Submitted by José Edmilson Lobato Júnior (edmilsonjr@hotmail.com) on 2015-01-06T19:21:59Z No. of bitstreams: 1 2014 - Dissertação - DEFESA.docx: 3720569 bytes, checksum: 7e072a8de7b7e262ed9c374fd083c6d3 (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-01-19T13:25:37Z (GMT) No. of bitstreams: 1 2014 - Dissertação - DEFESA.docx: 3720569 bytes, checksum: 7e072a8de7b7e262ed9c374fd083c6d3 (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-01-19T16:04:32Z (GMT) No. of bitstreams: 1 2014 - Dissertação - DEFESA.docx: 3720569 bytes, checksum: 7e072a8de7b7e262ed9c374fd083c6d3 (MD5) / Made available in DSpace on 2015-01-19T16:06:16Z (GMT). No. of bitstreams: 1 2014 - Dissertação - DEFESA.docx: 3720569 bytes, checksum: 7e072a8de7b7e262ed9c374fd083c6d3 (MD5) Previous issue date: 2014-06-05 / In the early 90's started a movement across Brazil, powered by a letter of intent UN, which suggested the decoupling of Criminal Expertise Institution of Civil Police. After this move, there was a wide range of administrative structures of expert institutions, both operational and administrative, confirming a status quo complete empiricism in all sectors. This study we attempted to initiate a proposal for standardization of inputs fingerprint nature contained in the briefcase crime scene donated by SENASP to all institutes of Criminology Brazil. In an attempt to reach the largest possible representation among Brazilian Institutes of Criminalistics, we used interviews and an online questionnaire directed to the National Forum of Experts Criminal (FNPC), where the sample was obtained from almost all states of the federation. The result was achieved qualitatively quite representative, where the trademark SIRCHIE proved with considerable preference between the Expert class. However, it is necessary to continue this work, seeking to achieve other variables not examined in this context, but may show the apparent reasons that perpetuate the lack of concrete processes of standardization in the National Skill. / Nos início da década de 90 teve início um movimento em todo Brasil, impulsionado por uma carta de intenções da ONU, que sugeria a desvinculação da Perícia Criminal da Instituição Polícia Civil. Após este movimento, verificou-se uma grande diversidade de estruturas que originaram as Instituições Periciais, tanto operacionalmente como administrativamente, chancelando um status quo de completo empirismo em todos os setores periciais. Buscou-se neste trabalho, iniciar uma análise da possibilidade de padronização dos materiais utilizados na coleta de impressões digitais latentes contidos na maleta de local de crime doada pela SENASP a todos os Institutos de Criminalística do Brasil. Na tentativa de alcançar a maior representatividade possível dentre os Institutos de Criminalística brasileiros, utilizou-se um questionário eletrônico direcionado ao Fórum Nacional de Peritos Criminais (FNPC), onde se obteve amostra de quase todos os estados da federação. No resultado alcançado qualitativamente, a marca comercial SIRCHIE demonstrou ter preferência considerável entre a classe Pericial.

The usage of automatic fingerprint identification systems as a means of identification and/or verification have increased substantially during the last couple of years. It is well known that small deviations may occur within a fingerprint over time, a problem referred to as template ageing. This problem, and other reasons for deviations between two images of the same fingerprint, complicates the identification/verification process, since distinct features may appear somewhat different in the two images that are matched. Commonly used to try and minimise this type of problem are different kinds of fingerprint image enhancement algorithms. This thesis tests different methods within the scale-space framework and evaluate their performance as fingerprint image enhancement methods. The methods tested within this thesis ranges from linear scale-space filtering, where no prior information about the images is known, to scalar and tensor driven diffusion where analysis of the images precedes and controls the diffusion process. The linear scale-space approach is shown to improve correlation values, which was anticipated since the image structure is flattened at coarser scales. There is however no increase in the number of accurate matches, since inaccurate features also tends to get higher correlation value at large scales. The nonlinear isotropic scale-space (scalar dependent diffusion), or the edge- preservation, approach is proven to be an ill fit method for fingerprint image enhancement. This is due to the fact that the analysis of edges may be unreliable, since edge structure is often distorted in fingerprints affected by the template ageing problem. The nonlinear anisotropic scale-space (tensor dependent diffusion), or coherence-enhancing, method does not give any overall improvements of the number of accurate matches. It is however shown that for a certain type of template ageing problem, where the deviating structure does not significantly affect the ridge orientation, the nonlinear anisotropic diffusion is able to accurately match correlation pairs that resulted in a false match before they were enhanced.

Composantes sine qua non des contenus multimédias distribués et/ou partagés via un réseau, les techniques de fingerprinting permettent d'identifier tout contenu numérique à l'aide d'une signature (empreinte) de taille réduite, calculée à partir des données d'origine. Cette signature doit être invariante aux transformations du contenu. Pour des vidéos, cela renvoie aussi bien à du filtrage, de la compression, des opérations géométriques (rotation, sélection de sous-région… ) qu'à du sous-échantillonnage spatio-temporel. Dans la pratique, c'est l'enregistrement par caméscope directement dans une salle de projection qui combine de façon non linéaire toutes les transformations pré-citées.Par rapport à l'état de l'art, sous contrainte de robustesse à l'enregistrement en salle de cinéma, trois verrous scientifiques restent à lever : (1) unicité des signatures, (2) appariement mathématique des signatures, (3) scalabilité de la recherche au regard de la dimension de la base de données.La principale contribution de cette thèse est de spécifier, concevoir, implanter et valider TrackART, une nouvelle méthode de traçage des contenus vidéo relevant ces trois défis dans un contexte de traçage de contenus cinématographiques.L'unicité de la signature est obtenue par sélection d'un sous-ensemble de coefficients d'ondelettes, selon un critère statistique de leurs propriétés. La robustesse des signatures aux distorsions lors de l'appariement est garantie par l'introduction d'un test statistique Rho de corrélation. Enfin, la méthode développée est scalable : l'algorithme de localisation met en œuvre une représentation auto-adaptative par sac de mots visuels. TrackART comporte également un mécanisme de synchronisation supplémentaire, capable de corriger automatiquement le jitter introduit par les attaques de désynchronisation variables en temps.La méthode TrackART a été validée dans le cadre d'un partenariat industriel, avec les principaux professionnels de l'industrie cinématographique et avec le concours de la Commission Technique Supérieure de l'Image et du Son. La base de données de référence est constituée de 14 heures de contenu vidéo. La base de données requête correspond à 25 heures de contenu vidéo attaqué, obtenues en appliquant neuf types de distorsion sur le tiers des vidéo de la base de référence.Les performances de la méthode TrackART ont été mesurées objectivement dans un contexte d'enregistrement en salle : la probabilité de fausse alarme est inférieure à 16*10^-6, la probabilité de perte inférieure à 0,041, la précision et le rappel sont égal à 93%. Ces valeurs représentent une avancée par rapport à l'état de l'art qui n'exhibe aucune méthode de traçage robuste à l'enregistrement en salle et valident une première preuve de concept de la méthodologie statistique développée. / Sine qua non component of multimedia content distribution on the Internet, video fingerprinting techniques allow the identification of content based on digital signatures(fingerprints) computed from the content itself. The signatures have to be invariant to content transformations like filtering, compression, geometric modifications, and spatial-temporal sub-sampling/cropping. In practice, all these transformations are non-linearly combined by the live camcorder recording use case.The state-of-the-art limitations for video fingerprinting can be identified at three levels: (1) the uniqueness of the fingerprint is solely dealt with by heuristic procedures; (2) the fingerprinting matching is not constructed on a mathematical ground, thus resulting in lack of robustness to live camcorder recording distortions; (3) very few, if any, full scalable mono-modal methods exist.The main contribution of the present thesis is to specify, design, implement and validate a new video fingerprinting method, TrackART, able to overcome these limitations. In order to ensure a unique and mathematical representation of the video content, the fingerprint is represented by a set of wavelet coefficients. In order to grant the fingerprints robustness to the mundane or malicious distortions which appear practical use-cases, the fingerprint matching is based on a repeated Rho test on correlation. In order to make the method efficient in the case of large scale databases, a localization algorithm based on a bag of visual words representation (Sivic and Zisserman, 2003) is employed. An additional synchronization mechanism able to address the time-variants distortions induced by live camcorder recording was also designed.The TrackART method was validated in industrial partnership with professional players in cinematography special effects (Mikros Image) and with the French Cinematography Authority (CST - Commision Supérieure Technique de l'Image et du Son). The reference video database consists of 14 hours of video content. The query dataset consists in 25 hours of replica content obtained by applying nine types of distortions on a third of the reference video content. The performances of the TrackART method have been objectively assessed in the context of live camcorder recording: the probability of false alarm lower than 16 10-6, the probability of missed detection lower than 0.041, precision and recall equal to 0.93. These results represent an advancement compared to the state of the art which does not exhibit any video fingerprinting method robust to live camcorder recording and validate a first proof of concept for the developed statistical methodology.

Ziel der vorliegenden Arbeit war die Untersuchung von Mikroalgen-Kulturstämmen unterhalb des Artniveaus. Für Service-Kulturensammlungen sind diese Untersuchungen wichtig, um die genetische Diversität der vorhandenen bzw. neu aufzunehmender Kulturen zu erfassen, Stämme eindeutig zu identifizieren oder auch Kontaminationen auszuschließen. Zu diesem Zweck wurden häufig in der Forschung genutzte Stämme von Mikroalgen mithilfe von genetischen Fingerabdrücken, die mit der AFLP (amplified fragment length polymorphism) Technik generiert wurden, untersucht. Mittels AFLP was es möglich, verschiedene Stämme derselben Art (Chlorella vulgaris) zu unterscheiden und für jedes Isolat einen charakteristischen genetischen Fingerabdruck zu erhalten. Da diese genomischen Variationen innerhalb derselben Art auch mit Unterschieden in physiologischen/biochemischen Eigenschaften korreliert sein können, ist es wichtig genau festzuhalten, welches Isolat für einen Versuch oder eine biotechnologische Anwendung verwendet wurde. Mit der AFLP-Technik war es ferner möglich, phänotypisch unterschiedliche Mutanten von Parachlorella kessleri und den entsprechenden Wildtyp zu unterscheiden. Normalerweise ist es mit AFLP nicht möglich, Algenkulturen zu untersuchen, die mit anderen Organismen kontaminiert sind. Das Problem hierbei ist, dass bei den generierten Fragmenten nicht unterschieden werden kann, ob sie von der Alge stammen oder von der Kontamination. Es konnte hier jedoch gezeigt werden, dass kontaminierte Kulturen trotzdem mit AFLP untersucht werden können. Hierzu war es nötig, Kulturen mit unterschiedlichem Verhältnis von Algen- und Bakterienzellen zu untersuchen. Außerdem konnte gezeigt werden, dass Infektionen von Algenkulturen mit Viren, die meistens schwer nachzuweisen sind, mittels AFLP festgestellt werden können. Kryokonservierung ist mittlerweile die Methode der Wahl zur Langzeitaufbewahrung von Mikroalgen. Es wird davon ausgegangen, dass diese Aufbewahrungsmethode genetisch stabile Kulturen über Jahrzehnte hinweg garantieren kann. Jedoch wurde bis heute die genetische Stabilität von kryokonservierten Algen noch nicht überprüft und deshalb in der vorliegenden Arbeit mithilfe der AFLP-Technik untersucht.

An iterative travel time forecasting scheme, named the Advanced Multilane Prediction based Real-time Fastest Path (AMPRFP) algorithm, is presented in this dissertation. This scheme is derived from the conventional kernel estimator based prediction model by the association of real-time nonlinear impacts that caused by neighboring arcs’ traffic patterns with the historical traffic behaviors. The AMPRFP algorithm is evaluated by prediction of the travel time of congested arcs in the urban area of Jacksonville City. Experiment results illustrate that the proposed scheme is able to significantly reduce both the relative mean error (RME) and the root-mean-squared error (RMSE) of the predicted travel time. To obtain high quality real-time traffic information, which is essential to the performance of the AMPRFP algorithm, a data clean scheme enhanced empirical learning (DCSEEL) algorithm is also introduced. This novel method investigates the correlation between distance and direction in the geometrical map, which is not considered in existing fingerprint localization methods. Specifically, empirical learning methods are applied to minimize the error that exists in the estimated distance. A direction filter is developed to clean joints that have negative influence to the localization accuracy. Synthetic experiments in urban, suburban and rural environments are designed to evaluate the performance of DCSEEL algorithm in determining the cellular probe’s position. The results show that the cellular probe’s localization accuracy can be notably improved by the DCSEEL algorithm. Additionally, a new fast correlation technique for overcoming the time efficiency problem of the existing correlation algorithm based floating car data (FCD) technique is developed. The matching process is transformed into a 1-dimensional (1-D) curve matching problem and the Fast Normalized Cross-Correlation (FNCC) algorithm is introduced to supersede the Pearson product Moment Correlation Co-efficient (PMCC) algorithm in order to achieve the real-time requirement of the FCD method. The fast correlation technique shows a significant improvement in reducing the computational cost without affecting the accuracy of the matching process.

The main goal of this thesis is to design and implement a mechanism that provides protection against stateless tracking with browser fingerprint. Implemented tool has a form of module that takes part of JavaScript Restrictor extension. The module allows to specify heuristics used for evaluation of visited sites that may contain browser fingerprint extraction. If suspicious activity is detected, all subsequent HTTP requests from that site are blocked to prevent the extracted fingerprint from being sent to the server. The implementation and defined heuristics were tested. The resulting module represents an effective tool against stateless tracking. The main limitation of the implementation is possible corruption of sites by blocking HTTP requests.

Cieľom tejto diplomovej práce je vygenerovanie datasetu syntetických snímkov odtlačkov prstov, ktoré vykazujú známky kožných ochorení. Práca sa zaoberá poškodením spôsobeným kožnými ochoreniami v odtlačkoch prstov a generovaním syntetických odtlačkov prstov. Odtlačky prstov s prejavom kožných ochorení boli generované s využitím modelu založeného na Wasserstein GAN s penalizáciou gradientu. Na trénovanie GAN modelu bola použitá unikátna databáza odtlačkov prstov s prejavom kožných ochorení vytvorená na FIT VUT. Daný model bol trénovaný na troch typoch kožných ochorení: atopický ekzém, psoriáza a dyshidrotický ekzém. Sieť generátoru z natrénovaného WGAN-GP modelu bola použitá na vygenerovanie datasetov syntetických odtlačkov prstov. Tieto syntetické odtlačky boli porovnané s reálnymi odtlačkami s využitím NFIQ a FiQiVi nástrojov na určenie kvality spoločne s porovnaním rozložení lokácií a orientácii markantov v snímkoch odtlačkov prstov.

This master’s thesis deals with the issue of scanning human fingerprints, which is currently very topical and represents the most widespread biometric technology. The theoretical part of the work acquaints the reader with basics of dactyloscopy and biometrics and concerns technologies used for fingerprinting, image preprocessing methods and commercially available contactless optical scanners. The practical part is a realisation of a contactless optical scanner based on a Raspberry Pi minicomputer, implementation of preprocessing algorithms in Python and testing of the device from the perspective of image quality.

This master‘s thesis deals with biometric fingerprint liveness detection. The theoretical part of the work describes fingerprint recognition biometric systems, fingerprint liveness detection issues and methods for fingerprint liveness detection. The practical part of the work describes proposed set of discriminant features and preprocessing of fingerprint image. Proposed approach using neural network to detect a liveness. The algorithm is tested on LivDet database comprising real and fake images acquired with tree sensors. Classification performance approximately 93% was obtained.

The work deals with verification of the metadata of audio recordings in an environment of mobile phones running Android. The text work is divided into three main parts. The first part is focused more on theoretical basis, where the main emphasis on the description of Android, application structure, description of the audio MP3 file and theoretical basis of Fourier transform in signal processing. The second part is devoted to solution design, analysis and implementation. The third part deals with testing.