures. Drawing on literatures in criminology and political sociology, I theorise that a key determinant of this choice | between redistribution and policing/punishment as alternative approaches to dealing with crime | is the level of ethnic heterogeneity in the population. Taken together, therefore, this paper's argument implies that inequality will have differential effects on support for redistribution in different contexts: in cases where the population is homogenous, fear of crime - and by consequence inequality - will boost demand for redistribution, whilst no such effects will follow in contexts of high heterogeneity. Using a two-step statistical methodology, I analyse Eurobarometer and ESS data from 21 OECD countries and find persuasive empirical support for my theoretical expectations. Fear of crime is more strongly associated to support for redistribution when the level of population heterogeneity remains low, whilst the opposite holds true for the relationship between fear of crime and support for policing and punishment.
2
Csakany, Peter. "Classification of 3D objects from mesh surfaces." Thesis, Heriot-Watt University, 2001. http://hdl.handle.net/10399/463.
Full text
3
Feilzer, Martina Yvonne. "The 'crime scene' experiment : improving public knowledge through the provision of factual information on crime and criminal justice." Thesis, University of Oxford, 2007. http://ora.ox.ac.uk/objects/uuid:d8834e3d-996d-400a-8c71-8fb1f3bb6a2b.
Full textAbstract:
This thesis explores the relationship between crime, media, and public opinion of crime and criminal justice. It sets out current levels of knowledge and contemporary debates in each of the three academic disciplines devoted to these aspects of social life, and discusses how they relate to each other. It focuses on the capacity of information and public education to influence levels of public knowledge of and, as a secondary concern, attitudes towards, crime and criminal justice. The empirical research at the heart of the thesis employed a mixed methods research study drawing on quantitative – experimental research using a public opinion survey – as well as qualitative research methods – in-depth interviews and contextual data. The experimental research, the Oxford Public Opinion Survey and the publication of the Crime Scene column, was designed to measure the impact of providing factual information about crime and criminal justice to the public in a naturalistic way, i.e. by using a local newspaper column as the conveyor of such information. The key finding from the research was that readership of the column was low and that the column had no measurable impact on readers. Overall, the research findings suggest that interest in, take-up, and retention of factual information on crime and criminal justice is not as high as previous empirical research has suggested. The Crime Scene study has implications for sociological theories of crime and punishment which rely on simplistic orthodoxies concerning the media’s importance in influencing public opinion on crime and criminal justice and the related assumption that ‘the public’ is straightforwardly punitive.
4
Brandin, Martin, and Roger Hamrén. "Classification of Ground Objects Using Laser Radar Data." Thesis, Linköping University, Department of Electrical Engineering, 2003. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-1572.
Full textAbstract:
<p>Accurate 3D models of natural environments are important for many modelling and simulation applications, for both civilian and military purposes. When building 3D models from high resolution data acquired by an airborne laser scanner it is de-sirable to separate and classify the data to be able to process it further. For example, to build a polygon model of a building the samples belonging to the building must be found.</p><p>In this thesis we have developed, implemented (in IDL and ENVI), and evaluated algorithms for classification of buildings, vegetation, power lines, posts, and roads. The data is gridded and interpolated and a ground surface is estimated before the classification. For the building classification an object based approach was used unlike most classification algorithms which are pixel based. The building classifica-tion has been tested and compared with two existing classification algorithms. </p><p>The developed algorithm classified 99.6 % of the building pixels correctly, while the two other algorithms classified 92.2 % respective 80.5 % of the pixels correctly. The algorithms developed for the other classes were tested with thefollowing result (correctly classified pixels): vegetation, 98.8 %; power lines, 98.2 %; posts, 42.3 %; roads, 96.2 %.</p>
5
Kühne, Gerald. "Motion based segmentation and classification of video objects." [S.l. : s.n.], 2002. http://www.bsz-bw.de/cgi-bin/xvms.cgi?SWB10605031.
Full text
6
Edwards, Joseph Richard 1971. "Acoustic classification of buried objects with mobile sonar platforms." Thesis, Massachusetts Institute of Technology, 2006. http://hdl.handle.net/1721.1/37568.
Full textAbstract:
Thesis (Ph. D. in Ocean Engineering)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2006.<br>Includes bibliographical references (p. 229-237).<br>In this thesis, the use of highly mobile sonar platforms is investigated for the purpose of acoustically classifying compact objects on or below the seabed. The extension of existing strategies, including synthetic aperture sonar and conventional imaging, are explored within the context of the buried object problem. In particular, the need to employ low frequencies for seabed penetration is shown to have a significant impact both due to the relative length of the characteristic scattering mechanisms and due to the interface effects on the target scattering. New sonar strategies are also shown that exploit incoherent wide apertures that are created by multiple sonar platforms. For example, target shape can be inverted by mapping the scattered field from the target with a team of receiver vehicles. A single sonar-adaptive sonar platform is shown to have the ability to perform hunting and classification tasks more efficiently than its pre-programmed counterpart. While the monostatic sonar platform is often dominated by the source component, the bistatic or passive receiver platform behavior is controlled by the target response. The sonar-adaptive platform trajectory, however, can result in the platform finishing its classification effort out of position to complete further tasks.<br>(cont.) Within the context of a larger mission, the use of predetermined adaptive behaviors is shown to provide improved detection and classification performance while minimizing the risk to the overall mission. Finally, it is shown that multiple sonar-adaptive platforms can be used to create new sonar strategies for hunting and classifying objects by shape and content. The ability to sample the scattered field from the target across a wide variety of positions allows an analysis of the aspect-dependent behavior of the target. The aspect-dependence of the specular returns indicate the shape of the target, while the secondary returns from an elastic target are also strongly aspect-dependent. These features are exploited for improved classification performance in the buried object hunting mission.<br>by Joseph R. Edwards.<br>Ph.D.in Ocean Engineering
7
Schwanebeck, Wieland. "Beatrix Hesse. The English Crime Play in the Twentieth Century. Crime Files: Review." De Gruyter, 2017. https://tud.qucosa.de/id/qucosa%3A71155.
Full textAbstract:
In a 2014 episode of Toast of London, a highly successful Channel 4 sitcom, protagonist Steven Toast, a dim-witted London-based actor, joins the cast of legendary stage-play The Moose Trap, which is, of course, modelled on Agatha Christie’s successful West End whodunit, The Mousetrap. When Toast accidentally spoils the play’s ending during an interview, the play soon finds itself without an audience, which leads to some of the geriatric cast members literally dying on stage.
8
Jacobsen, Dawn. "Classification of Toolmark Surfaces on Zipper Teeth." Thesis, University of North Texas, 1999. https://digital.library.unt.edu/ark:/67531/metadc2248/.
Full textAbstract:
This study proposes the classification of the toolmark under the heads of zipper teeth as a subclass characteristic as outlined by the Association of Firearm and Toolmark Examiners (AFTE). Two separate cases in which zipper teeth were found at crime scenes prompted this study. Brass zipper teeth manufactured by YKK were taken from 20 pairs of jeans and studied using a Reichert comparison microscope at 4X power. Photographs were taken and over 750 comparisons made. It was found that the toolmarks on each side on the 20 zippers were unique and independent of all other sides. The observations made in this study indicate that classifying zipper teeth toolmarks as a subclass characteristic is valid.
9
Cichocki, Radoslaw. "Classification of objects in images based on various object representations." Thesis, Blekinge Tekniska Högskola, Avdelningen för programvarusystem, 2006. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-5774.
Full textAbstract:
Object recognition is a hugely researched domain that employs methods derived from mathematics, physics and biology. This thesis combines the approaches for object classification that base on two features – color and shape. Color is represented by color histograms and shape by skeletal graphs. Four hybrids are proposed which combine those approaches in different manners and the hybrids are then tested to find out which of them gives best results.<br>Mail the author at radoslaw.cichocki(at)gmail.com
10
LIU, HUI. "Classification and prediction for flying objects based on behavior model." Thesis, Mittuniversitetet, Avdelningen för elektronikkonstruktion, 2014. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-22207.
Full textAbstract:
This master thesis relates to machine vision system and object classification. The aim of this paper is to classify the flying objects in images in a computer vision system, for example, an eagle, kite or airplane. In this thesis, large amounts of data will be analyzed and a behavior model will be built for each object as important steps towards improving and automating the object classification system. The application of this thesis is to reduce the deaths of golden and bald eagles due to wind blades. In this thesis work, a new effective method is presented, namely, a stereo vision system, which is applied in feature selection based on this object classification. Several features are primarily extracted, including the flying height, speed, size and degree of changes in the object parameters. For image processing and feature extraction, the video acquisition is the first and essential step. Due to the limitation both of equipment and location, the captured videos still do not allow for the collection of sufficient data. For the classification of two objects, a Support Vector Machine (SVM) and Library for Support Vector Machine (LIBSVM) have been employed and implemented in MATLAB. In addition, a preliminary study in relation to the idea of multi-class classification has been conceived and tested by means of an experiment. In relation to building a behavior model, the various feature properties and characteristics were beneficial with regards to developing the accuracy and robustness of the final classification and recognition results. The results gathered from these two methods in terms of SVM and LIBSVM are compared and analyzed in order to identify their differences and to determine a better solution. Additionally, the possible future work for this project will be discussed. Results show that 98% of the flying objects can be currently classified by using OVO SVMs and the OVR SVMs. Based on the results of the classification, 85.82% of the flying objects could be predicted correctly. Key words: machine vision system, object classification, behavior model, stereo vision system, image processing, feature extraction, SVM, LIBSVM, MATLAB
11
FONSECA, LUIS MARCELO VITAL ABREU. "CLASSIFICATION OF OBJECTS IN REAL CONTEXT BY CONVOLUTIONAL NEURAL NETWORKS." PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO, 2017. http://www.maxwell.vrac.puc-rio.br/Busca_etds.php?strSecao=resultado&nrSeq=30251@1.
Full textAbstract:
PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO<br>COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR<br>PROGRAMA DE EXCELENCIA ACADEMICA<br>A classificação de imagens em contexto real é o ápice tecnológico do reconhecimento de objetos. Esse tipo de classificação é complexo, contendo diversos problemas de visão computacional em abundância. Este projeto
propõe solucionar esse tipo de classificação através do uso do conhecimento no aprendizado de máquina aplicado ao dataset do MS COCO. O algoritmo implementado neste projeto consiste de um modelo de Rede Neural Convolutiva que consegue aprender características dos objetos e realizar predições sobre suas classes. São elaborados alguns experimentos que comparam diferentes resultados de predições a partir de diferentes técnicas de aprendizado. É também realizada uma comparação dos resultados da implementação com o estado da arte na segmentação de objetos em contexto.<br>The classification of objects in real contexts is the technological apex of object recognition. This type of classification is complex, containing diverse computer vision problems in abundance. This project proposes
to solve that type of classification through the use of machine learning knowledge applied to the MS COCO dataset. The implemented algorithm in this project consists of a Convolutional Neural Network model that
is able to learn characteristics of the objects and predict their classes. Some experiments are made that compare different results of predictions using different techniques of learning. There is also a comparison of the results from the implementation with state of art in contextual objects segmentation.
12
Davies, Matthew William. "Elected Police and Crime Commissioners : an experiment in democratic policing." Thesis, University of Oxford, 2015. http://ora.ox.ac.uk/objects/uuid:72bf870f-4ce8-4cf6-9e5c-5564d4273100.
Full textAbstract:
In this thesis, I explore the ways in which Police and Crime Commissioners (PCCs) have met a declared policy intention to create greater democratic accountability around policing and crime. I conceptualise PCCs as a piece of a broader democratic puzzle and explore both how they have been positioned and shaped within the broader policing and crime nexus across England and Wales. In considering the positioning of PCCs, I use data from case studies and interviews with 32 (out of 41) PCCs to identify how they have begun to develop relationships with the public and local, regional and national partners. The findings suggest that with the exception of their abilities to join up local crime reduction services, PCCs occupy an awkward space - not local enough to be meaningfully representative of the public they serve, but not outwardly-facing enough to manage wider co-ordination of policing. Subsequently, I investigate the shape of the PCC model to deliver greater accountability by focusing on the ways in which PCCs have begun to envisage the role and develop relationships with other key stakeholders. Varied responses from PCCs across the country reflected the broad-ranging nature of the role, which in some cases appeared to undermine their ability to fully perform all aspects of the job. I argue that this became particularly accentuated in emerging relationships with chief constables and Police and Crime Panels, where the single PCC model exposes accountability to dangers of personalities and politics. I conclude by arguing that while many PCCs have facilitated various components of democratic accountability within the management of policing and crime-reduction services, the PCC model appears to be misplaced and misshaped to effectively complete the puzzle of democratic policing.
13
Zarate, Tenorio Barbara Astrid. "Social policy, protest participation and violent crime in Latin America." Thesis, University of Oxford, 2015. https://ora.ox.ac.uk/objects/uuid:5dcad1e6-ef54-4fba-a6b4-38d68cf7d0c7.
Full textAbstract:
This thesis consists of four self-contained articles which focus on different aspects of citizens' demand for and governments' supply of social policy in Latin American democracies. The underlying questions that link the four papers are a) do social and economic grievances affect citizens' propensity to protest? and, b) do democratic governments in the region use social policy as an instrument to mitigate social discontent and violent crime? In the first two papers, I use public opinion data in order to examine the determinants of citizens' participation in protest with a special focus on dissatisfaction with the quality of public services, demands for inequality-reduction policies and economic deprivation. The results show that among other factors, protest participation is motivated by citizens' discontent over the quality of basic social services, support for redistributive policies, and relative economic deprivation. The third and fourth papers analyze the "supply side" of social spending in the region focusing on collective protest and violent crime, respectively. The third article argues that under democracy, organized labor is in a better position relative to other groups in society to obtain social policy concessions as a consequence of their collective action efforts. The results show that whereas social security spending increases as a consequence of labor militancy, cutbacks in human capital spending are less likely as peaceful large-scale demonstrations increase. The fourth paper argues that political leaders use education spending as an instrument to mitigate violent crime. It also argues that the effect of violent crime on education spending is larger when leftist governments are in power. The empirical analysis provides support for these arguments.
14
Kreulich, Klaus. "XML Formatting Objects - medienunabhängige Dokumentaufbereitung." Universitätsbibliothek Chemnitz, 2003. http://nbn-resolving.de/urn:nbn:de:swb:ch1-200300936.
Full text
15
Nordenmark, Victor, and Adam Forsgren. "Radar-detection based classification of moving objects using machine learning methods." Thesis, KTH, Maskinkonstruktion (Inst.), 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-180882.
Full textAbstract:
In this MSc thesis, the possibility to classify moving objects based on radar detection data is investigated. The intention is a light-weight, low-level system that relies on cheap hardware and calculations of low complexity. Scania, the company that has commissioned this project, is interested in the usage potential of such a system in autonomous vehicle applications. Specifically, the class information is desired in order to enhance the moving object tracker, a subsystem that represents a crucial skillset of an autonomously driving truck. Objects are classified as belonging to one of four classes: Pedestrian, bicyclist, personal vehicle and truck. The major system input consists of sensor data from a set off our short-range mono-pulse Doppler radars operating at 77 GHz. Using a set of training and validation data gathered and labeled within this project, a classification system based on the machine learning method of Support vector machines is created. Several other supporting software structures are also created and evaluated. In the validation phase, the system is shown to discern well between the four classes. System simulations performed on logged radar data show promising performance also in situations not reflected within the labeled dataset.To further investigate the feasibility of the system, it has been implemented and tested on the prototype test vehicle Astator, and performance has been evaluated with regards to both real-time constraints and classification accuracy. Overall, the system shows promise in the scenarios for which it was intended, both with respect to real-time and classification performance. In more complex scenarios however, sensor noise is increasingly apparent and affects the system performance in a negative way. The noise is extra apparent in heavy traffic and high velocity scenarios.<br>I detta examensarbete undersöks möjligheten att klassificera rörliga objekt baserat på data från Dopplerradardetektioner. Slutmålet är ett system som använder billig hårdvara och utför beräkningar av låg komplexitet. Scania, företaget som har beställt detta projekt, är intresserat av användningspotentialen för ett sådant system i applikationer för autonoma fordon. Specifikt vill Scania använda klassinformationen för att lättare kunna följa rörliga objekt, vilket är en väsentlig färdighet för en autonomt körande lastbil. Objekten delas in i fyra klasser: fotgängare, cyklist, bil och lastbil. Indatan till systemet består väsentligen av en plattform med fyra stycken monopulsdopplerradars som arbetar med en vågfrekvens på 77 GHz. Ett klassificeringssystem baserat på maskininlärningskonceptet Support vector machines har skapats. Detta system har tränats och validerats på ett dataset som insamlats för projektet, innehållandes datapunkter med klassetiketter. Ett antal stödfunktioner till detta system har också skapats och testats. Klassificeraren visas kunna skilja väl på de fyra klasserna i valideringssteget. Simuleringar av det kompletta systemet gjort på inspelade loggar med radardata visar lovande resultat, även i situationer som inte finns representerade i träningsdatan. För att vidare undersöka systemet har det implementerats och testats på prototyplastbilen Astator, och prestandan har utvärderats utifrån både realtidstidsperpsektiv och klassificeringsnoggranhet. Generellt uppvisar systemet lovande resultat i scenarier som liknar slutanvändningsområdet. I mer komplexa trafiksituationer och då lastbilen färdas i högre hastigheter leder dock en högre förekomst av sensorbrus till att systemets prestanda försämras.
16
Walters, Mark Austin. "Hate crimes hurt more : can restorative practices help repair the harms?" Thesis, University of Oxford, 2012. http://ora.ox.ac.uk/objects/uuid:5b28fa3f-38b0-4af4-ad88-7922e1d834b1.
Full textAbstract:
The current retributive approach to tackling hate crime, while intuitively grounded in the principle of proportionately, does little to either repair the harms caused by incidents of hate or engender greater levels of acceptance of those deemed as “different”. This thesis therefore explores whether restorative justice, a relatively new theory and practice of criminal justice, is better placed to tackle the causes and consequences of hate victimisation. The 18 month empirical study, carried out to examine the thesis’ aims, uses a triangulation approach by incorporating observations of restorative justice meetings, semi-structured interviews with victim participants and semi-structured interviews with restorative practitioners who have experience facilitating hate crime cases. The mainly qualitative data collated provides for a detailed evaluation of the various processes found within restorative practices that: 1) helped to alleviate the distress caused by hate victimisation and 2) prevented the recurrence of hate-motivated incidents. A broad conceptualisation of hate crime was used within the thesis that included “hate incidents”. This allowed me to explore the utility of restorative practices in cases involving serious violence and the more pervasive “low-level”, but nonetheless highly deleterious, non-criminal incidents of hate that are frequently committed against minority group individuals. There were also several unanticipated findings from the study. First, data emerged which highlighted various aspects of the restorative practice which were unforeseen as being central to the successful application of restorative processes, these are discussed throughout the thesis. Second, great insight was gained into the nature of hate victimisation, helping to unravel some of the complex socio-cultural factors pivotal to both the cause and effect of hate victimisation. It is hoped that these additional findings provide important epistemological advancements in both fields of study.
17
Saunders-Hastings, Katherine E. "Order and insecurity under the mara : violence, coping, and community in Guatemala City." Thesis, University of Oxford, 2015. http://ora.ox.ac.uk/objects/uuid:41c30581-4e46-483a-b77e-c241fa88a819.
Full textAbstract:
Drawing on sixteen months of ethnographic fieldwork in a poor and notorious neighbourhood, this dissertation examines how evolving dynamics of urban violence have affected life in a Guatemala City gang territory. The maras of Guatemala, El Salvador, and Honduras - the gangs that help give these countries some of the world's highest homicide rates - have changed dramatically in their group cultures and criminal economies since they appeared in the 1990s. I trace what I call the mara's predatory turn: the elaboration of an extortion economy, which has had far-reaching consequences for the relationship between gang cliques and their barrios. This transformation has re-shaped the experience of chronic insecurity in the communities that maras operate from: inhabitants report that it is now less manageable, less predictable, and more frightening. They speak of a heightening of danger in their lives brought about by the decline of certain local norms and mechanisms that had previously moderated gang violence and bolstered community resilience. Local narratives of insecurity and decline illuminate when, how, and why violence disrupts and disorders social life. What many informants emphasized was not a cataclysmic appearance of violence in their lives, but rather a catastrophic breakdown in the mechanisms that had controlled it. In this distressing context, residents struggle to minimize their insecurity and to reclaim or create forms of order. I examine two principal ways that they seek to do so: by working to maintain a moral order based on narratives about the neighbourhood and its values or 'codes', and by looking to external providers of order in the state and its security forces. Exploring the complex relationships and interactions between inhabitants, gang members, and state forces in this barrio, I contribute to academic debates about local and state responses to insecurity in Latin America and propose modifications to prevailing models of state and criminal 'governance' in marginal urban communities.
18
Heeb, Alexis. "Violent crime, public perceptions and citizen security strategies in Colombia during the 1990s." Thesis, University of Oxford, 2002. http://ora.ox.ac.uk/objects/uuid:9aa285b1-15e6-402f-a4e8-2f7322c9adc3.
Full textAbstract:
This research deals with the topic of violent crime in urban Colombia. Although many references are made to the conflict between the State, guerrilla groups and paramilitary organisations in rural areas, and to the problem of drug-cartels and illegal-drug production, the main aim of the thesis is to show recent trends in violent crime and discuss citizen security strategies followed during the decade of the 1990s. Chapter 1 focuses on urban homicides. In Colombia, 40 percent of the 25,000 annual homicides are committed in the ten largest cities. The cities of Medellin, Bogota and Cali account for almost 30 percent of this total. Although the victims are mainly young men from the poorest socio-economic levels, homicides are not necessarily correlated to the areas where the poor live. The probability of getting involved in a homicide, either as a victim or as victimiser, is significantly higher in places where access to economic resources is greater. Chapter 2 analyses the problem of kidnappings. The chapter looks at recent progresses in Colombian anti-kidnapping legislation and focuses on the authors, the victims and the riskzones where most cases take place. Although these crimes affect mainly the rich and the middle class, kidnappers have recently started to target victims from all social backgrounds. This strategy creates fear among citizens and permits kidnappers to extort more fees from people who could be at risk of being kidnapped. Chapter 3 looks at the issue of perception and fear of crime. This question provides a better understanding of the concept of risk and the subjectivity of decision-making when facing insecurity. If citizen security strategies have had little impact during the last decade, it has been partly because of poor levels of co-operation and communication with the population. The consequence of this has been an increasing perception of insecurity and distrust among citizens. Chapter 4 assesses citizen security strategies followed during the administrations of Presidents Gaviria (1990-1994), Samper (1994-1998) and Pastrana (1998-2002). The reforms implemented since the adoption of a new Constitution in 1991 have had important impacts on security strategies as they have given more discretionary powers to civilian authorities, especially at the municipal level. Larger cities like Bogotá, Cali and Medellin, implemented a series of successful programmes that have reduced the levels of violent crime, notably homicides. Other crimes, like kidnappings, have not decreased since the government has lacked a coherent strategy to combine its peace negotiations with insurgent organisations with its legitimate right to fight violent crime.
19
Behzad, Nima. "3D Context of Objects : A prior for Object Detection and Place Classification." Thesis, Örebro universitet, Institutionen för naturvetenskap och teknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-92522.
Full textAbstract:
Contextual information is helpful for object detection and object-based placerepresentation. 3D data significantly helps to capture geometrical informationabout scenes. In this work, a feature descriptor for object context in full 3Dpointclouds of places is introduced together with a method to extract featuresand build the context model.The proposed model is evaluated in experiments on pointclouds from differenttypes of places which include different object categories. Results showthe promising ability of the model to predict the possible context of objects inpointclouds or complete 3D maps of an environment.Among various applications for this, the author suggests object contextmodels to be used in place categorization and semantic mapping and discussesa method for it. To the knowledge of the author, this work is unique regardingits use of full 3D pointcloud of scenes and also introducing this descriptor tobe used to represent places.
20
Vu, Trung-Dung. "Vehicle perception : localization, mapping with detection, classification and tracking of moving objects." Phd thesis, Grenoble INPG, 2009. http://www.theses.fr/2009INPG0183.
Full textAbstract:
La perception de l'environnement dynamique d'un véhicule est une étape très importante dans la construction d'un système d'assistance à la conduite automobile. Le travail de thèse aborde les problèmes associés de localisation, de cartographie (appelé « mapping »), de détection, de classification, et de suivi des objets mobiles perçus à l'aide des capteurs embarqués. Les modèles et algorithmes développés se concentrent sur l'utilisation de capteurs télémétriques à balayage laser. Dans un premier temps, nous proposons une méthode basée sur une grille d'occupation pour résoudre le problème de « mapping » avec détection des objets mobiles. Afin de corriger les données odométriques du véhicule qui sont toujours entachées d'erreurs, une nouvelle méthode rapide de « scan matching » est introduite, apte à fonctionner de manière fiable dans des environnements dynamiques. Cela permet de localiser correctement le véhicule, et ainsi de mettre à jour de manière incrémentale la carte courante et de détecter les objets mobiles. La deuxième contribution est une méthode simultanée de détection, classification et suivi des objets mobiles. Une approche fondée sur les modèles d'objets est introduite permettant d'interpréter la séquence de données laser par des hypothèses de trajectoires d'objets mobiles. La technique de « Data-Driven Markov chain Monte Carlo » (DDMCMC) est utilisée pour résoudre de manière efficace le problème d'association des données dans un espace « spatio-temporel » et fournir la solution la plus probable. La méthode a été validée sur des données réelles avec des résultats très prometteur. Les travaux sont valorisés à travers une collaboration industrielle avec Daimler AG dans le cadre du projet européen PREVENT. Le module de perception a été intégré sur un véhicule réel dans une application automobile particulière, appelée le « Pre-Crash ». Celle-ci a donné lieu à des démonstrations lors des journées de présentation des résultats du projet<br>Perceiving or understanding the environment surrounding of a vehicle is a very important step in building driving assistant systems or autonomous vehicles. In this thesis, we study problems of simultaneous localization and mapping (SLAM) with detection, classification and tracking moving objects in context of dynamic outdoor environments focusing on using laser scanner as a main perception sensor. It is believed that if one is able to accomplish these tasks reliably in real time, this will open a vast range of potential automotive applications. The first contribution of this research is made by a grid-based approach to solve both problems of SLAM with detection of moving objects. To correct vehicle location from odometry we introduce a new fast incremental scan matching method that works reliably in dynamic outdoor environments. After good vehicle location is estimated, the surrounding map is updated incrementally and moving objects are detected without a priori knowledge of the targets. Experimental results on datasets collected from different scenarios demonstrate the efficiency of the method. The second contribution follows the first result after a good vehicle localization and a reliable map are obtained. We now focus on moving objects and present a method of simultaneous detection, classification and tracking moving objects. A model-based approach is introduced to interpret the laser measurement sequence over a sliding window of time by hypotheses of moving object trajectories. The data-driven Markov chain Monte Carlo (DDMCMC) technique is used to solve the data association in the spatio-temporal space to effectively find the most likely solution. We test the proposed algorithm on real-life data of urban traffic and present promising results. The third contribution is an integration of our perception module on a real vehicle for a particular safety automotive application, named Pre-Crash. This work has been performed in the framework of the European Project PReVENT-ProFusion in collaboration with Daimler AG. A comprehensive experimental evaluation based on relevant crash and non-crash scenarios is presented which confirms the robustness and reliability of our proposed method
21
Wong, Rebecca W. Y. "The organisation of the illegal tiger parts trade in China." Thesis, University of Oxford, 2013. http://ora.ox.ac.uk/objects/uuid:2a373846-393c-47ab-b297-8e389d906b49.
Full textAbstract:
The thesis is a study of how Chinese illegal tiger parts trading networks are organized. In particular, this thesis tests in a qualitative manner the causal relationship between three independent variables and the network organizations of these markets. The three independent variables are “ethnicity”, “level of enforcement” and “proximity to the source country”. The thesis also discusses the dynamics of the illegal transactions of tiger parts products. Legitimate meditators or dispute resolutions mechanisms are lacking in the underworld so the risks, which the parties undertake during trading, are far higher. This thesis explores how illegal transactions are enforced, carried out and honored in this trade. In order to map the organization of the tiger trade, I conducted fieldwork in three trading hubs across China: Lhasa. Kunming and Xining. I discovered five tiger parts trading networks, three of which specialized in the trading of tiger skins and two in tiger bones. Within these networks, the level of perceived but not the actual level of risk influences the decisions of the actors in the network. Entry into the network is easy when the perceived level of enforcement is low. In these settings, there is no ethnic restriction for entering the network; the supplier is willing to trade with anyone with a trustworthy reputation. On the other hand, accessibility to the network is strictly controlled when actors perceive a high level of enforcement in their operating environment. Under this setting, the organization of the network becomes more exclusive and ethnically homogenous, as shown in the Tibetan tiger skin-trading network in Lhasa and the tiger bone-trading network in Kunming. The proximity of the tiger source country to the re-distribution sites (fieldwork cities) also influences the organization of the networks. When the level of enforcement is low and the tiger source country is far away from the re-distribution sites, a monetary deposit is required in order to show that the buyer is serious about his/her request, as shown by the tiger skin-trading network in Kunming.
22
Tuchscheerer, Sven. "Human Factors in Automotive Crime and Security." Doctoral thesis, Universitätsbibliothek Chemnitz, 2011. http://nbn-resolving.de/urn:nbn:de:bsz:ch1-qucosa-70586.
Full textAbstract:
Als vor 120 Jahren, am 5. August 1888, Bertha Benz eine 106 Kilometer lange Fahrt mit dem Prototypenfahrzeug ihres Mannes Carl Benz unternahm legte sie den Grundstein für die erste Überlandfahrt mit einem Benzinautomobil überhaupt. Damit trug sie wesentlich zum Abbau von Vorurteilen in der Gesellschaft gegenüber Benzinfahrzeugen bei und verhalf der Firma ihres Mannes zum Erfolg. Gleichzeitig beging sie auch den ersten dokumentierten Fahrzeugdiebstahl, da die Nutzung ohne das Wissen ihres Mannes erfolgte. Strafrechtliche Sanktionen musste Frau Benz nicht fürchten, schließlich war dieser Fall durchaus von Vorteil für die Vermarktung. Das erkannte auch Herr Benz und verzichtete auf eine Anzeige.
Am 23.12.2008 wurde ein PKW Toyota Landcruiser gegen 01:20 Uhr an der Landstrasse L46 zwischen Herforst und Rothaus gefunden. Der Täter, welcher das Fahrzeug am 17.12.2008 bei einem Bauunternehmer stahl, stellte es – vermutlich auf Grund von Treibstoffmangel – dort ab. Ihm drohen bis zu drei Jahren Haft und / oder eine Geldstrafe.
Am 24.02.2008 schrieb eine Benutzerin namens Kitty in einem Internetforum zum Thema Erfahrungen mit Autodiebstahl: „…mir hat man nur bei meinem jetzigen Fahrzeug schon die Nummernschilder geklaut, was mich dann 80 Euro gekostet hat.“ ("AW: Auto geklaut", 2008).
Diese drei Beispiele repräsentieren ansatzweise, wie vielfältig die Fälle sind, welche gemeinhin unter dem Begriff Fahrzeugkriminalität gefasst werden. Sie unterscheiden sich hinsichtlich der Motive der Täter, dem Besitzer zugefügten Schaden oder den strafrechtlichen Konsequenzen. Zusätzlich zu dieser Vielfalt ist die Fahrzeugkriminalität dem Einfluss der Zeit unterworfen, in Bezug auf die Anzahl im Verkehr befindlicher Fahrzeuge, deren Nutzung, deren Diebstahlschutzvorrichtung oder Vorgehen der Täter. Somit ist das Konstrukt Fahrzeugkriminalität durch einen hohen Komplexitätsgrad gekennzeichnet.
Aus dieser Komplexität heraus begründet sich das fundamentale Ziel der vorliegenden Arbeit, eine detaillierte und systematische Beschreibung von Fahrzeugkriminalität zu geben. Darauf aufbauend werden Ursachen- und Wirkungszusammenhänge identifiziert, beschrieben und in einem numerischen Relativ abgebildet. Dabei wird ein ganzheitlicher Ansatz verfolgt, in Abgrenzung zu bisherigen wissenschaftlichen Arbeiten, welche einen Erkenntnisgewinn zur Beschreibung, Erklärung und Vorhersage von Fahrzeugkriminalität für einen mehr oder weniger eng umschriebenen Teilaspekt leisten. In diesem Ansatz werden gezielt psychologische Aspekte in den Fokus gestellt und mit gesellschaftlichen, technischen, legislativen, judikativen und exekutiven Rahmenfaktoren in Beziehung gesetzt. Damit kann der Arbeit ein wissenschaftlicher Erkenntnisgewinn im Sinne einer ganzheitlichen Analyse beigemessen werden, in den Ansätze zur Beschreibung und Erklärung separater Aspekte integriert und bewertet werden können.
Eine – damit eng verknüpfte – Voraussetzung ist die Entwicklung und Evaluation von Untersuchungsmethoden um die relevanten Aspekte in den beteiligten Personengruppen so detailliert und umfassend wie zur Erklärung nötig sowie so ökonomisch wie möglich zu erfassen.
Im Zentrum der Betrachtung steht die Interaktion zwischen den beiden hauptsächlich beteiligten Personengruppen: Fahrzeugbesitzer und Täter. Betrachtete Aspekte in diesen Gruppen sind neben Motiven, Risiko- und Gewinneinschätzungen auch Faktoren wie finanzieller Konnex oder Ausstattung mit Sicherheitstechnik, bzw. Werkzeug zu dessen Überwindung, Furcht vor Fahrzeugkriminalität oder finanzielle und zeitliche Schäden durch Fahrzeugkriminalität.
23
Dumortier, Alexis Jean Louis. "Detection, classification and localization of seabed objects with a virtual time reversal mirror." Thesis, Massachusetts Institute of Technology, 2009. http://hdl.handle.net/1721.1/55316.
Full textAbstract:
Thesis (S.M.)--Joint Program in Oceanography/Applied Ocean Science and Engineering (Massachusetts Institute of Technology, Dept. of Mechanical Engineering; and the Woods Hole Oceanographic Institution), 2009.<br>Includes bibliographical references (p. 88-91).<br>The work presented in this thesis addresses the problem of the detection, classification and localization of seabed objects in shallow water environments using a time reversal approach in a bistatic configuration. The waveguide is insonified at low frequency ('kHz) with an omnidirectional source and the resulting scattered field is sampled by a receiving array towed behind an Autonomous Underwater Vehicle (AUV). The recorded signals are then processed to simulate onboard the AUV, the time reversed transmissions which serve to localize the origin of the scattered field on the seabed and estimate the position of the targets present. The clutter rejection based upon the analysis of the singular values of the Time Reversal operator is investigated with simulated data and field measurements collected off the coast of Palmaria (Italy) in January 2008.<br>by Alexis J. Dumortier.<br>S.M.
24
Fehlman, William L. "Classification of non-heat generating outdoor objects in thermal scenes for autonomous robots." W&M ScholarWorks, 2008. https://scholarworks.wm.edu/etd/1539623338.
Full textAbstract:
We have designed and implemented a physics-based adaptive Bayesian pattern classification model that uses a passive thermal infrared imaging system to automatically characterize non-heat generating objects in unstructured outdoor environments for mobile robots. In the context of this research, non-heat generating objects are defined as objects that are not a source for their own emission of thermal energy, and so exclude people, animals, vehicles, etc. The resulting classification model complements an autonomous bot's situational awareness by providing the ability to classify smaller structures commonly found in the immediate operational environment. Since GPS depends on the availability of satellites and onboard terrain maps which are often unable to include enough detail for smaller structures found in an operational environment, bots will require the ability to make decisions such as "go through the hedges" or "go around the brick wall." A thermal infrared imaging modality mounted on a small mobile bot is a favorable choice for receiving enough detailed information to automatically interpret objects at close ranges while unobtrusively traveling alongside pedestrians. The classification of indoor objects and heat generating objects in thermal scenes is a solved problem. A missing and essential piece in the literature has been research involving the automatic characterization of non-heat generating objects in outdoor environments using a thermal infrared imaging modality for mobile bots. Seeking to classify non-heat generating objects in outdoor environments using a thermal infrared imaging system is a complex problem due to the variation of radiance emitted from the objects as a result of the diurnal cycle of solar energy. The model that we present will allow bots to "see beyond vision" to autonomously assess the physical nature of the surrounding structures for making decisions without the need for an interpretation by humans.;Our approach is an application of Bayesian statistical pattern classification where learning involves labeled classes of data (supervised classification), assumes no formal structure regarding the density of the data in the classes (nonparametric density estimation), and makes direct use of prior knowledge regarding an object class's existence in a bot's immediate area of operation when making decisions regarding class assignments for unknown objects. We have used a mobile bot to systematically capture thermal infrared imagery for two categories of non-heat generating objects (extended and compact) in several different geographic locations. The extended objects consist of objects that extend beyond the thermal camera's field of view, such as brick walls, hedges, picket fences, and wood walls. The compact objects consist of objects that are within the thermal camera's field of view, such as steel poles and trees. We used these large representative data sets to explore the behavior of thermal-physical features generated from the signals emitted by the classes of objects and design our Adaptive Bayesian Classification Model. We demonstrate that our novel classification model not only displays exceptional performance in characterizing non-heat generating outdoor objects in thermal scenes but it also outperforms the traditional KNN and Parzen classifiers.
25
Salazar, Rafael Perera. "Predictive classification using mixtures of normal distributions." Thesis, University of Oxford, 1998. http://ora.ox.ac.uk/objects/uuid:b5c4da17-ad7f-4950-ba3e-7bbecfdcb1d8.
Full textAbstract:
Classification using mixture distributions to model each class has not received too much attention in the literature. The most important attempts use normal distributions as com- ponents in these mixtures. Recently developed methods have allowed the use of these kinds of models as a flexible approach for density estimation. Most of the methods de- veloped so far use plug-in estimates for the parameters and assume that the number of components in the mixture is known. We obtain a predictive classifier for the classes by using Markov Chain Monte Carlo techniques which allow us to obtain a sampling chain for the parameters. This fully Bayesian approach to classification has the advantage that the number of components for each class is taken as another variable parameter and integrated out of the classification. To achieve this we use a birth-and-death/Gibbs sampler algorithm developed by Stephens (1997). We use five different datasets, two simulated ones to test the methods on a single class and three real datasets to test the methods for classification. We look at different models to de- fine which gives better flexibility in the modelling and an overall better classification. We look at different types of priors for the means and dispersion matrices of the components. Joint conjugate priors and an independent conjugate priors for the means and dispersion matrices for the components are used. We use a model with a common dispersion matrix for all the components and another one with a reparametrisation of these dispersion ma- trices into size, shape and orientation (Banfield and Raftery (1993)). We allow the sizes to differ while keeping a common shape and orientation for the dispersion matrices of the components in a class. We found that this type of modelling with independent conjugate priors for the means and dispersions while allowing the sizes of the dispersions to vary gave the best results for classification purposes as it allowed great flexibility and separation between the compo- nents of the classes.
26
Kravchenko, Andrey. "BERyL : unified approach to web block classification." Thesis, University of Oxford, 2015. https://ora.ox.ac.uk/objects/uuid:f885aa3b-a590-4acf-a01b-5af3bae93cdf.
Full textAbstract:
Web blocks such as navigation menus, advertisements, and headers and footers are key components of web pages which define not only the appearance of a web page but also the way in which humans interact with different parts of the page. For machines, however, classifying and interacting with these blocks is a surprisingly hard task. Yet, web block classification has varied applications in the fields of wrapper induction, assistance for visually impaired people, mobile web browsing, web page topic clustering and web searching. Our system for web block classification, BERyL, performs the automated classification of web blocks through a combination of machine learning and declarative, model-driven feature extraction based on Datalog rules. BERyL uses refined feature sets for the classification of individual blocks to achieve accurate classification for all of the block types that we have so far observed. The high accuracy is achieved through these carefully selected features. Some are even tuned to the specific block type. At the same time, BERyL avoids the high cost of feature engineering through a model-driven rather than programmatic approach to feature extraction. Not only does this reduce the time for feature engineering, the model-driven, declarative approach also allows for semi-automatic optimisation of the feature extraction system. BERyL also employs a holistic approach to web block classification where individual blocks are considered within the context of a web page powered by the knowledge representation rules specific to that domain. We validate these claims for a broad range of web blocks in an extensive evaluation.
27
Simancik, Frantisek. "Consequence-based reasoning for ontology classification." Thesis, University of Oxford, 2013. http://ora.ox.ac.uk/objects/uuid:5e16d4ca-d80b-413b-9701-28febcf15613.
Full textAbstract:
Description logics (DLs) are knowledge representation languages that provide the theoretical underpinning for modern ontology languages such as OWL and serve as the basis for the development of ontology reasoners and tools. Most modern ontology reasoners are based on optimized tableau algorithms, which perform reasoning by trying to build counter-models. More recently, another kind of reasoning algorithms has been introduced that, instead of building counter-models, directly derive logical consequences of axioms in the ontology using inference rules. Such consequence-based algorithms were first introduced for the EL family of DLs, and later extended to more expressive Horn DLs. However, up until now, consequence-based algorithms could not handle non-Horn features such as disjunctions. We consider several complementary aspects of consequence-based reasoning in this thesis. Firstly, we describe the parallelized consequence-based reasoner ELK, which is currently the fastest reasoner for EL ontologies. Secondly, we demonstrate how consequence-based algorithms can be extended to handle disjunctions using inference rules reminiscent of ordered resolution. Finally, we combine our consequence-based framework with methods based on tree decompositions, and thus obtain what we believe are the first fixed-parameter tractability results for subsumption reasoning in DLs.
28
Clibbon, Alex P. "Methods of classification of the cardiotocogram." Thesis, University of Oxford, 2016. https://ora.ox.ac.uk/objects/uuid:550bb5ea-bee8-4eb8-95e2-f16c54d7cd68.
Full textAbstract:
This Thesis compares CTG classification techniques proposed in the literature and their potential extensions. A comparison between four classifiers previously assessed - Adaboost, Artificial Neural Networks (ANN), Random Forest (RF), Support Vector Machine (SVM) - and two proposed classifiers - Bayesian ANN (BANN), Relevance Vector Machine - was conducted using a database of 7,568 cases and two open source databases. The Random Forest (RF) achieved the highest average result and was proposed as a benchmark classifier. The proposal to use model certainty to introduce a third, unclassified, class was investigated using the BANN. An increase in the classification accuracy was demonstrated, however the proportion of cases in the unclassified class was too great to be of practical value. The information content of time series was explored using a Hidden Markov Model (HMM). The average performance of the HMM was comparable with the performance of the benchmark with a smaller distribution across validation folds, demonstrating that time-series information provides more stable estimates of class than stationary methods. Finally a method of system identification was implemented. Significant differences between feature trends and histograms in low pH (< 7.1) and healthy pH (≥ 7.1) cases were observed. These features were used as classifier inputs, and achieved performance similar to existing feature sets. When these features were aligned according the onset of stage 2 labour three unique trend patterns were discovered.
29
Leuthäuser, Max. "Pure Embedding of Evolving Objects." International Academy, Research, and Industry Association, 2017. https://tud.qucosa.de/id/qucosa%3A70692.
Full textAbstract:
Scripting languages are extraordinarily popular due to their very flexible object model. Dynamic extensions (i.e., adding, removing and manipulating behavior and state) allow for the evolution and adaption of objects to context changes at runtime. Introducing this flexibility into a statically typed, object-oriented language would improve programmability and separation of concerns beyond the level of what one could usually gain with inheritance, mixins, traits or manually adapted designpatterns. They often lead to object-schizophrenia or the need for hand-crafted, additional management code. Although there were already attempts bringing flexible objects into statically typed languages with the benefits of an explicitly crafted core calculus or type system, they need their own compiler and tooling which limits the usability, e.g., when dealing with existing legacy code. This work presents an embedding of dynamically evolving objects via a lightweight library approach, which is pure in the sense, that there is no need for a specific compiler or tooling. It is written in Scala, which is both a modern object-oriented and functional programming language. Our approach is promising to solve practical problems arising in the area of dynamical extensibility and adaption like role-based programming.
30
Delaitre, Vincent. "Modeling and recognizing interactions between people, objects and scenes." Thesis, Paris, Ecole normale supérieure, 2015. http://www.theses.fr/2015ENSU0003/document.
Full textAbstract:
Nous nous intéressons dans cette thèse à la modélisation des interactions entre personnes, objets et scènes. Nous montrons l’intérêt de combiner ces trois sources d’information pour améliorer la classification d’action et la compréhension automatique des scènes. Dans la première partie, nous cherchons à exploiter le contexte fourni par les objets et la scène pour améliorer la classification des actions humaines dans les photographies. Nous explorons différentes variantes du modèle dit de “bag-of-features” et proposons une méthode tirant avantage du contexte scénique. Nous proposons ensuite un nouveau modèle exploitant les objets pour la classification d’action basé sur des paires de détecteurs de parties du corps et/ou d’objet. Nous évaluons ces méthodes sur notre base de données d’images nouvellement collectée ainsi que sur trois autres jeux de données pour la classification d’action et obtenons des résultats proches de l’état de l’art. Dans la seconde partie de cette thèse, nous nous attaquons au problème inverse et cherchons à utiliser l’information contextuelle fournie par les personnes pour aider à la localisation des objets et à la compréhension des scènes. Nous collectons une nouvelle base de données de time-lapses comportant de nombreuses interactions entre personnes, objets et scènes. Nous développons une approche permettant de décrire une zone de l’image par la distribution des poses des personnes qui interagissent avec et nous utilisons cette représentation pour améliorer la localisation d’objets. De plus, nous démontrons qu’utiliser des informations provenant des personnes détectées peut améliorer plusieurs étapes de l’algorithme utilisé pour la compréhension des scènes d’intérieur. Pour finir, nous proposons des annotations 3D de notre base de time-lapses et montrons comment estimer l’espace utilisé par différentes classes d’objets dans une pièce. Pour résumer, les contributions de cette thèse sont les suivantes : (i) nous mettons au point des modèles pour la classification d’image tirant avantage du contexte scénique et des objets environnants et nous proposons une nouvelle base de données pour évaluer leurs performances, (ii) nous développons un nouveau modèle pour améliorer la localisation d’objet grâce à l’observation des acteurs humains interagissant avec une scène et nous le testons sur un nouveau jeu de vidéos comportant de nombreuses interactions entre personnes, objets et scènes, (iii) nous proposons la première méthode pour évaluer les volumes occupés par différentes classes d’objets dans une pièce, ce qui nous permet d’analyser les différentes étapes pour la compréhension automatique de scène d’intérieur et d’en identifier les principales sources d’erreurs<br>In this thesis, we focus on modeling interactions between people, objects and scenes and show benefits of combining corresponding cues for improving both action classification and scene understanding. In the first part, we seek to exploit the scene and object context to improve action classification in still images. We explore alternative bag-of-features models and propose a method that takes advantage of the scene context. We then propose a new model exploiting the object context for action classification based on pairs of body part and object detectors. We evaluate our methods on our newly collected still image dataset as well as three other datasets for action classification and show performance close to the state of the art. In the second part of this thesis, we address the reverse problem and aim at using the contextual information provided by people to help object localization and scene understanding. We collect a new dataset of time-lapse videos involving people interacting with indoor scenes. We develop an approach to describe image regions by the distribution of human co-located poses and use this pose-based representation to improve object localization. We further demonstrate that people cues can improve several steps of existing pipelines for indoor scene understanding. Finally, we extend the annotation of our time-lapse dataset to 3D and show how to infer object labels for occupied 3D volumes of a scene. To summarize, the contributions of this thesis are the following: (i) we design action classification models for still images that take advantage of the scene and object context and we gather a new dataset to evaluate their performance, (ii) we develop a new model to improve object localization thanks to observations of people interacting with an indoor scene and test it on a new dataset centered on person, object and scene interactions, (iii) we propose the first method to evaluate the volumes occupied by different object classes in a room that allow us to analyze the current 3D scene understanding pipeline and identify its main source of errors
31
Cale, Michelle. "'Saved from a life of vice and crime' : reformatory and industrial schools for girls, c.1854-c.1901." Thesis, University of Oxford, 1993. https://ora.ox.ac.uk/objects/uuid:a551dd78-6ebc-4b0d-a2fe-693e74d5e19c.
Full textAbstract:
Reformatory and industrial schools were semi-penal Victorian institutions designed, firstly, to reclaim juveniles from a nascent criminal career and, secondly, to prevent neglected children from slipping into criminality. Most existing studies of these schools have been principally concerned with the campaigning philanthropists, such as Mary Carpenter, central government activity, and institutions for boys. This thesis utilises hitherto unused archival sources relating to individual institutions for girls in order to look at various aspects of reformatory life from the perspective of those by whom it was daily experienced. In addition to a consideration of the lives of the inmates, there are discussions of the motivations of the voluntary managers and the pay and conditions of the staff. A data base of industrial school cases from the Children's Society is analysed, in conjunction with other committal records, to ascertain which children were most likely to find themselves incarcerated. The importance of the respectability of a child's mother is particularly highlighted. The role of sexuality is discussed, in relation to attitudes towards girl delinquents, the selection of girls for committal, and the prevention of immorality within the schools. The internal disciplinary regime is considered in two chapters, the first concentrating on the forms of punishment and reward common throughout the reformatory system; and the second on the kinds of outburst and disturbance which contemporaries labelled hysterical, but which could be interpreted as calculated resistance to authority. Special attention is paid to the issue of corporal punishment. The socialising education in feminine, domestic skills which was provided in these institutions is the subject of a further chapter. Finally, the destinations of the 'reformed' girls on release is analysed, again using the Children's Society cases. The little which is known of their adult lives and various tokens of 'success' or 'failure' are delineated in an attempt to assess whether the schools accomplished their numerous aims.
32
Braun, Marco, and Frank Cichos. "Gold nanostructure assisted thermophoretic trapping of single nano-objects." Diffusion fundamentals 20 (2013) 62, S. 1, 2013. https://ul.qucosa.de/id/qucosa%3A13639.
Full text
33
Follert, Florian. "On the punishment of white-collar and tax crime." Universität Leipzig, 2019. https://ul.qucosa.de/id/qucosa%3A33907.
Full textAbstract:
Economic and tax crimes account for a significant proportion of criminal activity and result in considerable economic damage. In Germany, two of the most prominent offenders in this area in recent years were Thomas Middelhoff and Uli Hoeneß, both of whom served jail sentences. Taking advantage of the widespread media coverage of both cases, the following paper draws on economic theory to compare prison terms with fines. It argues that fines are preferred from an economic perspective and can therefore be considered a useful firstchoice punishment in cases involving white-collar and tax crimes. The paper sees itself in this regard as a plea.
34
Chavez, Garcia Ricardo Omar. "Multiple sensor fusion for detection, classification and tracking of moving objects in driving environments." Thesis, Grenoble, 2014. http://www.theses.fr/2014GRENM034/document.
Full textAbstract:
Les systèmes avancés d'assistance au conducteur (ADAS) aident les conducteurs à effectuer des tâches de conduite complexes et à éviter ou atténuer les situations dangereuses. Le véhicule détecte le monde extérieur au moyen de capteurs, et ensuite construit et met à jour un modèle interne de la configuration de l'environnement. La perception de véhicule consiste à établir des relations spatiales et temporelles entre le véhicule et les obstacles statiques et mobiles dans l'environnement. Cette perception se compose de deux tâches principales : la localisation et cartographie simultanées (SLAM) traite de la modélisation de pièces statiques; et la détection et le suivi d'objets en mouvement (DATMO) est responsable de la modélisation des pièces mobiles dans l'environnement. Afin de réaliser un bon raisonnement et contrôle, le système doit modéliser correctement l'environnement. La détection précise et la classification des objets en mouvement est un aspect essentiel d'un système de suivi d'objets. Classification des objets en mouvement est nécessaire pour déterminer le comportement possible des objets entourant le véhicule, et il est généralement réalisée au niveau de suivi des objets. La connaissance de la classe d'objets en mouvement au niveau de la détection peut aider à améliorer leur suivi. La plupart des solutions de perception actuels considèrent informations de classification seulement comme information additional pour la sortie final de la perception. Aussi, la gestion de l'information incomplète est une exigence importante pour les systèmes de perception. Une information incomplète peut être originaire de raisons liées à la détection, tels que les problèmes d calibrage et les dysfonctionnements des capteurs; ou des perturbations de la scène, comme des occlusions, des problèmes de météo et objet déplacement. Les principales contributions de cette thèse se concentrent sur la scène DATMO. Précisément, nous pensons que l'inclusion de la classe de l'objet comme un élément clé de la représentation de l'objet et la gestion de l'incertitude de plusieurs capteurs de détections, peut améliorer les résultats de la tâche de perception. Par conséquent, nous abordons les problèmes de l'association de données, la fusion de capteurs, la classification et le suivi à différents niveaux au sein de la phase de DATMO. Même si nous nous concentrons sur un ensemble de trois capteurs principaux: radar, lidar, et la caméra, nous proposons une architecture modifiables pour inclure un autre type ou nombre de capteurs. Premièrement, nous définissons une représentation composite de l'objet pour inclure des informations de classe et de l'état d'objet deouis le début de la tâche de perception. Deuxièmement, nous proposons, mettre en œuvre, et comparons deux architectures de perception afin de résoudre le problème de DATMO selon le niveau où l'association des objets, la fusion et la classification des informations sont inclus et appliquées. Nos méthodes de fusion de données sont basées sur la théorie de l'evidence, qui est utilisé pour gérer et inclure l'incertitude de la détection du capteur et de la classification des objets. Troisièmement, nous proposons une approche d'association de données bassée en la théorie de l'evidence pour établir une relation entre deux liste des détections d'objets. Quatrièmement, nous intégrons nos approches de fusion dans le cadre d'une application véhicule en temps réel. Cette intégration a été réalisée dans un réelle démonstrateur de véhicule du projet European InteractIVe. Finalement, nous avons analysé et évalué expérimentalement les performances des méthodes proposées. Nous avons comparé notre fusion rapproche les uns contre les autres et contre une méthode state-of-the-art en utilisant des données réelles de scénarios de conduite différents. Ces comparaisons sont concentrés sur la détection, la classification et le suivi des différents objets en mouvement: piétons, vélos, voitures et camions<br>Advanced driver assistance systems (ADAS) help drivers to perform complex driving tasks and to avoid or mitigate dangerous situations. The vehicle senses the external world using sensors and then builds and updates an internal model of the environment configuration. Vehicle perception consists of establishing the spatial and temporal relationships between the vehicle and the static and moving obstacles in the environment. Vehicle perception is composed of two main tasks: simultaneous localization and mapping (SLAM) deals with modelling static parts; and detection and tracking moving objects (DATMO) is responsible for modelling moving parts in the environment. In order to perform a good reasoning and control, the system has to correctly model the surrounding environment. The accurate detection and classification of moving objects is a critical aspect of a moving object tracking system. Therefore, many sensors are part of a common intelligent vehicle system. Classification of moving objects is needed to determine the possible behaviour of the objects surrounding the vehicle, and it is usually performed at tracking level. Knowledge about the class of moving objects at detection level can help improve their tracking. Most of the current perception solutions consider classification information only as aggregate information for the final perception output. Also, management of incomplete information is an important requirement for perception systems. Incomplete information can be originated from sensor-related reasons, such as calibration issues and hardware malfunctions; or from scene perturbations, like occlusions, weather issues and object shifting. It is important to manage these situations by taking them into account in the perception process. The main contributions in this dissertation focus on the DATMO stage of the perception problem. Precisely, we believe that including the object's class as a key element of the object's representation and managing the uncertainty from multiple sensors detections, we can improve the results of the perception task, i.e., a more reliable list of moving objects of interest represented by their dynamic state and appearance information. Therefore, we address the problems of sensor data association, and sensor fusion for object detection, classification, and tracking at different levels within the DATMO stage. Although we focus on a set of three main sensors: radar, lidar, and camera, we propose a modifiable architecture to include other type or number of sensors. First, we define a composite object representation to include class information as a part of the object state from early stages to the final output of the perception task. Second, we propose, implement, and compare two different perception architectures to solve the DATMO problem according to the level where object association, fusion, and classification information is included and performed. Our data fusion approaches are based on the evidential framework, which is used to manage and include the uncertainty from sensor detections and object classifications. Third, we propose an evidential data association approach to establish a relationship between two sources of evidence from object detections. We observe how the class information improves the final result of the DATMO component. Fourth, we integrate the proposed fusion approaches as a part of a real-time vehicle application. This integration has been performed in a real vehicle demonstrator from the interactIVe European project. Finally, we analysed and experimentally evaluated the performance of the proposed methods. We compared our evidential fusion approaches against each other and against a state-of-the-art method using real data from different driving scenarios. These comparisons focused on the detection, classification and tracking of different moving objects: pedestrian, bike, car and truck
35
Cabrera, Vives Guillermo. "Extraction and classification of objects from astronomical images in the presence of labeling bias." Tesis, Universidad de Chile, 2015. http://repositorio.uchile.cl/handle/2250/133321.
Full textAbstract:
Doctor en Ciencias, Mención Computación<br>Giga, tera y petabytes de datos astronómicos están empezando a fluir desde la nueva generación de telescopios. Los telescopios de rastreo escanean una amplia zona del cielo con el fin de mapear la galaxia, nuestro universo, y detectar fuentes variables como la explosion de estrellas (o supernovas) y asteroides. Al igual que en otros campos de la ciencia observacional, lo único que podemos hacer es observar estas fuentes a través de la luz que emiten y que podemos capturar en nuestras cámaras. Debido a la gran distancia a la que estos objetos se encuentran, aún cuando podemos tener una caracterización estimada de estas fuentes, es imposible conocer las propiedades reales de ellas.
En esta tesis, proponemos un método para la extracción de los llamados perfiles de Sérsic de fuentes astronómicas y su aplicación a clasificación morfológica de objetos. Este perfil de Sérsic es un modelo paramétrico radial asociado con la morfología de galaxias. La novedad de nuestro enfoque es que convierte la imagen 2D en un perfil radial 1D utilizando curvas de nivel elípticas, por lo que incluso cuando el espacio de parámetros de Sérsic es el mismo, la complejidad se ve reducida 10 veces en comaración a ajustes de modelos en 2D de la literatura. Probamos nuestro método sobre simulaciones y obtenemos un error de entre un 40% y un 50% en los parámetros de Sérsic, mientras que obtenemos un chi cuadrado reducido de 1,01. Estos resultados son similares a los obtenidos por otros autores, lo que sugiere que el modelo de Sérsic es degenerado. A su vez, aplicamos nuestro método a imágenes del SDSS y mostramos que somos capaces de extraer la componente suave del perfil de las galaxias, pero, como era de
esperar, fallamos en obtener su estructura más fina.
También mostramos que las etiquetas creadas por los seres humanos son sesgadas en términos de parámetros observables: al observar galaxias pequeñas, débiles o distantes, la estructura fina de estos objetos se pierde, produciendo un sesgo en el etiquetado sistemático hacia objetos más suaves. Creamos una métrica para evaluar el nivel de sesgo en los catálogos de las etiquetas y demostramos que incluso etiquetas obtenidas por expertos muestran cierto sesgo, mientras que el sesgo es menor para etiquetas obtenidas a partir de modelos de aprendizaje supervisado. Aun cuando este sesgo ha sido notado en la literatura, hasta donde sabemos, esta es la primera vez que ha sido cuantificado. Proponemos dos métodos para des-sesgar etiquetas. El primer método se basa en seleccionar una sub-muestra no-sesgada de los datos para entrenar un modelo de clasificación, y el segundo método ajusta simultáneamente un modelo de sesgo y de clasificación a los datos. Demostramos que ambos métodos obtienen el sesgo más bajo en comparación con otros conjuntos de datos y procedimientos de procesamiento.
36
Saldanha, Richard A. "Graph-theoretic methods in discrimination and classification." Thesis, University of Oxford, 1998. https://ora.ox.ac.uk/objects/uuid:3a06dee1-00e9-4b56-be8e-e991a570ced6.
Full textAbstract:
This thesis is concerned with the graphical modelling of multivariate data. The main aim of graphical modelling is to provide an easy to understand visual representation of, often complex, data relationships by fitting graphs to data. The graphs consist of nodes denoting random variables and connecting lines or edges are used to depict variable dependencies. Equivalently, the absence of particular edges in a graph describe conditional independencies between random variables. The resulting structure is called a conditional independence graph. The use of conditional independence graphs as a guide to discrete (mainly binary), normal and mixed conditional Gaussian model building is described. The problem of parameter estimation in fitting conditional Gaussian models is considered. A FORTRAN 77 program called CGM is developed and used to fit conditional Gaussian models. Submodel specification, model selection criteria and goodness-of-fit are explored. A procedure for discriminating between groups is constructed using fitted conditional Gaussian models. A Bayesian classification procedure is considered and is used to compute posterior classification probabilities. Standard bias-correcting error rates are used to test the performance of estimated classification rules. The graph-theoretic methodology described in this thesis is applied to a Scandinavian study of intrauterine foetal growth retardation also known as a small-for-gestational age (SGA) birth. Possible pre-pregnancy risk factors associated with SGA births are investigated using conditional independence graphs and an attempt is made to classify SGA births using fitted conditional Gaussian models.
37
Hussein, Abdul Aziz. "Identifying Crime Hotspot: Evaluating the suitability of Supervised and Unsupervised Machine learning." University of Cincinnati / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1624914607243042.
Full text
38
Kumara, Muthukudage Jayantha. "Automated Real-time Objects Detection in Colonoscopy Videos for Quality Measurements." Thesis, University of North Texas, 2013. https://digital.library.unt.edu/ark:/67531/metadc283843/.
Full textAbstract:
The effectiveness of colonoscopy depends on the quality of the inspection of the colon. There was no automated measurement method to evaluate the quality of the inspection. This thesis addresses this issue by investigating an automated post-procedure quality measurement technique and proposing a novel approach automatically deciding a percentage of stool areas in images of digitized colonoscopy video files. It involves the classification of image pixels based on their color features using a new method of planes on RGB (red, green and blue) color space. The limitation of post-procedure quality measurement is that quality measurements are available long after the procedure was done and the patient was released. A better approach is to inform any sub-optimal inspection immediately so that the endoscopist can improve the quality in real-time during the procedure. This thesis also proposes an extension to post-procedure method to detect stool, bite-block, and blood regions in real-time using color features in HSV color space. These three objects play a major role in quality measurements in colonoscopy. The proposed method partitions very large positive examples of each of these objects into a number of groups. These groups are formed by taking intersection of positive examples with a hyper plane. This hyper plane is named as 'positive plane'. 'Convex hulls' are used to model positive planes. Comparisons with traditional classifiers such as K-nearest neighbor (K-NN) and support vector machines (SVM) proves the soundness of the proposed method in terms of accuracy and speed that are critical in the targeted real-time quality measurement system.
39
Stebbins, Glenn Thurston. "CLASSIFICATION OF PRISON INMATES ACCORDING TO PRISON RULES AND REGULATIONS (ENVIRONMENT)." Thesis, The University of Arizona, 1985. http://hdl.handle.net/10150/291276.
Full text
40
Stewart, Adam James. "Multifrequency search & classification of the transient radio sky." Thesis, University of Oxford, 2017. https://ora.ox.ac.uk/objects/uuid:f5a803dd-0e05-4104-be0e-fe221914f1e5.
Full textAbstract:
The radio sky can exhibit dynamic changes that provide an excellent opportunity to study the most exotic events in the Universe. In recent years the development and completion of advanced radio facilities has enabled the transient radio sky to be probed to unprecedented levels, especially at low frequencies with telescopes such as the Low-Frequency Array (LOFAR) revolutionising observations in this regime. This work primarily focuses on the detection of radio transients both at low and high frequencies. I describe the analysis of a 60 MHz LOFAR monitoring campaign centred at the North Celestial Pole, and report the discovery of the first LOFAR transient source. I also present the initial transient search of the CHILES survey that was performed with the VLA at 1.4 GHz, supported by simultaneous optical observations. Three strong transient candidates are discussed along with identifying a further 72 possible candidates. Furthermore, I investigate the relation between the radio and optical flux properties of known transient and variable sources, with the aim of developing a classification method for these types of events. I show that from just radio and optical flux measurements, it is possible to separate certain classes of objects, with additional information provided by analysing how individual classes evolve in the radio and optical parameter space over time. Lastly, I explore the future prospects of the LOFAR and CHILES transient searches, detailing what techniques could be used to further define the transient parameter space. I also discuss how the radio and optical classification method can be advanced and incorporated into automatic transient pipelines, which may be a necessity in the future Square Kilometre Array era.
41
Abderrahmane, Zineb. "Visuo-Haptic recognition of daily-life objects : a contribution to the data scarcity problem." Thesis, Montpellier, 2018. http://www.theses.fr/2018MONTS036/document.
Full textAbstract:
Il est important pour les robots de pouvoir reconnaître les objets rencontrés dans la vie quotidienne afin d’assurer leur autonomie. De nos jours, les robots sont équipés de capteurs sophistiqués permettant d’imiter le sens humain du toucher. C’est ce qui permet aux robots interagissant avec les objets de percevoir les propriétés (telles la texture, la rigidité et la matière) nécessaires pour leur reconnaissance. Dans cette thèse, notre but est d’exploiter les données haptiques issues de l’interaction robot-objet afin de reconnaître les objets de la vie quotidienne, et cela en utilisant les algorithmes d’apprentissage automatique. Le problème qui se pose est la difficulté de collecter suffisamment de données haptiques afin d’entraîner les algorithmes d’apprentissage supervisé sur tous les objets que le robot doit reconnaître. En effet, les objets de la vie quotidienne sont nombreux et l’interaction physique entre le robot et chaque objet pour la collection des données prend beaucoup de temps et d’efforts. Pour traiter ce problème, nous développons un système de reconnaissance haptique permettant de reconnaître des objets à partir d'aucune, de une seule, ou de plusieurs données d’entraînement. Enfin, nous intégrons la vision afin d’améliorer la reconnaissance d'objets lorsque le robot est équipé de caméras<br>Recognizing surrounding objects is an important skill for the autonomy of robots performing in daily-life. Nowadays robots are equipped with sophisticated sensors imitating the human sense of touch. This allows the recognition of an object based on information ensuing from robot-object physical interaction. Such information can include the object texture, compliance and material. In this thesis, we exploit haptic data to perform haptic recognition of daily life objects using machine learning techniques. The main challenge faced in our work is the difficulty of collecting a fair amount of haptic training data for all daily-life objects. This is due to the continuously growing number of objects and to the effort and time needed by the robot to physically interact with each object for data collection. We solve this problem by developing a haptic recognition framework capable of performing Zero-shot, One-shot and Multi-shot Learning. We also extend our framework by integrating vision to enhance the robot’s recognition performance, whenever such sense is available
42
Domberg, Jana Charlotte. "Methodische Probleme bei der Operationalisierung von Kriminalitätsfurcht: ein Vergleich des Standardindikators mit der deliktbezogenen Operationalisierung." Universität Leipzig, 2018. https://ul.qucosa.de/id/qucosa%3A21202.
Full textAbstract:
Zwar hat die deliktspezifische Abfrage der Kriminalitätsfurcht eine ausführliche theoretische Begründung erfahren, jedoch mangelt es an einer empirischen Überprüfung, die sich zusätzlich auch auf den Vergleich mit dem modifizierten Standardindikator konzentriert. In der folgenden Arbeit soll ein solcher Vergleich der beiden Operationalisierungen anhand eines Datensatzes erfolgen.:1. Einleitung
2. Theoretische Ansätze
3. Methode
4. Datenanalyse
5. Zusammenfassung, Methodenkritik und Ausblick
43
Densley, James Andrew. "Under the hood : the mechanics of London's street gangs." Thesis, University of Oxford, 2011. http://ora.ox.ac.uk/objects/uuid:cea29f30-a98d-4f20-828b-6556a0ac51f4.
Full textAbstract:
Based upon two years of ethnographic fieldwork in London, England, which incorporated nearly 200 interviews with gang members, gang associates, and police officers, among others, this thesis addresses three questions presently unresolved in the street gangs literature: What is the business of gangs? How are gangs organised? And how do gangs recruit? With regard the business of gangs, this thesis illustrates how recreation, crime, enterprise, and extra-legal governance represent sequential stages in the evolutionary cycle of London’s street gangs. Gang member testimony emphasises how gangs typically begin life as neighbourhood-based peer groups, but also how, in response to external threats and financial commitments, gangs grow to incorporate street-level drugs distribution businesses that very much resemble the multi-level marketing structure of direct-sales companies. People join gangs to make money, achieve status, and obtain protection. Gangs engage in turf wars, acquire violent resources, and develop hierarchical structures in order to maintain provision of these desirable goods and services. Gang organisation, in turn, becomes a function of gang business. To better understand the nature and extent of gang organisation, this thesis moves on to discuss the presence of subgroups, hierarchy and leadership, pecuniary and non-pecuniary incentives, rules, responsibilities, and restrictions, and consequences for absconding within gangs. It further presents how, in order to convey reputation and achieve intimidation, gangs seek association with elements of popular culture that help promote their image. Finally, through the novel application of signalling theory to the gang recruitment process, this thesis demonstrates how gangs face a primary trust dilemma in their uncertainty over the quality of recruits. Given that none of the trust-warranting properties for gang membership can be readily discovered from observation, gangs look for observable signs correlated with these properties. Gangs face a secondary trust dilemma in their uncertainty over the reliability of signs because certain agents (e.g., police informants, rival gang members, and adventure-seekers) have incentives to mimic them. To overcome their informational asymmetry gangs thus screen for signs that are too costly for mimics to fake but affordable for the genuine article. The thesis concludes with a discussion of gang desistance and intervention in the context of escalating youth violence in London.
44
Rao, Aparna. "The defendant's bad character in the wake of the Criminal Justice Act 2003." Thesis, University of Oxford, 2014. http://ora.ox.ac.uk/objects/uuid:818b81e9-91bc-41dd-9f40-1bb57b0d45b2.
Full textAbstract:
This thesis examines the interpretation and application of the Criminal Justice Act 2003 (UK), Part 11 Chapter 1, which came into force on 15 December 2004. Part 11 Chapter 1 concerns evidence of bad character, a concept approximately comparable with common law similar fact evidence, in relation to all parties who may be connected with a criminal trial. The admission and use of similar fact evidence has often been the subject of controversy, and the significant changes made by the CJA 2003 have attracted their own body of support and criticism. The nine chapters of this thesis attempt an in-depth study into the impact of the legislation on the robustness and effectiveness of the criminal trial, and consider whether the criminal trial is suited to the level of exposure of bad character now facilitated by the CJA. In particular, the thesis focuses upon the key provisions governing the uses of bad character evidence of the defendant: the seven gateways set out in s 101 of the CJA. The operation of those gateways and their accompanying explanatory provisions is examined through a combination of engagement with the Law Commission’s Report 273 (which preceded the enactment of the legislation), the range of Court of Appeal cases dealing with the legislation, and academic commentary. It was foreshadowed by commentators and early case law that the new provisions might not be easy to interpret or apply, and subsequent cases have borne out this prediction. An analysis of the bad character provisions suggests that, even though the CJA was intended to provide clarity in regulation, it has itself led to confusion in some important respects. Certain central terms lack definition, and some provisions have unintended consequences. The case law reflects this in its frequent, often brief, and sometimes inconsistent analysis of the specific parts of the legislation, which can make it difficult to determine the defendant’s guilt or innocence in a precise and scrupulous manner.
45
Marshall, Jeremy H. "Natural classification and the reality of higher taxa." Thesis, University of Oxford, 1989. http://ora.ox.ac.uk/objects/uuid:240dd5d5-41f2-4378-a6d0-4c11c1e5351a.
Full textAbstract:
Having outlined the present situation as regards rival taxonomic philosophies, and some of its historical background, the thesis examines this attempt to recategorize taxa as individual-like entities, and finds it wanting. The properties of species which render them regardable as individuals do not readily extend to more inclusive levels, or, if they do, are not readily restricted solely to cladistic taxa. Cladistic systematization, in moving away from the notion of a taxon as a class of similar entities, may cease to convey the information expected of a classification system. The practice of biology requires a more flexible and more stable taxonomy than can be provided by strict adherence to cladistic rules, and taxa are-better regarded as 'historical classes', delineated neither by pure unanalysed similarity nor by logical transformation of hypotheses of phylogenetic relationship, but by a considered pragmatic synthesis of the two, employing the notion of convexity as a criterion of acceptability.
46
Grimmett, Hugo. "Introspective classification for robot perception and decision making." Thesis, University of Oxford, 2016. https://ora.ox.ac.uk/objects/uuid:8efd23ff-e853-492b-8401-eb96fd392582.
Full textAbstract:
In robotics, a classifier is often a core component of the decision-making framework. Precision and recall have been widely adopted as canonical metrics to quantify the performance of a classifier, but for applications involving mission-critical decision making, good performance in relation to these metrics is insufficient. The use of a classification framework which produces scores with inappropriate confidences will ultimately lead to the robot making bad decisions, thereby compromising robot or user safety. In order to select a classifier which will make decisions reflecting the nature of the costs, we should pay careful attention to the ways in which it generates scores. We introduce and motivate the importance of a classifier's introspective capacity: the ability to give an appropriate assessment of confidence with any test case. Classification made confidently must be correct, and mistakes should be made with high uncertainty. A classifier's capacity to do so must remain consistent despite unusual or surprising test cases. We propose that a key ingredient for introspection is a classifier's potential to increase its uncertainty with the distance between a test datum and its training data. We define the ideal introspective behaviour, and derive idealised classifiers which serve to benchmark a number of commonly used classification frameworks in a variety of decision-making tasks. We show that classifiers that offer predictive variance at test-time are more cautious and less over-confident than those which consider a single hypothesis or discriminant. However, in high-cost (or high-risk) decision making, none of the classifiers evaluated in this thesis are sufficiently introspective to prevent all potential catastrophic mistakes. We show that in sequential decision-making, when the mapping from score to class is explicitly stated, a classifier's ability to behave consistently despite non-stationary test data is of primary importance.
47
Armas, Romero Ana. "Ontology module extraction and applications to ontology classification." Thesis, University of Oxford, 2015. http://ora.ox.ac.uk/objects/uuid:4ec888f4-b7c0-4080-9d9a-3c46c91f67e3.
Full textAbstract:
Module extraction is the task of computing a (preferably small) fragment <i>M</i> of an ontology <i>O</i> that preserves a class of entailments over a signature of interest ∑. Existing practical approaches ensure that <i>M</i> preserves all second-order entailments of <i>O</i> over ∑, which is a stronger condition than is required in many applications. In the first part of this thesis, we propose a novel approach to module extraction which, based on a reduction to a datalog reasoning problem, makes it possible to compute modules that are tailored to preserve only specific kinds of entailments. This leads to obtaining modules that are often significantly smaller than those produced by other practical approaches, as shown in an empirical evaluation. In the second part of this thesis, we consider the application of module extraction to the optimisation of ontology classification. Classification is a fundamental reasoning task in ontology design, and there is currently a wide range of reasoners that provide this service. Reasoners aimed at so-called lightweight ontology languages are much more efficient than those aimed at more expressive ones, but they do not offer completeness guarantees for ontologies containing axioms outside the relevant language. We propose an original approach to classification based on exploiting module extraction techniques to divide the workload between a general purpose reasoner and a more efficient reasoner for a lightweight language in such a way that the bulk of the workload is assigned to the latter. We show how the proposed approach can be realised using two particular module extraction techniques, including the one presented in the first part of the thesis. Furthermore, we present the results of an empirical evaluation that shows that this approach can lead to a significant performance improvement in many cases.
48
Fennell, Jac, Cathy Treadaway, and Aidan Taylor. "Open Doors: designing playful objects for dementia." TUDpress, 2019. https://tud.qucosa.de/id/qucosa%3A36604.
Full textAbstract:
This design-based submission comprises a prototype design for a playful object for use in dementia care and the ‘Open Doors’ documentary video that explains the underpinning research collaboration. LUMA is a hand-held playful object, designed for people living with advanced dementia, and is one of six design outputs from the recently completed AHRC-funded LAUGH project. This qualitative design research used participatory and co-design methods and Compassionate Design methodology to investigate how playful objects can be designed to support the wellbeing of people living with advanced dementia. One of the major challenges facing society is how to provide appropriate care for the increasing numbers of people living with dementia and to ensure that they are able to live well, right until the end of their lives. The aim of the research was to investigate ways to stimulate, engage and bring pleasure to people living with advanced dementia through the creation of simple hand-held devices. LUMA is an object that was developed in collaboration with members of the Men’s Shed in Tondu. The accompanying ‘Open Doors’ film explains how the LAUGH design team exchanged their digital fabrication expertise with members of the Men’s Shed who were skilled in hand crafting wooden objects. The creative collaboration resulted in LUMA, a hand-held interactive object that brings the outside experience of nature inside through touch, light and sound.
49
Dura, Martinez Esther. "Reconstruction and classification of man-made objects and textured seafloors from side-scan sonar images." Thesis, Heriot-Watt University, 2002. http://hdl.handle.net/10399/409.
Full text
50
Wentzlof, Chloe Ann. "A Comparision of Classification Models for Predicting Criminal Case Disposition of Police Crime in the United States." Bowling Green State University / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1554993250668438.
Full text