To see the other types of publications on this topic, follow the link: Fake image.
Dissertations / Theses on the topic 'Fake image'
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 'Fake image.'
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
Wu, Yi. "Artificial-Pulse-Noise based Multipath Suppression Method for Through-Wall Imaging Radar." Thesis, The University of Sydney, 2019. http://hdl.handle.net/2123/19989.
Full textAbstract:
Through-Wall Radar Imaging (TWRI) uses UWB radar technology to detect the targets behind the wall, complete the panoramic view of the target area after the wall. As a new generation of perspective imaging technology, through-wall radar imaging can be widely used in many fields. However, in through-wall radar imaging, a very commonly seen problem is the fake image. The fake image is caused by the sidewall and back wall reflecting the object signal. The matched-filter results cannot distinguish the fake image from the real image in current research. This thesis proposes a new method to suppress fake images caused by the multipath signal in single target and multi-target scenarios. I consider if adding artificial-pulse-noise(APN) into the received raw radar data，after matched-filter processing，the multipath echoes will be suppressed. This thesis first discusses the multipath radar signal model. By further analyzing the matching filter results of direct path signal and multipath signal with and without APN, it can be theoretically proved that adding APN can suppress the fake image. With the radar data from numerical simulation and experimental results, adding APN has been proved to effectively suppress multipath signals in the matched-filter processing. By using the proposed method in through-wall image algorithm, numerical simulation and experimental results show that the proposed method can suppress the ghost images caused by multi-path reflections from the side and back walls in enclosed rooms in a more effective way. In this thesis, a new TWRI algorithm is proposed to suppress fake images without increasing computational complexity. This means that it is easier to use in reality. Using this method, the existing through-wall radar system can be upgraded, which will make the through-wall radar become more important in anti-terrorism and disaster relief.
2
Silva, Simone Faustino da. "Diva, Presidenta e fake: a construÃÃo da imagem de Dilma Rousseff pelo perfil Dilma Bolada no Twitter." Universidade Federal do CearÃ, 2015. http://www.teses.ufc.br/tde_busca/arquivo.php?codArquivo=15389.
Full textAbstract:
nÃo hÃEm um processo contÃnuo de construÃÃo da opiniÃo e moldagem de representaÃÃes nas redes de comunicaÃÃo digital, destaca-se a produÃÃo de conteÃdo e a interaÃÃo realizada por meio da apropriaÃÃo da identidade de terceiros. A cada dia, povoam a internet novos perfis nÃo-oficiais de personalidades conhecidas, estando os atores polÃticos (continuamente sob julgamento pÃblico) especialmente vulnerÃveis a aÃÃes dessa natureza. Em consonÃncia com essa realidade, a presente dissertaÃÃo realiza um estudo qualitativo, ancorado metodologicamente na abordagem da AnÃlise de ConteÃdo (AC) de Bardin (2011) e executado pela prÃpria pesquisadora com o apoio do software Qualitative Solutions Research NVivo em sua versÃo 10.0, facilitador do processo de organizaÃÃo e categorizaÃÃo. Dessa forma, investigou-se a construÃÃo feita da imagem pÃblica da Presidenta Dilma Rousseff na rede social Twitter pelo perfil fake Dilma Bolada, um dos mais populares da internet. A escolha da rede justifica-se pela perspectiva personalizada adotada pela conta âfalsaâ. Este, inclusive, à um caso notÃrio perfil fake cuja visibilidade jà ultrapassa aquela obtida pela conta oficial em algumas redes, como o Facebook e o Instagram. Com base em um recorte temporal que vai de 1 de janeiro atà 5 de julho de 2014 (Ãltimo dia para registro de candidatos a Presidente e Vice, segundo o TSE), foram mapeados no conteÃdo das postagens indÃcios de uma construÃÃo estratÃgica, mÃtica, midiatizada e personalizada da imagem pÃblica, em uma anÃlise relacionada à literatura revisada nos capÃtulos introdutÃrios.
3
Linnander, Mathilda. "Last Night in Sweden : A Critical Discourse Analysis of the Image of Sweden in International Media." Thesis, Högskolan i Jönköping, Högskolan för lärande och kommunikation, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:hj:diva-40977.
Full textAbstract:
This is a study of how the image of Sweden is constructed in international media. Using the country as a swinging bat in debates on socialism and progressiveness is nothing new but has had an upswing during recent years as a result of the global rise of right-wing forces. With the help of Critical Discourse Analysis, four articles from the United States and the United Kingdom are analysed. These are then presented according to Fairclough’s three-layered model. With the help of previous research on Sweden in international media, fake news and nation branding, these findings are then explained and put into context.The study finds that the image of Sweden presented in media tends to follow the narrative of Good Sweden and Bad Sweden. On the one hand is the classic welfare state in the north, which takes care of its people and with high levels of trust between the actors. On the other hand is a country in ruins as a result of letting in too many immigrants. Both narratives rely heavily on stereotypes. The discussion tends to use Sweden as an example, when it is really about ideologies and values. Another result shown by the study is that fake news is a common trace in news about Sweden, not only in alternative media but also in the established elite media. This can be seen as a result of the hardening situation in the media business as well as the rise of right-wing forces.
4
Silva, Simone Faustino da. "Diva, Presidenta e fake: a construção da imagem de Dilma Rousseff pelo perfil Dilma Bolada no twitter." www.teses.ufc.br, 2015. http://www.repositorio.ufc.br/handle/riufc/14624.
Full textAbstract:
SILVA, Simone Faustino da. Diva, Presidenta e fake: a construção da imagem de Dilma Rousseff pelo perfil Dilma Bolada no twitter. 2015. 188f. – Dissertação (Mestrado) – Universidade Federal do Ceará, Programa de Pós-graduação em Comunicação Social, Fortaleza (CE), 2015.Submitted by Márcia Araújo (marcia_m_bezerra@yahoo.com.br) on 2015-12-22T12:14:40Z No. of bitstreams: 1 2015_dis_sfsilva.pdf: 2961489 bytes, checksum: f295f32a6239810f3559738a295d4da7 (MD5)
Approved for entry into archive by Márcia Araújo(marcia_m_bezerra@yahoo.com.br) on 2015-12-22T12:31:18Z (GMT) No. of bitstreams: 1 2015_dis_sfsilva.pdf: 2961489 bytes, checksum: f295f32a6239810f3559738a295d4da7 (MD5)
Made available in DSpace on 2015-12-22T12:31:18Z (GMT). No. of bitstreams: 1 2015_dis_sfsilva.pdf: 2961489 bytes, checksum: f295f32a6239810f3559738a295d4da7 (MD5) Previous issue date: 2015
Em um processo contínuo de construção da opinião e moldagem de representações nas redes de comunicação digital, destaca-se a produção de conteúdo e a interação realizada por meio da apropriação da identidade de terceiros. A cada dia, povoam a internet novos perfis não-oficiais de personalidades conhecidas, estando os atores políticos (continuamente sob julgamento público) especialmente vulneráveis a ações dessa natureza. Em consonância com essa realidade, a presente dissertação realiza um estudo qualitativo, ancorado metodologicamente na abordagem da Análise de Conteúdo (AC) de Bardin (2011) e executado pela própria pesquisadora com o apoio do software Qualitative Solutions Research NVivo em sua versão 10.0, facilitador do processo de organização e categorização. Dessa forma, investigou-se a construção feita da imagem pública da Presidenta Dilma Rousseff na rede social Twitter pelo perfil fake Dilma Bolada, um dos mais populares da internet. A escolha da rede justifica-se pela perspectiva personalizada adotada pela conta “falsa”. Este, inclusive, é um caso notório perfil fake cuja visibilidade já ultrapassa aquela obtida pela conta oficial em algumas redes, como o Facebook e o Instagram. Com base em um recorte temporal que vai de 1º de janeiro até 5 de julho de 2014 (último dia para registro de candidatos a Presidente e Vice, segundo o TSE), foram mapeados no conteúdo das postagens indícios de uma construção estratégica, mítica, midiatizada e personalizada da imagem pública, em uma análise relacionada à literatura revisada nos capítulos introdutórios.
5
McIntyre, A. H. "Applying psychology to forensic facial identification : perception and identification of facial composite images and facial image comparison." Thesis, University of Stirling, 2012. http://hdl.handle.net/1893/9077.
Full textAbstract:
Eyewitness recognition is acknowledged to be prone to error but there is less understanding of difficulty in discriminating unfamiliar faces. This thesis examined the effects of face perception on identification of facial composites, and on unfamiliar face image comparison. Facial composites depict face memories by reconstructing features and configurations to form a likeness. They are generally reconstructed from an unfamiliar face memory, and will be unavoidably flawed. Identification will require perception of any accurate features, by someone who is familiar with the suspect and performance is typically poor. In typical face perception, face images are processed efficiently as complete units of information. Chapter 2 explored the possibility that holistic processing of inaccurate composite configurations will impair identification of individual features. Composites were split below the eyes and misaligned to impair holistic analysis (cf. Young, Hellawell, & Jay, 1987); identification was significantly enhanced, indicating that perceptual expertise with inaccurate configurations exerts powerful effects that can be reduced by enabling featural analysis. Facial composite recognition is difficult, which means that perception and judgement will be influence by an affective recognition bias: smiles enhance perceived familiarity, while negative expressions produce the opposite effect. In applied use, facial composites are generally produced from unpleasant memories and will convey negative expression; affective bias will, therefore, be important for facial composite recognition. Chapter 3 explored the effect of positive expression on composite identification: composite expressions were enhanced, and positive affect significantly increased identification. Affective quality rather than expression strength mediated the effect, with subtle manipulations being very effective. Facial image comparison (FIC) involves discrimination of two or more face images. Accuracy in unfamiliar face matching is typically in the region of 70%, and as discrimination is difficult, may be influenced by affective bias. Chapter 4 explored the smiling face effect in unfamiliar face matching. When multiple items were compared, positive affect did not enhance performance and false positive identification increased. With a delayed matching procedure, identification was not enhanced but in contrast to face recognition and simultaneous matching, positive affect improved rejection of foil images. Distinctive faces are easier to discriminate. Chapter 5 evaluated a systematic caricature transformation as a means to increase distinctiveness and enhance discrimination of unfamiliar faces. Identification of matching face images did not improve, but successful rejection of non-matching items was significantly enhanced. Chapter 6 used face matching to explore the basis of own race bias in face perception. Other race faces were manipulated to show own race facial variation, and own race faces to show African American facial variation. When multiple face images were matched simultaneously, the transformation impaired performance for all of the images; but when images were individually matched, the transformation improved perception of other race faces and discrimination of own race faces declined. Transformation of Japanese faces to show own race dimensions produced the same pattern of effects but failed to reach significance. The results provide support for both perceptual expertise and featural processing theories of own race bias. Results are interpreted with reference to face perception theories; implications for application and future study are discussed.
6
Wysocki, Bruna. "Interação face a face: um estudo das estratégias discursivas na reconstrução da imagem." Universidade de São Paulo, 2008. http://www.teses.usp.br/teses/disponiveis/8/8142/tde-05052008-153601/.
Full textAbstract:
O presente trabalho tem por proposta analisar o discurso formulado por um expresidente que pretende reconstruir sua imagem abalada em decorrência do impeachment que sofreu. Ao considerarmos um encontro social, em que os interlocutores interagem face a face, observamos que o interactante procura articular estratégias interacionais, a fim de preservar sua face e protegê-la de eventuais ameaças; ao mesmo tempo, coordena estratégias argumentativas com o intuito de interferir na concepção inicial que seus interlocutores possuem a respeito de sua imagem. Para atingirmos nossos objetivos, consideramos, da Sociolingüística Interacional, as teorias de preservação de faces abordadas por Goffman (1974) e, ao levarmos em conta que estratégias argumentativas também colaboram para a reconstrução da imagem, recorremos aos pressupostos da Teoria da Argumentação, segundo estudos de Perelman e Olbrechts-Tyteca (2002). Com base nesses estudos, partimos para a aplicação dos conceitos em um corpus constituído por uma entrevista televisiva, transmitida pelo Sistema Brasileiro de Televisão (SBT), em agosto de 1998, no \"Programa Livre\", em que o ex-presidente Fernando Collor de Melo é entrevistado por estudantes de ensino médio e cursinho. O corpus foi gravado e transcrito de acordo com as normas publicadas pelo Projeto da Norma Urbana Culta - NURC-SP.This paper has the purpose of analyzing the discourse delivered by a former president that plans to rebuild his image, since it was shattered as a result of an impeachment sustained by him. Upon considering a social gathering, in which the interlocutors interact face to face, we have noted that the interacting person tries to coordinate interactional strategies in order to preserve his face and protect it against any threats; at the same time, he organizes strategic arguments for the purpose of interfering with the initial assumption that his interlocutors have made in regard to his image. To achieve our goals, we have used, from the Interactional Sociolinguistics, the theory of faces\' preservation as approached by Goffman (1974) and, by taking into account that strategic arguments too cooperate for the reconstruction of the image, we have relied on the assumptions of the Argumentation Theory, according to studies by Perelman & Olbrechts-Tyteca (2002). With basis on this information, we set out to apply the concepts in a corpus made up by a television interview broadcast by SBT, a television network, in August 1998, in the \"Programa Livre\" talk show, in which former President Fernando Collor de Melo was interviewed by high school students. The corpus was recorded and transcribed according to the rules published by Urban Educated Norm Project - NURC - SP
7
Mahmood, Muhammad Tariq. "Face Detection by Image Discriminating." Thesis, Blekinge Tekniska Högskola, Avdelningen för för interaktion och systemdesign, 2006. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-4352.
Full textAbstract:
Human face recognition systems have gained a considerable attention during last few years. There are very many applications with respect to security, sensitivity and secrecy. Face detection is the most important and first step of recognition system. Human face is non rigid and has very many variations regarding image conditions, size, resolution, poses and rotation. Its accurate and robust detection has been a challenge for the researcher. A number of methods and techniques are proposed but due to a huge number of variations no one technique is much successful for all kinds of faces and images. Some methods are exhibiting good results in certain conditions and others are good with different kinds of images. Image discriminating techniques are widely used for pattern and image analysis. Common discriminating methods are discussed.SIPL, Mechatronics, GIST 1 Oryong-Dong, Buk-Gu, Gwangju, 500-712 South Korea tel. 0082-62-970-2997
8
Youmaran, Richard. "Algorithms to Process and Measure Biometric Information Content in Low Quality Face and Iris Images." Thesis, Université d'Ottawa / University of Ottawa, 2011. http://hdl.handle.net/10393/19729.
Full textAbstract:
Biometric systems allow identification of human persons based on physiological or behavioral characteristics, such as voice, handprint, iris or facial characteristics. The use of face and iris recognition as a way to authenticate user’s identities has been a topic of research for years. Present iris recognition systems require that subjects stand close (<2m) to the imaging camera and look for a period of about three seconds until the data are captured. This cooperative behavior is required in order to capture quality images for accurate recognition. This will eventually restrict the amount of practical applications where iris recognition can be applied, especially in an uncontrolled environment where subjects are not expected to cooperate such as criminals and terrorists, for example. For this reason, this thesis develops a collection of methods to deal with low quality face and iris images and that can be applied for face and iris recognition in a non-cooperative environment. This thesis makes the following main contributions: I. For eye and face tracking in low quality images, a new robust method is developed. The proposed system consists of three parts: face localization, eye detection and eye tracking. This is accomplished using traditional image-based passive techniques such as shape information of the eye and active based methods which exploit the spectral properties of the pupil under IR illumination. The developed method is also tested on underexposed images where the subject shows large head movements. II. For iris recognition, a new technique is developed for accurate iris segmentation in low quality images where a major portion of the iris is occluded. Most existing methods perform generally quite well but tend to overestimate the occluded regions, and thus lose iris information that could be used for identification. This information loss is potentially important in the covert surveillance applications we consider in this thesis. Once the iris region is properly segmented using the developed method, the biometric feature information is calculated for the iris region using the relative entropy technique. Iris biometric feature information is calculated using two different feature decomposition algorithms based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). III. For face recognition, a new approach is developed to measure biometric feature information and the changes in biometric sample quality resulting from image degradations. A definition of biometric feature information is introduced and an algorithm to measure it proposed, based on a set of population and individual biometric features, as measured by a biometric algorithm under test. Examples of its application were shown for two different face recognition algorithms based on PCA (Eigenface) and Fisher Linear Discriminant (FLD) feature decompositions.
9
Tan, Teewoon. "HUMAN FACE RECOGNITION BASED ON FRACTAL IMAGE CODING." University of Sydney. Electrical and Information Engineering, 2004. http://hdl.handle.net/2123/586.
Full textAbstract:
Human face recognition is an important area in the field of biometrics. It has been an active area of research for several decades, but still remains a challenging problem because of the complexity of the human face. In this thesis we describe fully automatic solutions that can locate faces and then perform identification and verification. We present a solution for face localisation using eye locations. We derive an efficient representation for the decision hyperplane of linear and nonlinear Support Vector Machines (SVMs). For this we introduce the novel concept of $\rho$ and $\eta$ prototypes. The standard formulation for the decision hyperplane is reformulated and expressed in terms of the two prototypes. Different kernels are treated separately to achieve further classification efficiency and to facilitate its adaptation to operate with the fast Fourier transform to achieve fast eye detection. Using the eye locations, we extract and normalise the face for size and in-plane rotations. Our method produces a more efficient representation of the SVM decision hyperplane than the well-known reduced set methods. As a result, our eye detection subsystem is faster and more accurate. The use of fractals and fractal image coding for object recognition has been proposed and used by others. Fractal codes have been used as features for recognition, but we need to take into account the distance between codes, and to ensure the continuity of the parameters of the code. We use a method based on fractal image coding for recognition, which we call the Fractal Neighbour Distance (FND). The FND relies on the Euclidean metric and the uniqueness of the attractor of a fractal code. An advantage of using the FND over fractal codes as features is that we do not have to worry about the uniqueness of, and distance between, codes. We only require the uniqueness of the attractor, which is already an implied property of a properly generated fractal code. Similar methods to the FND have been proposed by others, but what distinguishes our work from the rest is that we investigate the FND in greater detail and use our findings to improve the recognition rate. Our investigations reveal that the FND has some inherent invariance to translation, scale, rotation and changes to illumination. These invariances are image dependent and are affected by fractal encoding parameters. The parameters that have the greatest effect on recognition accuracy are the contrast scaling factor, luminance shift factor and the type of range block partitioning. The contrast scaling factor affect the convergence and eventual convergence rate of a fractal decoding process. We propose a novel method of controlling the convergence rate by altering the contrast scaling factor in a controlled manner, which has not been possible before. This helped us improve the recognition rate because under certain conditions better results are achievable from using a slower rate of convergence. We also investigate the effects of varying the luminance shift factor, and examine three different types of range block partitioning schemes. They are Quad-tree, HV and uniform partitioning. We performed experiments using various face datasets, and the results show that our method indeed performs better than many accepted methods such as eigenfaces. The experiments also show that the FND based classifier increases the separation between classes. The standard FND is further improved by incorporating the use of localised weights. A local search algorithm is introduced to find a best matching local feature using this locally weighted FND. The scores from a set of these locally weighted FND operations are then combined to obtain a global score, which is used as a measure of the similarity between two face images. Each local FND operation possesses the distortion invariant properties described above. Combined with the search procedure, the method has the potential to be invariant to a larger class of non-linear distortions. We also present a set of locally weighted FNDs that concentrate around the upper part of the face encompassing the eyes and nose. This design was motivated by the fact that the region around the eyes has more information for discrimination. Better performance is achieved by using different sets of weights for identification and verification. For facial verification, performance is further improved by using normalised scores and client specific thresholding. In this case, our results are competitive with current state-of-the-art methods, and in some cases outperform all those to which they were compared. For facial identification, under some conditions the weighted FND performs better than the standard FND. However, the weighted FND still has its short comings when some datasets are used, where its performance is not much better than the standard FND. To alleviate this problem we introduce a voting scheme that operates with normalised versions of the weighted FND. Although there are no improvements at lower matching ranks using this method, there are significant improvements for larger matching ranks. Our methods offer advantages over some well-accepted approaches such as eigenfaces, neural networks and those that use statistical learning theory. Some of the advantages are: new faces can be enrolled without re-training involving the whole database; faces can be removed from the database without the need for re-training; there are inherent invariances to face distortions; it is relatively simple to implement; and it is not model-based so there are no model parameters that need to be tweaked.
10
Tan, Teewoon. "HUMAN FACE RECOGNITION BASED ON FRACTAL IMAGE CODING." Thesis, The University of Sydney, 2003. http://hdl.handle.net/2123/586.
Full textAbstract:
Human face recognition is an important area in the field of biometrics. It has been an active area of research for several decades, but still remains a challenging problem because of the complexity of the human face. In this thesis we describe fully automatic solutions that can locate faces and then perform identification and verification. We present a solution for face localisation using eye locations. We derive an efficient representation for the decision hyperplane of linear and nonlinear Support Vector Machines (SVMs). For this we introduce the novel concept of $\rho$ and $\eta$ prototypes. The standard formulation for the decision hyperplane is reformulated and expressed in terms of the two prototypes. Different kernels are treated separately to achieve further classification efficiency and to facilitate its adaptation to operate with the fast Fourier transform to achieve fast eye detection. Using the eye locations, we extract and normalise the face for size and in-plane rotations. Our method produces a more efficient representation of the SVM decision hyperplane than the well-known reduced set methods. As a result, our eye detection subsystem is faster and more accurate. The use of fractals and fractal image coding for object recognition has been proposed and used by others. Fractal codes have been used as features for recognition, but we need to take into account the distance between codes, and to ensure the continuity of the parameters of the code. We use a method based on fractal image coding for recognition, which we call the Fractal Neighbour Distance (FND). The FND relies on the Euclidean metric and the uniqueness of the attractor of a fractal code. An advantage of using the FND over fractal codes as features is that we do not have to worry about the uniqueness of, and distance between, codes. We only require the uniqueness of the attractor, which is already an implied property of a properly generated fractal code. Similar methods to the FND have been proposed by others, but what distinguishes our work from the rest is that we investigate the FND in greater detail and use our findings to improve the recognition rate. Our investigations reveal that the FND has some inherent invariance to translation, scale, rotation and changes to illumination. These invariances are image dependent and are affected by fractal encoding parameters. The parameters that have the greatest effect on recognition accuracy are the contrast scaling factor, luminance shift factor and the type of range block partitioning. The contrast scaling factor affect the convergence and eventual convergence rate of a fractal decoding process. We propose a novel method of controlling the convergence rate by altering the contrast scaling factor in a controlled manner, which has not been possible before. This helped us improve the recognition rate because under certain conditions better results are achievable from using a slower rate of convergence. We also investigate the effects of varying the luminance shift factor, and examine three different types of range block partitioning schemes. They are Quad-tree, HV and uniform partitioning. We performed experiments using various face datasets, and the results show that our method indeed performs better than many accepted methods such as eigenfaces. The experiments also show that the FND based classifier increases the separation between classes. The standard FND is further improved by incorporating the use of localised weights. A local search algorithm is introduced to find a best matching local feature using this locally weighted FND. The scores from a set of these locally weighted FND operations are then combined to obtain a global score, which is used as a measure of the similarity between two face images. Each local FND operation possesses the distortion invariant properties described above. Combined with the search procedure, the method has the potential to be invariant to a larger class of non-linear distortions. We also present a set of locally weighted FNDs that concentrate around the upper part of the face encompassing the eyes and nose. This design was motivated by the fact that the region around the eyes has more information for discrimination. Better performance is achieved by using different sets of weights for identification and verification. For facial verification, performance is further improved by using normalised scores and client specific thresholding. In this case, our results are competitive with current state-of-the-art methods, and in some cases outperform all those to which they were compared. For facial identification, under some conditions the weighted FND performs better than the standard FND. However, the weighted FND still has its short comings when some datasets are used, where its performance is not much better than the standard FND. To alleviate this problem we introduce a voting scheme that operates with normalised versions of the weighted FND. Although there are no improvements at lower matching ranks using this method, there are significant improvements for larger matching ranks. Our methods offer advantages over some well-accepted approaches such as eigenfaces, neural networks and those that use statistical learning theory. Some of the advantages are: new faces can be enrolled without re-training involving the whole database; faces can be removed from the database without the need for re-training; there are inherent invariances to face distortions; it is relatively simple to implement; and it is not model-based so there are no model parameters that need to be tweaked.
11
Broderick, Shawn D. "A Comparison of Mathematical Discourse in Online and Face-to-Face Environments." Diss., CLICK HERE for online access, 2009. http://contentdm.lib.byu.edu/ETD/image/etd2820.pdf.
Full text
12
Baldasso, Rosane Pérez. "Emprego da fotoantropometria para compreensão do perfil de envelhecimento facial." Universidade de São Paulo, 2016. http://www.teses.usp.br/teses/disponiveis/23/23153/tde-06032017-134731/.
Full textAbstract:
Muitas são as tentativas de projetar a aparência de uma pessoa através do tempo. Um recurso muito utilizado para este fim é a simulação de progressão de idade, que consiste na modificação de uma fotografia de uma pessoa, representando o efeito do envelhecimento sobre a sua aparência por meio de processamento de imagem digital ou de desenhos artísticos. Porém, as metodologias utilizadas usualmente são subjetivas. A inexistência de uma metodologia cientificamente consolidada capaz de simular mudanças ocorridas na face, por meio de imagens, evidencia a necessidade de estudos na área, visando subsidiar o perito criminal para a execução da técnica e sua aplicação como auxiliar nos exames de Identificação Facial Forense (IFF). A principal premissa envolvida em qualquer método de Identificação Humana é a da \"unicidade\" da informação utilizada como meio de comparação. Por unicidade, entende-se a individualidade (ou não repetitividade) do parâmetro estudado, que deve ser mensurada com a realização de estudos populacionais específicos, sendo que algumas características faciais se mantêm ao longo dos anos. O presente trabalho teve como objetivo avaliar modificações no padrão métrico da face do indivíduo adulto em função da idade, em específico no que diz respeito ao crescimento das orelhas, crescimento de nariz e alteração da espessura dos lábios, visando a compreensão do processo de envelhecimento e o desenvolvimento de metodologias para estimativas de progressão de idade em casos de desaparecidos e foragidos, dando subsídios ao perito criminal para execução da técnica de simulação de envelhecimento da face em imagens. Para tanto, foram utilizadas imagens faciais padronizadas de 700 indivíduos adultos brasileiros, do sexo feminino e do sexo masculino, da região Sul, ancestralidade branca e em idades de 20, 30, 40, 50, 60, 70 e 80 anos, todas em norma frontal oriundas do Sistema Nacional de Passaporte (SINPA), gerenciado pela Polícia Federal. Os resultados apontaram um aumento gradual e mensurável da altura das orelhas, intimamente relacionado com o aumento dos lóbulos, aumento da largura do nariz, bem como a diminuição da porção mucosa dos lábios, levemente mais evidente no lábio inferior, sendo estas alterações progressivas ao longo dos grupos etários estudados e diferente entre os sexos. Assim, concluiu-se ser possível determinar metricamente as modificações que ocorrem na face do indivíduo adulto em função da idade, em específico no que diz respeito a alterações de nariz, lábios e orelhas, visando a compreensão do processo de envelhecimento e fornecendo subsídios para execução da técnica de simulação de progressão de idade da face em imagens.There are many attempts to design the look of a person over time. A widely used resource for this purpose is the simulation of age progression, which is the modification of a person\'s photograph, representing the effect of aging on their appearance through digital image processing or artistic designs. However, the methodologies commonly used are subjective. The absence of a scientifically consolidated methodology capable of simulating changes in the face, through images, highlights the need for studies in the area, in order to support the criminal expert for technical execution, for their application to assist in the Forensic Facial Identification (FFI). The main premise involved in any method of human identification is the \"uniqueness\" of information used as a comparison. For uniqueness means individuality (or non-repeatability) of the studied parameter, which should be measured to perform specific population studies, with some facial features that are maintained over the years. This study evaluated changes in the metric pattern of the face of adult people according to age, in particular with regard to the growth of ears, nose and change in the thickness of the lips, aiming to understand the aging process and the development of methodologies for age progression to estimate in cases of missing persons and fugitives, giving subsidies to perform the face of the aging simulation technique images. Therefore, were used facial images of 700 South Brazilian adults of both sexes, white ancestry and 20 30, 40, 50, 60, 70 and 80 ages, all in the frontal picture, coming from the National Passport System managed by the Federal Police. Results showed a gradual and measurable increase the height of the ears closely related with the increase of the lobes, increased nose width, as well as decreasing mucous portion of the lips slightly more evident in the lower lip, and these incremental changes over the age groups studied and are different between the sexes. Thus, it was found possible to determine metrically changes that occur in the face of adult individuals depending on age, in particular as regards the nose changes, lips and ears, aiming to understand the aging process and providing data to running the technique of simulating the progression of age in face images.
13
Lopes, Daniel Pedro Ferreira. "Face verication for an access control system in unconstrained environment." Master's thesis, Universidade de Aveiro, 2017. http://hdl.handle.net/10773/23395.
Full textAbstract:
Mestrado em Engenharia Eletrónica e TelecomunicaçõesO reconhecimento facial tem vindo a receber bastante atenção ao longo dos últimos anos não só na comunidade cientifica, como também no ramo comercial. Uma das suas várias aplicações e o seu uso num controlo de acessos onde um indivíduo tem uma ou várias fotos associadas a um documento de identificação (também conhecido como verificação de identidade). Embora atualmente o estado da arte apresente muitos estudos em que tanto apresentam novos algoritmos de reconhecimento como melhorias aos já desenvolvidos, existem mesmo assim muitos problemas ligados a ambientes não controlados, a aquisição de imagem e a escolha dos algoritmos de deteção e de reconhecimento mais eficazes. Esta tese aborda um ambiente desafiador para a verificação facial: um cenário não controlado para o acesso a infraestruturas desportivas. Uma vez que não existem condições de iluminação controladas nem plano de fundo controlado, isto torna um cenário complicado para a implementação de um sistema de verificação facial. Esta tese apresenta um estudo sobre os mais importantes algoritmos de detecção e reconhecimento facial assim como técnicas de pré-processamento tais como o alinhamento facial, a igualização de histograma, com o objetivo de melhorar a performance dos mesmos. Também em são apresentados dois métodos para a aquisição de imagens envolvendo a seleção de imagens e calibração da câmara. São apresentados resultados experimentais detalhados baseados em duas bases de dados criadas especificamente para este estudo. No uso de técnicas de pré-processamento apresentadas, foi possível presenciar melhorias até 20% do desempenho dos algoritmos de reconhecimento referentes a verificação de identidade. Com os métodos apresentados para os testes ao ar livre, foram conseguidas melhorias na ordem dos 30%.
Face Recognition has been received great attention over the last years, not only on the research community, but also on the commercial side. One of the many uses of face recognition is its use on access control systems where a person has one or several photos associated to an Identi cation Document (also known as identity veri cation). Although there are many studies nowadays, both presenting new algorithms or just improvements of the already developed ones, there are still many open problems regarding face recognition in uncontrolled environments, from the image acquisition conditions to the choice of the most e ective detection and recognition algorithms, just to name a few. This thesis addresses a challenging environment for face veri cation: an unconstrained environment for sports infrastructures access. As there are no controlled lightning conditions nor controlled background, this makes a di cult scenario to implement a face veri cation system. This thesis presents a study of some of the most important facial detection and recognition algorithms as well as some pre-processing techniques, such as face alignment and histogram equalization, with the aim to improve their performance. It also introduces some methods for a more e cient image acquisition based on image selection and camera calibration, specially designed for addressing this problem. Detailed experimental results are presented based on two new databases created speci cally for this study. Using pre-processing techniques, it was possible to improve the recognition algorithms performances up to 20% regarding veri cation results. With the methods presented for the outdoor tests, performances had improvements up to 30%
14
PINHO, JOSÉ LUIZ BUONOMO DE. "IMAGE QUALITY METRICS FOR FACE RECOGNITION." PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO, 2012. http://www.maxwell.vrac.puc-rio.br/Busca_etds.php?strSecao=resultado&nrSeq=22825@1.
Full textAbstract:
O Reconhecimento Facial é o processo de identificação de uma pessoa a
partir da imagem de sua face. Na forma mais usual, o processo de identificação
consiste em extrair informações dessa imagem e compará-las com informações
relativas a outras imagens armazenadas numa base de dados e por fim indicar na
saída a imagem da base mais similar à imagem de entrada. O desempenho desse
processo está diretamente ligado à qualidade das imagens, tanto das que estão
armazenadas na base de dados, quanto da imagem do indivíduo cuja identidade
está sendo determinada. Por isso, convém que a qualidade das imagens faciais seja
avaliada antes que estas sejam submetidas ao procedimento de reconhecimento.
A maioria dos métodos apresentados até o momento na literatura baseia-se em um
conjunto de critérios, cada um voltado a um atributo isolado da imagem. A
qualidade da imagem é considerada adequada se aprovada por todos os critérios
individualmente. Desconsidera-se, portanto, o efeito cumulativo de diversos
fatores que afetam a qualidade das imagens e, por conseguinte, o desempenho do
reconhecimento facial. Essa monografia propõe uma metodologia para o projeto
de métricas de qualidade de imagens faciais que expressem num único índice o
efeito combinado de diversos fatores que afetam o reconhecimento. Tal índice é
dado por uma função de um conjunto de atributos extraídos diretamente da
imagem. O presente estudo analisa experimentalmente uma função linear e uma
rede neural do tipo back-propagation como alternativas para a estimativa de
qualidade a partir dos atributos. Experimentos conduzidos sobre a base de dados
IMM para o algoritmo de reconhecimento baseado em padrões binários locais
comprovam a o bom desempenho da metodologia.Face Recognition is the process of identifying people based on facial images. In its most usual form the identification procedure consists of extracting information from an input face image and comparing them to the records of other face images stored in a face data base, and finally indicating the most similar one to the input image. The performance of this process is directly dependent on the input image quality, as well as on the images in the data base. Thus, it is important that the quality of a face image is tested before it is given to the recognition procedure, either as a input image or as a new record in the face database. Most methods proposed thus far based on a set of criteria, each one devoted to an isolated attribute. The image quality is considered adequate if approved by all criteria individually. Thus, the cumulative effect of different factors affecting the image quality is no regarded. This dissertation proposes a methodology for the design of quality metrics of facial images that Express in a single scalar the combined effect of multiple factors affecting the quality. Such score is given by a function of attributes extracted directly from the image. This study investigates a linear and a non-linear approach for quality assessment. Experiments conducted upon the IMM face database for a Local Binary Pattern face recognition algorithm demonstrate the good performance of the proposed methodology.
15
Mutelo, Risco Mulwani. "Biometric face image representation and recognition." Thesis, University of Newcastle upon Tyne, 2011. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.548004.
Full text
16
Ribeiro, Ricardo Ferreira. "Face detection on infrared thermal image." Master's thesis, Universidade de Aveiro, 2017. http://hdl.handle.net/10773/23551.
Full textAbstract:
Mestrado em Engenharia Eletrónica e TelecomunicaçõesInfrared cameras or thermal imaging cameras are devices that use infrared radiation to capture an image. This kind of sensors are being developed for almost a century now. They started to be used in the military environment, but at that time it took too long to create a single image. Nowadays, the infrared sensors have reached a whole new technological level and are used for other than military purposes. These sensors are being used for face detection in this thesis. When comparing the use of thermal images regarding color images, it is possible to see advantages and limitations, such as capture images in total darkness and high price, respectively, which will be explored throughout this document. This work proposes the development or adaptation of several methods for face detection on infrared thermal images. The well known algorithm developed by Paul Viola and Michael Jones, using Haar feature-based cascade classi ers, is used to compare the traditional algorithms developed for visible light images when applied to thermal imaging. Three di erent algorithms for face detection are presented. Face segmentation is the rst step in these methods. A method for the segmentation and ltering of the face in the infrared thermal images resulting in a binary image is proposed. In the rst method, an edge detection algorithm is applied to the binary image and the face detection is based on these contours. In the second method, a template matching method is used for searching and nding the location of a template image with the shape of a human head in the binary image. In the last one, a matching algorithm is used. This algorithm correlates a template with the distance transform of the edge image. This algorithm incorporates edge orientation information resulting in the reduction of false detection and the cost variation is limited. The experimental results show that the proposed methods have promising outcome, but the second method is the most suitable for the performed experiments.
As camaras infravermelhas ou as camaras de imagem termica sao dispositivos que usam radiação infravermelha para capturar uma imagem. Este tipo de sensores estao a ser desenvolvidos há quase um século. Começaram a ser usados para fins militares, mas naquela época demorava demasiado tempo para criar uma única imagem. Hoje em dia, os sensores infravermelhos alcançaram um nível tecnológico totalmente novo e são usados para fins além de militares. Esses sensores estão ser usados para detecção facial nesta dissertação. Comparando o uso de imagens térmicas relativamente a imagens coloridas, é possível ver vantagens e limitações, tal como a captura de imagens na escuridão e o preço elevado, respectivamente, que serão exploradas durante este documento. Este trabalho propõe o desenvolvimento ou adaptação de vários métodos para a detecção facial em imagens térmicas. O conhecido algoritmo desenvolvido por Paul Viola e Michael Jones, que utiliza cascatas de classificadores de Haar baseado em características, é usado para comparar os algoritmos tradicionais desenvolvidos para imagens de luz visível quando aplicados a imagens térmicas. São apresentados três métodos diferentes para a detecção facial. A segmentação do rosto e o primeiro passo nestes métodos. E proposto um método para a segmentação e filtragem do rosto nas imagens térmicas que tem como resultado uma imagem binária. No primeiro método, é aplicado um algoritmo de detecção de contornos a imagem binária e a detecção facial é baseada nesses contornos. No segundo método, é usado um método de correspondência de padrões para pesquisar e encontrar a localização de uma imagem padrão com a forma da cabeça humana na imagem binária. No último, é usado um algoritmo de correspondência. Este algoritmo correlaciona um padrão com a transformada de distância da imagem de contornos. Este algoritmo incorpora informações de orientação de contornos que resulta na redução de falsas detecções e a variação do custo é limitada. Os resultados experimentais mostram que os métodos propostos têm resultados promissores, mas o segundo método é o mais adequado para as experiências realizadas.
17
Rana, Santu. "Multilinear analysis of face image ensembles." Thesis, Curtin University, 2010. http://hdl.handle.net/20.500.11937/1662.
Full textAbstract:
Machine based face recognition is an important area of research that has attracted significant attention over the past few decades. Recently, multilinear models of face images have gained prominence as an alternative method for face recognition. Against linear techniques, multilinear models offer the advantage of having more complex models. Against kernel and manifold based non-linear techniques, the advantage lies in having more intuitive and computationally frugal modelling. In this thesis, we present an in-depth analysis and understanding of different properties associated with multilinear analysis and propose three different face recognition algorithms and a unified framework addressing open issues in face recognition.We first propose a face recognition algorithm primarily to address the limitations of the existing multilinear based algorithms in the form of their inability to handle test images that are in unseen conditions. The algorithm is based on the construction of a new representational basis multilinear eigenmodes, enabling representation and classification of faces at unseen conditions. Subsequently, we propose a second algorithm to address the high computational complexity of the first algorithm. We define a set of person-specific bases to represent person-specific images under all variations, and based on this propose an efficient recognition algorithm. Next, we propose a framework of face recognition based on an interpretation of the multilinear analysis as a factor analysis paradigm. We the reformulate all the multilinear based algorithms to link them to a single optimization framework. A theoretical comparison of these algorithms is performed revealing the fundamental differences between them and their applicability in different face recognition scenarios. Experiments performed to compare multilinear analysis based methods to the leading linear and non-linear techniques reveal the superiority of our second algorithm on both measures of recognition accuracy and test speed.Next, we address the issue of inadequate training samples that arise in many application scenarios. We introduce a novel “friendly-hostile” paradigm, in which we propose a mechanism to compensate for the low number of training samples of hostile people by learning the structure of face images from a large training set of the friendly people. The formulation is built on a novel synthesis paradigm that is based on the unique factorization properties of the multilinear analysis. Experimental results show significant performance gain in comparison to the conventional methods. We also discuss issues concerning unbalanced datasets, wherein some people may be under-represented than others in the training set. This results in different apriori bias per class, affecting conventional recognition algorithms. Based on theory and experiments we demonstrate that our algorithm does not get affected by any such imbalance in bias and produces consistent performance in all situations.
18
Nyland, Robert Scott. "The Gratification Niches of Internet Social Networking, E-mail, and Face-to-face Communication." Diss., CLICK HERE for online access, 2007. http://contentdm.lib.byu.edu/ETD/image/etd2151.pdf.
Full text
19
Bryan, Angela E. "Face-to-face communication in Australian workplaces : a social rules approach /." St. Lucia, Qld, 2002. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe16510.pdf.
Full text
20
Pereira, Diogo Camara. "Face recognition using infrared imaging." Thesis, Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2002. http://library.nps.navy.mil/uhtbin/hyperion-image/02Dec%5FPereira.pdf.
Full textAbstract:
Thesis (Electrical Engineer and M.S. in Electrical Engineering)--Naval Postgraduate School, December 2002.Thesis advisor(s): Monique P. Fargues, Gamani Karunasiri, Roberto Cristi. Includes bibliographical references (p. 93-95). Also available online.
21
Ebrahimpour-Komleh, Hossein. "Fractal techniques for face recognition." Thesis, Queensland University of Technology, 2006. https://eprints.qut.edu.au/16289/1/Hossein_Ebrahimpour-Komleh_Thesis.pdf.
Full textAbstract:
Fractals are popular because of their ability to create complex images using only several simple codes. This is possible by capturing image redundancy and presenting the image in compressed form using the self similarity feature. For many years fractals were used for image compression. In the last few years they have also been used for face recognition. In this research we present new fractal methods for recognition, especially human face recognition.
This research introduces three new methods for using fractals for face recognition, the use of fractal codes directly as features, Fractal image-set coding and Subfractals. In the first part, the mathematical principle behind the application of fractal image codes for recognition is investigated. An image Xf can be represented as Xf = A x Xf + B which A and B are fractal parameters of image Xf . Different fractal codes can be presented for any arbitrary image. With the defnition of a fractal transformation, T(X) = A(X - Xf ) + Xf , we can define the relationship between any image produced in the fractal decoding process starting with any arbitrary image X0 as Xn = Tn(X) = An(X - Xf ) + Xf . We show that some choices for A or B lead to faster convergence to the final image.
Fractal image-set coding is based on the fact that a fractal code of an arbitrary gray-scale image can be divided in two parts - geometrical parameters and luminance parameters. Because the fractal codes for an image are not unique, we can change the set of fractal parameters without significant change in the quality of the reconstructed image. Fractal image-set coding keeps geometrical parameters
the same for all images in the database. Differences between images are captured in the non-geometrical or luminance parameters - which are faster to compute. For recognition purposes, the fractal code of a query image is applied to all the images in the training set for one iteration. The distance between an image and the result after one iteration is used to define a similarity measure between this image and the query image.
The fractal code of an image is a set of contractive mappings each of which transfer a domain block to its corresponding range block. The distribution of selected domain blocks for range blocks in an image depends on the content of image and the fractal encoding algorithm used for coding. A small variation in a part of the input image may change the contents of the range and domain blocks in the fractal encoding process, resulting in a change in the transformation parameters in the same part or even other parts of the image. A subfractal is a set of fractal codes related to range blocks of a part of the image. These codes are calculated to be independent of other codes of the other parts of the same image. In this case the domain blocks nominated for each range block must be located in the same part of the image which the range blocks come from.
The proposed fractal techniques were applied to face recognition using the MIT and XM2VTS face databases. Accuracies of 95% were obtained with up to 156 images.
22
Ebrahimpour-Komleh, Hossein. "Fractal techniques for face recognition." Queensland University of Technology, 2006. http://eprints.qut.edu.au/16289/.
Full textAbstract:
Fractals are popular because of their ability to create complex images using only several simple codes. This is possible by capturing image redundancy and presenting the image in compressed form using the self similarity feature. For many years fractals were used for image compression. In the last few years they have also been used for face recognition. In this research we present new fractal methods for recognition, especially human face recognition.
This research introduces three new methods for using fractals for face recognition, the use of fractal codes directly as features, Fractal image-set coding and Subfractals. In the first part, the mathematical principle behind the application of fractal image codes for recognition is investigated. An image Xf can be represented as Xf = A x Xf + B which A and B are fractal parameters of image Xf . Different fractal codes can be presented for any arbitrary image. With the defnition of a fractal transformation, T(X) = A(X - Xf ) + Xf , we can define the relationship between any image produced in the fractal decoding process starting with any arbitrary image X0 as Xn = Tn(X) = An(X - Xf ) + Xf . We show that some choices for A or B lead to faster convergence to the final image.
Fractal image-set coding is based on the fact that a fractal code of an arbitrary gray-scale image can be divided in two parts - geometrical parameters and luminance parameters. Because the fractal codes for an image are not unique, we can change the set of fractal parameters without significant change in the quality of the reconstructed image. Fractal image-set coding keeps geometrical parameters
the same for all images in the database. Differences between images are captured in the non-geometrical or luminance parameters - which are faster to compute. For recognition purposes, the fractal code of a query image is applied to all the images in the training set for one iteration. The distance between an image and the result after one iteration is used to define a similarity measure between this image and the query image.
The fractal code of an image is a set of contractive mappings each of which transfer a domain block to its corresponding range block. The distribution of selected domain blocks for range blocks in an image depends on the content of image and the fractal encoding algorithm used for coding. A small variation in a part of the input image may change the contents of the range and domain blocks in the fractal encoding process, resulting in a change in the transformation parameters in the same part or even other parts of the image. A subfractal is a set of fractal codes related to range blocks of a part of the image. These codes are calculated to be independent of other codes of the other parts of the same image. In this case the domain blocks nominated for each range block must be located in the same part of the image which the range blocks come from.
The proposed fractal techniques were applied to face recognition using the MIT and XM2VTS face databases. Accuracies of 95% were obtained with up to 156 images.
23
Pershits, Edward. "Recognition of Face Images." Thesis, University of North Texas, 1994. https://digital.library.unt.edu/ark:/67531/metadc277785/.
Full textAbstract:
The focus of this dissertation is a methodology that enables computer systems to classify different up-front images of human faces as belonging to one of the individuals to which the system has been exposed previously. The images can present variance in size, location of the face, orientation, facial expressions, and overall illumination. The approach to the problem taken in this dissertation can be classified as analytic as the shapes of individual features of human faces are examined separately, as opposed to holistic approaches to face recognition. The outline of the features is used to construct signature functions. These functions are then magnitude-, period-, and phase-normalized to form a translation-, size-, and rotation-invariant representation of the features. Vectors of a limited number of the Fourier decomposition coefficients of these functions are taken to form the feature vectors representing the features in the corresponding vector space. With this approach no computation is necessary to enforce the translational, size, and rotational invariance at the stage of recognition thus reducing the problem of recognition to the k-dimensional clustering problem. A recognizer is specified that can reliably classify the vectors of the feature space into object classes. The recognizer made use of the following principle: a trial vector is classified into a class with the greatest number of closest vectors (in the sense of the Euclidean distance) among all vectors representing the same feature in the database of known individuals. A system based on this methodology is implemented and tried on a set of 50 pictures of 10 individuals (5 pictures per individual). The recognition rate is comparable to that of most recent results in the area of face recognition. The methodology presented in this dissertation is also applicable to any problem of pattern recognition where patterns can be represented as a collection of black shapes on the white background.
24
Bilson, Amy Jo. "Image size and resolution in face recognition /." Thesis, Connect to this title online; UW restricted, 1987. http://hdl.handle.net/1773/9166.
Full text
25
Omar, Luma Qassam Abedalqader. "Face liveness detection under processed image attacks." Thesis, Durham University, 2018. http://etheses.dur.ac.uk/12812/.
Full textAbstract:
Face recognition is a mature and reliable technology for identifying people. Due to high-definition cameras and supporting devices, it is considered the fastest and the least intrusive biometric recognition modality. Nevertheless, effective spoofing attempts on face recognition systems were found to be possible. As a result, various anti-spoofing algorithms were developed to counteract these attacks. They are commonly referred in the literature a liveness detection tests. In this research we highlight the effectiveness of some simple, direct spoofing attacks, and test one of the current robust liveness detection algorithms, i.e. the logistic regression based face liveness detection from a single image, proposed by the Tan et al. in 2010, against malicious attacks using processed imposter images. In particular, we study experimentally the effect of common image processing operations such as sharpening and smoothing, as well as corruption with salt and pepper noise, on the face liveness detection algorithm, and we find that it is especially vulnerable against spoofing attempts using processed imposter images. We design and present a new facial database, the Durham Face Database, which is the first, to the best of our knowledge, to have client, imposter as well as processed imposter images. Finally, we evaluate our claim on the effectiveness of proposed imposter image attacks using transfer learning on Convolutional Neural Networks. We verify that such attacks are more difficult to detect even when using high-end, expensive machine learning techniques.
26
Feng, Guo Can. "Face recognition using virtual frontal-view image." HKBU Institutional Repository, 1999. http://repository.hkbu.edu.hk/etd_ra/267.
Full text
27
Duffner, Stefan. "Face image analysis with convolutional neural networks." [S.l. : s.n.], 2007. http://nbn-resolving.de/urn:nbn:de:bsz:25-opus-48350.
Full text
28
Bartlett, Marian Stewart. "Face image analysis by unsupervised learning and redundancy reduction /." Diss., Connect to a 24 p. preview or request complete full text in PDF format. Access restricted to UC campuses, 1998. http://wwwlib.umi.com/cr/ucsd/fullcit?p9907603.
Full text
29
Moore, Thomas Brendan. "Learning Geometry-Free Face Re-lighting." Master's thesis, University of Central Florida, 2007. http://digital.library.ucf.edu/cdm/ref/collection/ETD/id/3353.
Full textAbstract:
The accurate modeling of the variability of illumination in a class of images is a fundamental problem that occurs in many areas of computer vision and graphics. For instance, in computer vision there is the problem of facial recognition. Simply, one would hope to be able to identify a known face under any illumination. On the other hand, in graphics one could imagine a system that, given an image, the illumination model could be identified and then used to create new images. In this thesis we describe a method for learning the illumination model for a class of images. Once the model is learnt it is then used to render new images of the same class under the new illumination. Results are shown for both synthetic and real images. The key contribution of this work is that images of known objects can be re-illuminated using small patches of image data and relatively simple kernel regression models. Additionally, our approach does not require any knowledge of the geometry of the class of objects under consideration making it relatively straightforward to implement. As part of this work we will examine existing geometric and image-based re-lighting techniques; give a detailed description of our geometry-free face re-lighting process; present non-linear regression and basis selection with respect to image synthesis; discuss system limitations; and look at possible extensions and future work.M.S.
School of Electrical Engineering and Computer Science
Engineering and Computer Science
Computer Science MS
30
Hinze, Carl Graham. "Re-thinking 'face' : pursuing an emic-etic understanding of Chinese Mian and Lian and English face /." St. Lucia, Qld, 2002. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe16544.pdf.
Full text
31
Schaffer, Paul. "Grain refiner fade in aluminium alloys /." [St. Lucia, Qld.], 2004. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe18566.pdf.
Full text
32
Alayza, Prager Cristina. "La imagen en primer plano: La pasión de Juana de Arco y el poder del rostro." Pontificia Universidad Católica del Perú, 2016. http://repositorio.pucp.edu.pe/index/handle/123456789/119568.
Full textAbstract:
Two peculiarities of the film The Passion of Joan of Arc (Carl Theodor Dreyer,1928) are explored in this essay: its unusual close-ups and its strongly expressivecharacter. We start from the idea that both characteristics are closely related andwe seek to clarify this relationship making use of Gilles Deleuze’s affection-image and Georges Didi-Huberman’s dialectical image.Dos particularidades de la película La pasión de Juana de Arco (Carl Theodor Dreyer, 1928) son exploradas en este ensayo: sus insólitos primeros planos y sucarácter fuertemente expresivo. Partimos de la idea de que ambas característicasestán íntimamente relacionadas y buscamos esclarecer esta relación sirviéndonos de la imagen-afección de Gilles Deleuze y de la imagen dialéctica de Georges Didi-Huberman.
33
Bennetts, Jennifer. "The identification and characterisation of novel genes in development /." [St. Lucia, Qld.], 2006. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe19375.pdf.
Full text
34
Buchala, Samarasena. "Computational analysis of face images." Thesis, University of Hertfordshire, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.431938.
Full text
35
Akinbola, Akintunde A. "Estimation of image quality factors for face recognition." Morgantown, W. Va. : [West Virginia University Libraries], 2005. https://eidr.wvu.edu/etd/documentdata.eTD?documentid=4308.
Full textAbstract:
Thesis (M.S.)--West Virginia University, 2005.Title from document title page. Document formatted into pages; contains vi, 56 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 52-56).
36
Li, Yang. "Face Recognition Based on Histogram And Spin Image." Thesis, University of York, 2007. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.485831.
Full textAbstract:
This thesis presents our research work on shape-based human face recognition exploiting curvature-based histogram and spin image. Instead of the popular 2D shape information represented by fiducial points, the novelty here is the use of 2.5D shape information obtained by shape-from-shading (SFS). Though surface normals generated by performing shape-from-shading on objects are not widely accepted as a precise shape representation for face recognition purposes, recent research in shape-from-shading [Ragheb and Hancock, 2003] [prados et aI., 2006] [Castelan, 2006] [Smith and Hancock, 2006] has made it possible to recover fairly accurate shape information under various conditions from face images in the real world. These contributions make our face recognition approaches based on 2.5D shape recovered from a single image possi.ble. Chapter 2 is a thorough review of the existing literature in the areas pf surface reconstruction using shape-from-shading, appearance-based and model-based face recognition on 2D and 3D data, and histogram-based image representation and recognition. The literature is pretty sparse on works on shape-from-shading in the face recognition area, however there are plenty of approaches based on 2D images ~ and 3D range data. With accurate height map being recovered from single face image [Prados et aI., 2006] [Castelan, 2006] and statistical model being proposed to recover surface normals [Smith and Hancock, 2006], we have enough 2.5D shape infonnation recovered' from 2D image based on which we can perfonn face recognition. In Chapter 3, we present our curvature-based histogram appro,ach as our first contribution, which employs principal curvature infonnation calculated from the Hessian matrix based on the recovered surface nonnals. Generalized entropies are introduced as similarity measurements, which give stable performance especially when the number of relative images varies. While curvature-based histogram proves to be a fairly good face recognition approach with concise representation and easy comparison, its performance is not always stable when applied to different databases. Therefore more advanced facerecognition approach is required to bring better and more stable identification results. In Chapter 4, we derive patch-based spin image as a local shape representation from the idea of global curvature-based histogram in Chapter 3. This representation is inspired by [Johnson and Hebert, 1999] and adapted in this thesis as a solution to face recognition problem. Instead of using 3D range data, the estimated height map reconstructed by shape-from-shading is employed in the spin image construction. Also the mean needle map model is used as the preprocessing to correct the errors and noise that exists in the surface nonnal estimates. Face surface is segmented into small patches and the spin image corresponding to the surface is composed of histograms constructed on each surface patch. In Appendix A, we propose dual spin image to address the difficulties of recognizing face in rotation, among which the two major problems of point correspondence and point occlusion are of particular importance. Therefore we propose the idea of neighbour area spin image to construct the pointwise neighbourhood surface feature collection and use the ratio of projected distances to relative angles to alleviate the errors introduced by surface rotation. We also present the linear model and the finite Gaussian mixture model to approximate novel dual spin image using the existing dual spin images. Face recognition is performed based on the model parameters. .The work in this thesis suggests that 2.5D shape feature recovered by shapefrom- shading can be used for the purpose offace recognition. Also the appearancebased approaches derived from histogram are effective for face recognition. The result suggests that shape information recovered from single image is sufficient for face recognition based on the condition that shape-from:-shading can successfully recover the surface normal fields and the height map from the image.
37
Song, Qing. "Features and statistical classifiers for face image analysis." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2000. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape4/PQDD_0035/NQ62459.pdf.
Full text
38
CastelaÌn, Mario. "Face shape recovery from a single image view." Thesis, University of York, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.431579.
Full text
39
Dasgupta, Sankarshan. "3D Face Reconstruction From Front And Profile Image." University of Dayton / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1626689472561706.
Full text
40
Du, Shan. "Image-based face recognition under varying pose and illuminations conditions." Thesis, University of British Columbia, 2008. http://hdl.handle.net/2429/2814.
Full textAbstract:
Image-based face recognition has attained wide applications during the past decades in commerce and law enforcement areas, such as mug shot database matching, identity authentication, and access control. Existing face recognition techniques (e.g., Eigenface, Fisherface, and Elastic Bunch Graph Matching, etc.), however, do not perform well when the following case inevitably exists. The case is that, due to some variations in imaging conditions, e.g., pose and illumination changes, face images of the same person often have different appearances. These variations make face recognition techniques much challenging. With this concern in mind, the objective of my research is to develop robust face recognition techniques against variations.
This thesis addresses two main variation problems in face recognition, i.e., pose and illumination variations. To improve the performance of face recognition systems, the following methods are proposed: (1) a face feature extraction and representation method using non-uniformly selected Gabor convolution features, (2) an illumination normalization method using adaptive region-based image enhancement for face recognition under variable illumination conditions, (3) an eye detection method in gray-scale face images under various illumination conditions, and (4) a virtual pose generation method for pose-invariant face recognition. The details of these proposed methods are explained in this thesis. In addition, we conduct a comprehensive survey of the existing face recognition methods. Future research directions are pointed out.
41
Ness, Hayley. "Improving composite images of faces produced by eyewitnesses." Thesis, University of Stirling, 2003. http://hdl.handle.net/1893/12555.
Full textAbstract:
When a witness views a crime, they are often asked to construct a facial likeness, or composite of the suspect. These composites are then used to stimulate recognition from someone who is familiar with the suspect. Facial composites are commonly used in large scale cases e. g. Jill Dando, Yorkshire Ripper, however a great deal of research has indicated that facial composites perform poorly and often do not portray an accurate likeness of the suspect. This thesis therefore examined methods of improving facial composites. In particular, it examined methods of increasing the likeness portrayed in composites, both during construction and at test. Experiments 1 to 3 examined the effectiveness of a new three-quarter-view database in PROfit. Experiment 1 examined whether the presentation of composites in a three-quarter- view composite will aid construction. Participant-witnesses were exposed to all views of a target and the results indicated that three-quarter-view composites performed as well as full-face composites but not better. Experiments 2 and 3 then examined whether the presentation of two composites (one in a full-face view and the other in a three-quarter-view) from the same participant-witness would increase performance above the level observed for a single composite. The results revealed that two views were better than one. In addition, experiment 3 examined the issue of encoding specificity and viewpoint dependency in composite construction. All participant-witnesses were exposed to either one view of a target (full-face or three-quarter) or all views and they were asked to construct both a full-face and a three-quarter- view composite. The results indicated that performance was better when all views of a face had been presented. When a target had been seen in a three-quarter-view, it was better to construct a three-quarter-view composite. However, when a target had been seen in a full-face view, performance for both full-face and three-quarter composites was poor. Experiments 4 to 8 examined whether the presentation of composites from multiple witnesses would increase performance. The results revealed that morphing composites from four different witnesses (4-Morphs) resulted in an image that performed as well as or better than the best single image. Further experimentation attempted to examine why multiple composites performed well. In particular, it was asked whether multiple composites performed well because they contained varied information or whether they performed well because they just contained more information. Multiple composites from both single and multiple witnesses using the same (PROfit) and different (PROfit, E-FIT, Sketch, EvoFIT) composite techniques were compared and the results revealed that multiple composites performed well because they contained different memorial representations. This combination of different memorial representations appeared to result in an image that was closer to the ideal, or prototypical image. Experiments 9 to 12 examined the relationship between verbal descriptions and composite quality. The results revealed that there was no clear relationship between the amount of description provided, the accuracy of the description and performance of the resulting composite. Further experimentation examined whether the presentation of a composite and a description would increase performance above the level observed for a single composite. The results revealed that the combination of a description and a composite from the same participant-witness did increase performance. This indicated that descriptions and composites might contain differing amounts and types of featural and configurational information. Both the theoretical and practical implications of these results are discussed. Experiments 1, 2 and 3 of this thesis have been submitted for publication. Ness, H., Hancock, P. J. B., Bowie, L. and Bruce, V. Are two views better than one? A study investigating recognition of full-face and three-quarter-view composites. Applied Cognitive Psychology. Experiment 4 of this thesis appears in Bruce, V., Ness, H., Hancock, P. J. B., Newman, C. and Rarity, J. (2002). Four heads are better than one: combining face composites results yields improvements in face likeness. Journal of Applied Psychology. 87 (5), 894-902. Other Publications Frowd, C. D., Carson, D., Ness, H., Richardson, J., Morrison, L., McLanaghan, S., Hancock, P. J. B. Evaluating Facial Composite Systems. Manuscript accepted for publication in Psychology, Crime and Law. Frowd, C. D., Carson, D., Ness, H., McQuiston, D., Richardson, J., Baldwin, H., Hancock, P. J. B. Contemporary Composite Techniques: The impact of a forensically relevant target delay. Manuscript accepted for publication in Legal and Criminological Psychology.
42
Bicudo, Alexandre Luiz da Costa [UNESP]. "Avaliação ultra-sonográfica convencional e dopplerfluxométrica de ovário de cadelas, durante a fase folicular do ciclo estral." Universidade Estadual Paulista (UNESP), 2008. http://hdl.handle.net/11449/99370.
Full textAbstract:
Made available in DSpace on 2014-06-11T19:30:11Z (GMT). No. of bitstreams: 0
Previous issue date: 2008-06-05Bitstream added on 2014-06-13T20:07:36Z : No. of bitstreams: 1
bicudo_alc_dr_botfmvz.pdf: 403167 bytes, checksum: 90885c2effdcd34acfe70d5474be4d99 (MD5)Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Nenhuma avaliação clínica ou a receptividade à cobertura pelo macho é suficientemente rigorosa para detectar com exatidão a ocorrência e o dia das ovulações nas cadelas. Algumas cadelas podem ovular no quinto dia do ciclo estral, enquanto outras, ovulam mais tardiamnente, como no trigésimo dia. A técnica mais recente para determinar a ovulação na cadela é a exploração ovariana por ultrasonografia. Portanto, o presente trabalho tem por objetivo avaliar o grau de confiabilidade e rigor do exame ultra-sonográfico, em modo B e a possibilidade de avaliações dopplerfluxométricas de artérias intra-ovarianas para uma detecção segura da ovulação nas cadelas. Para tanto foram utilizadas 5 cadelas, SRD, adultas, com peso de até 10 Kg e em início de fase folicular do ciclo estral. Durante a fase de proestro e estro foram realizados exames ultra-sonográficos (Logic 3, GE) com transdutor de 10Mhz, seriados, com no máximo 48horas de intervalo. Foram realizados cortes sagitais e transversais para se detectar e mensurar os folículos ovarianos. A partir da detecção do colapso folicular ou da substituição da imagem circunscrita anecóica de parede regular por uma estrutura hipoecogênica de contornos irregulares foi monitorado o início de formação luteal. Na fase luteal do ciclo foram realizadas mensurações dos corpos lúteos. Foram realizadas também imagens com o doppler colorido e espectral. Para avaliação do Doppler espectral o cursor foi colocado sobre a artéria ângulo insonação ≤ 60º, procedendo-se a avaliação da velocidade máxima do pico sistólico (VPS), dos índices de resistividade (IR) e pulsatilidade (IP). Foi realizada uma análise descritiva dos resultados obtidos. Pela imagem ultra-sonográfica convencional foi possível detectar 100% dos ovários das cadelas estudadas, possibilitando acompanhar o desenvolvimento folicular, colapso e formação de corpo lúteo em 85,7% dos folículos. A presença de corpos lúteos foi identificada em 100% das cadelas. A média da velocidade máxima do pico sistólico foi de 26,41cm/s, observada desde 120horas até 24horas após a ovulação. O IR apresentou declínio no dia da ovulação ou 24horas após, e o seu valor variou de 0,44 – 0,59 neste período. O IP apresentou muitas oscilações durante o proestro e estro, apresentando declínio desde 120horas antes até 24horas após a ovulação e seu valor nesse período foi de 0,57-1,1.
No clinical evaluation or the receptivity to the male can be rigorous enough to detect precisely the occurrence and the day of the ovulations in dogs. Some dogs can ovulate on the fifth day of the estral cycle, while others, ovulate later, on the thirtieth day. The latest technique to determine the ovulation in dogs is the ovarian exploration with ultrasound. Therefore, the present work evaluates the degree of dependability and accuracy of the ultrasound examination, in B mode and the possibility of dopplerfluxometric evaluations of intra-ovarian arteries for a safe detection of the ovulation in dogs. Therefore 5 mongrel adult bitches, with weight of up to 10 kg and at the onset of follicular phase, were used. During proestrus and estrus ultrasound examinations (Logic 3, GE) with transducer of 10Mhz were carried out in series with the maximum of 48-hour intervals. Sagittal and transversal cuts were taken to detect and measure the ovarian follicules. From the detection of the folicular collapse or the substitution of the anecoiccircumscribed image of regular wall for a hipoecogenic structure of irregular outline was monitored the beginning of luteal formation. In the luteal phase of the cycle measurements of the luteum bodies were carried out. Images with the spectral doppler had been carried also through colorful and. For evaluation of the spectral Doppler the cursor was placed on the artery angle insonation = 60º, proceeding it evaluation from the maximum speed of the systolic peak (SSP), of the resistivity rate (RR) and pulsatility rate (PR). A descriptive analysis of the results was carried out. For the conventional ultrasound image it was possible to detect 100% of the ovarius of the studied dogs, making possible to follow the development of the follicule, collapse and formation of luteum body in 85.7% of the follicules. The presence of luteum bodies was identified in 100% of the dogs. The average of the maximum speed of the systolic peak was of 26.41cm/s, observed since 120 hours until 24hours after the ovulation. RR presented decline on the day of the ovulation or 24 hours, and its value varied 0.44 - 0.59 in this period. The PR presented many oscillations during proestrus and estrus, presenting decline from 120 hours before until 24horas after the ovulation and its value in this period was of 0.57-1.1.
43
Shin, Joo-Heon. "TOPOLOGICAL PROPERTIES OF A NETWORK OF SPIKING NEURONS IN FACE IMAGE RECOGNITION." VCU Scholars Compass, 2010. http://scholarscompass.vcu.edu/etd/2133.
Full textAbstract:
We introduce a novel system for recognition of partially occluded and rotated images. The system is based on a hierarchical network of integrate-and-fire spiking neurons with random synaptic connections and a novel organization process. The network generates integrated output sequences that are used for image classification. The network performed satisfactorily given appropriate topology, i.e. the number of neurons and synaptic connections, which corresponded to the size of input images. Comparison of Synaptic Plasticity Activity Rule (SAPR) and Spike Timing Dependant Plasticity (STDP) rules, used to update connections between the neurons, indicated that the SAPR gave better results and thus was used throughout. Test results showed that the network performed better than Support Vector Machines. We also introduced a stopping criterion based on entropy, which significantly shortened the iterative process while only slightly affecting classification performance.
44
Boyd, Michelle Anne. "The Impact of Web-Based Tutorials in One Corporation's Transition to a Blended Learning Environment." Diss., CLICK HERE for online access, 2008. http://contentdm.lib.byu.edu/ETD/image/etd2429.pdf.
Full text
45
Kafetzi, Evi. "L'Ethos dans l'Argumentation : le cas du face à face Sarkozy / Royal 2007." Thesis, Université de Lorraine, 2013. http://www.theses.fr/2013LORR0053/document.
Full textAbstract:
En quête d'efficacité et d'influence, tout candidat aux électionsprésidentielles tente de se fabriquer et de donner à voir une image de soi conformeaux attentes des électeurs concernant le profil d'un futur chef d'État. Cette imagede soi séduisante construite à travers le discours, appelée ethos en rhétorique, faitpartie intégrante de l'argumentation au même titre que ses autres composantes, àsavoir le logos et le pathos. Le discours politique, en tant que porteur d'importantsenjeux, est le terrain de construction identitaire par excellence.Ce travail explore les stratégies communicatives dans l'activitéargumentative qu'est le débat politique télévisé. Les données sont constituées parle face à face télévisé du 2 mai 2007 entre Nicolas Sarkozy et Ségolène Royal, àla veille du deuxième tour des élections présidentielles françaises.Je me propose dans ce travail de dégager les règles et les mécanismes surlesquels repose la fabrication d'une image de soi télévisuelle par les praticiens dela persuasion que sont les hommes et les femmes politiques, afin de parvenir àleurs fins. Les outils langagiers dont les deux adversaires se servent lors du dueltélévisé en question ici, pour nous servir une image de soi conforme au modèleprésidentiel « idéal » sont analysés un par un. Ainsi, ayant une meilleureconnaissance des coulisses de la rhétorique audiovisuelle, l'électeur-téléspectateurdevient maître de sa décision et responsable de son choix, et apprend à se méfierdes sentiments et des impressions que lui inspirent les praticiens de la persuasionIn search of effectiveness and influence, every candidate who stands forpresidential elections attempts to create and give to the audience a self-imageconsistent with the elector's expectations concerning a future head of state'sprofile. This attractive self-image created through discourse, called ethos inrhetoric, is an integral part of argumentation, as well as its other components,logos and pathos. Political discourse, as a vector of important stakes, constitutesthe ground of identity construction par excellence.This work explores communication strategies in argumentation activity,and particularly in televised political debate. The data is constituted by thetelevised face to face of the 2nd of May 2007 between Nicolas Sarkozy andSégolène Royal, at the eve of the second ballot of the French presidentialelection.What I propose in this work is to draw up the rules and mechanisms thatgovern the making of one's televised self-image by politicians, spin doctors, inorder to achieve their ends. I propose to analyse, one by one, the linguistic toolsthat the two opponents use in order to give the audience a self-image consistentwith an ideal presidential model, during the televised duel that we're studyinghere. In this way, having a better knowledge of what goes on behind the scenesof audiovisual rhetoric, the elector-televiewer becomes master of his decisionand has the control of his choice and learns to beware of feelings andimpressions inspired by the professionals of persuasion
46
Chen, Shaokang. "Robust discriminative principal component analysis for face recognition /." [St. Lucia, Qld.], 2005. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe18934.pdf.
Full text
47
Costen, Nicholas Paul. "Spatial frequencies and face recognition." Thesis, University of Aberdeen, 1994. http://digitool.abdn.ac.uk/R?func=search-advanced-go&find_code1=WSN&request1=AAIU069146.
Full textAbstract:
If face images are degraded by spatial quantisation there is a non-linear acceleration of the decline of recognition accuracy as block-size increases, suggesting recognition requires a critical minimum range of object spatial frequencies. These may define the facial configuration, reflecting the structural properties allowing differentiation of faces. Experiment 1 measured speed and accuracy of recognition of six fronto-parallel faces shown with 11, 21 and 42 pixels/face, produced by quantisation, a Fourier low-pass filter and Gaussian blurring. Performance declined with image quality in a significant, non-linear manner, but faster for the quantised images. Experiment 2 found some of this additional decline was due to frequency-domain masking. Experiment 3 compared recognition for quantised, Fourier low-pass and high-pass versions, recognition was only impaired when the frequency limit exceeded the range 4.5-12.5 cycles/face. Experiment 4 found this was not due to contrast differences. Experiments 5, 6 and 7 used octave band-pass filters centred on 4.14, 9.67 and 22.15 cycles/face, varying view-point for both sequential matching and recognition. The spatial frequency effect was not found for matching, but was for recognition. Experiment 8 also measured recognition of band-passed images, presented with octave bands centred on 2.46-50.15 cycles/face and at 0-90 degrees from fronto-parallel. Spatial frequency effects were found at all angles, with best performance for semi-profile images and 11.10 cycles/face. Experiment 9 replicated this, with perceptually equal contrasts and the outer facial contour removed. Modeling showed this reflected a single spatial-frequency channel two octaves wide, centred on 9 cycles/face. Experiment 10 measured response time for successive matching of faces across a size-disparity, finding an asymmetrical effect.
48
Gunn, Steve R. "Dual active contour models for image feature extraction." Thesis, University of Southampton, 1996. https://eprints.soton.ac.uk/250089/.
Full textAbstract:
Active contours are now a very popular technique for shape extraction, achieved by minimising a suitably formulated energy functional. Conventional active contour formulations suffer difficulty in appropriate choice of an initial contour and values of parameters. Recent approaches have aimed to resolve these problems, but can compromise other performance aspects. To relieve the problem in initialisation, an evolutionary dual active contour has been developed, which is combined with a local shape model to improve the parameterisation. One contour expands from inside the target feature, the other contracts from the outside. The two contours are inter-linked to provide a balanced technique with an ability to reject weak’local energy minima. Additionally a dual active contour configuration using dynamic programming has been developed to locate a global energy minimum and complements recent approaches via simulated annealing and genetic algorithms. These differ from conventional evolutionary approaches, where energy minimisation may not converge to extract the target shape, in contrast with the guaranteed convergence of a global approach. The new techniques are demonstrated to extract successfully target shapes in synthetic and real images, with superior performance to previous approaches. The new technique employing dynamic programming is deployed to extract the inner face boundary, along with a conventional normal-driven contour to extract the outer face boundary. Application to a database of 75 subjects showed that the outer contour was extracted successfully for 96% of the subjects and the inner contour was successful for 82%. This application highlights the advantages new dual active contour approaches for automatic shape extraction can confer.
49
Gallão, Simone [UNESP]. "Dos modelos de gesso ás imagens tridimensionais." Universidade Estadual Paulista (UNESP), 2010. http://hdl.handle.net/11449/104484.
Full textAbstract:
Made available in DSpace on 2014-06-11T19:33:22Z (GMT). No. of bitstreams: 0
Previous issue date: 2010-09-13Bitstream added on 2014-06-13T19:44:07Z : No. of bitstreams: 1
gallao_s_dr_arafo.pdf: 1124962 bytes, checksum: d992196fadc340d0f4d06befe7cd817a (MD5)Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Esta pesquisa foi desenvolvida em três estudos que tiveram como objetivos: pesquisar a presença de bactérias e fungos em modelos de gesso; avaliar a precisão e a reprodutibilidade de medidas dentárias lineares em modelos de gesso originais e duplicados utilizando paquímetro digital e em modelos digitais medidos com o Software Brasileiro O3d e; correlacionar medidas faciais, em indivíduos leucodermas, de 12 a 17 anos, com medidas transversas das arcadas em oclusão normal e avaliar diferenças entre os gêneros. Da amostra total de 78 pares de modelos, foram utilizados 44 pares de modelos de gesso para o primeiro estudo, armazenados desde os anos 1980 e 1981. O material para a pesquisa de micro-organismos foi colhido com auxílio de swab e semeado em culturas de Infusão de Cérebro-Coração (BHI) e Caldo Sabouraud Glicose para posterior plaqueamento e identificação das colônias. Para o segundo estudo outros 34 pares de modelos de gesso foram selecionados, preenchendo os seguintes requisitos de inclusão: presença de todos os dentes permanentes de 1° molar a 1° molar; oclusão normal; sem coroas protéticas; nenhum tratamento ortodôntico prévio; apinhamentos ou espaçamentos menores ou iguais a 2 milímetros. Os modelos originais e os duplicados foram medidos com paquímetro digital (Digimess®). Os modelos de gesso duplicados foram escaneados a laser e medidos pelo Sistema O3d (Widialabs®). Dos modelos utilizados no segundo, foram selecionados 21 pares de modelos de gesso para o terceiro estudo, por apresentarem medidas cefalométricas que foram consultadas e tabuladas. No primeiro estudo ocorreu, em 25% dos modelos, crescimento bacteriano de Estafilococos coagulase negativa; Micrococcus sp.; Bacillus sp. e bacilos gram-negativos não fermentadores e, em 5,7%, crescimento fúngico de Cladophialophora sp.; Trichosporon beigelii; Aspergillus flavus e Aureobasidium pullulans...
This research was developed in three studies that had as objectives: search the presence of bacteria and fungi on dental models.; evaluate the accuracy and reproducibility of linear dental measures in original and duplicated gypsum models using digital caliper and in digital models measured by O3d Brazilian Software and correlate facial measures, in Caucasian individuals, between 12 and 17 years old, with transverse measures of normal occlusion arches and evaluate gender differences. From the original sample of 78 pairs of models, 44 pairs of dental models stored since 1980 and 1981 were used for the first study. The material for the microbiological search was collected with swab and cultivated in Brain-Heart Infusion cultures (BHI) and Sabouraud Dextrose Broth for subsequent plating and colonies identification. For the second study 34 pairs of dental models were selected following the inclusion requirements: presence of all the permanent teeth from 1st molar to 1st molar; normal occlusion; with no prosthetic crowns; no previous orthodontic treatment; 2 millimeters or less crowdings or smaller spacings. The original and the duplicated models were measured with digital caliper (Digimess™). The duplicated dental models were laser scanned and measured by O3d software (Widialabs™). From the models used for the second study 21 pairs of gypsum models were selected for the third study, because they had cephalometric measures that were consulted and tabbed. In the first study happened, in 25% of the models, bacterial growth of coagulase-negative Staphylococcus; Micrococcus sp.; Bacillus sp. and nonfermenting Gram-negative bacillus and, in 5,7%, fungal growth of Cladophialophora sp.; Trichosporon beigelii; Aspergillus flavus and Aureobasidium pullulans. So, it’s important that the dentist avoid crossed contamination and the digital models technology can avoid to store dental... (Complete abstract click electronic access below)
50
Shafi, Muhammad. "Face pose estimation in monocular images." Thesis, Loughborough University, 2010. https://dspace.lboro.ac.uk/2134/5894.
Full textAbstract:
People use orientation of their faces to convey rich, inter-personal information. For example, a person will direct his face to indicate who the intended target of the conversation is. Similarly in a conversation, face orientation is a non-verbal cue to listener when to switch role and start speaking, and a nod indicates that a person has understands, or agrees with, what is being said. Further more, face pose estimation plays an important role in human-computer interaction, virtual reality applications, human behaviour analysis, pose-independent face recognition, driver s vigilance assessment, gaze estimation, etc. Robust face recognition has been a focus of research in computer vision community for more than two decades. Although substantial research has been done and numerous methods have been proposed for face recognition, there remain challenges in this field. One of these is face recognition under varying poses and that is why face pose estimation is still an important research area. In computer vision, face pose estimation is the process of inferring the face orientation from digital imagery. It requires a serious of image processing steps to transform a pixel-based representation of a human face into a high-level concept of direction. An ideal face pose estimator should be invariant to a variety of image-changing factors such as camera distortion, lighting condition, skin colour, projective geometry, facial hairs, facial expressions, presence of accessories like glasses and hats, etc. Face pose estimation has been a focus of research for about two decades and numerous research contributions have been presented in this field. Face pose estimation techniques in literature have still some shortcomings and limitations in terms of accuracy, applicability to monocular images, being autonomous, identity and lighting variations, image resolution variations, range of face motion, computational expense, presence of facial hairs, presence of accessories like glasses and hats, etc. These shortcomings of existing face pose estimation techniques motivated the research work presented in this thesis. The main focus of this research is to design and develop novel face pose estimation algorithms that improve automatic face pose estimation in terms of processing time, computational expense, and invariance to different conditions.