Yes / The most recent face recognition systems are mainly dependent on feature representations obtained using either local handcrafted-descriptors, such as local binary patterns (LBP), or use a deep learning approach, such as deep belief network (DBN). However, the former usually suffers from the wide variations in face images, while the latter usually discards the local facial features, which are proven to be important for face recognition. In this paper, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the DBN is proposed to address the face recognition problem in unconstrained conditions. Firstly, a novel multimodal local feature extraction approach based on merging the advantages of the Curvelet transform with Fractal dimension is proposed and termed the Curvelet–Fractal approach. The main motivation of this approach is that theCurvelet transform, a newanisotropic and multidirectional transform, can efficiently represent themain structure of the face (e.g., edges and curves), while the Fractal dimension is one of the most powerful texture descriptors for face images. Secondly, a novel framework is proposed, termed the multimodal deep face recognition (MDFR)framework, to add feature representations by training aDBNon top of the local feature representations instead of the pixel intensity representations. We demonstrate that representations acquired by the proposed MDFR framework are complementary to those acquired by the Curvelet–Fractal approach. Finally, the performance of the proposed approaches has been evaluated by conducting a number of extensive experiments on four large-scale face datasets: the SDUMLA-HMT, FERET, CAS-PEAL-R1, and LFW databases. The results obtained from the proposed approaches outperform other state-of-the-art of approaches (e.g., LBP, DBN, WPCA) by achieving new state-of-the-art results on all the employed datasets.

Yes / n this paper, a powerful face recognition system for authentication and identification tasks is presented and a new facial feature extraction approach is proposed. A novel feature extraction method based on combining the characteristics of the Curvelet transform and Fractal dimension transform is proposed. The proposed system consists of four stages. Firstly, a simple preprocessing algorithm based on a sigmoid function is applied to standardize the intensity dynamic range in the input image. Secondly, a face detection stage based on the Viola-Jones algorithm is used for detecting the face region in the input image. After that, the feature extraction stage using a combination of the Digital Curvelet via wrapping transform and a Fractal Dimension transform is implemented. Finally, the K-Nearest Neighbor (K-NN) and Correlation Coefficient (CC) Classifiers are used in the recognition task. Lastly, the performance of the proposed approach has been tested by carrying out a number of experiments on three well-known datasets with high diversity in the facial expressions: SDUMLA-HMT, Faces96 and UMIST datasets. All the experiments conducted indicate the robustness and the effectiveness of the proposed approach for both authentication and identification tasks compared to other established approaches.

Made available in DSpace on 2017-07-21T18:53:03Z (GMT). No. of bitstreams: 1 Juliane Alves Souza.pdf: 2088285 bytes, checksum: 7df534f70b74bb908ee007c98dc1179a (MD5) Previous issue date: 2014-12-12 / Cereals have an important role in human nutrition. Wheat is one of the most cultivated and consumed cereals in the world being used in the preparation of the enormous diversity of products. The wheat flour fortification with iron and folic acid is obligatory in Brazil in according with current legislation. Starch is a reserve polysaccharide of vegetable consisting essentially of amylose and amylopectin. It is a matter abundant and replaceable material being the main source of carbohydrate in diet human. Starches in their native form have limited features for industrial use so it is important modification aiming to acquire improved stability, rheological properties, paste, gel texture, water retention, among others. The chemical modification, still very used, is not considered as safe a one way of modification, so the physical modification is well regarded by food industry. Heat Moisture Treatment (HMT) in starch is considered a natural method and is conducted with a moisture content of 30% and a maximum temperature of 120 ° C. Were studied native wheat and modified starches with presence and absence of folic acid and iron, by HMT at 121 °C for a period of 1 hour. The effects of HMT were studied using the following techniques: thermogravimetry and derivative thermogravimetry (TG/DTG), differential scanning calorimetry (DSC), x-ray diffraction (DRX), atomic force microscopy (NC-AFM), rapid visco analysis (RVA), the flame atomic absorption spectrometry (FEAA) and reflectance photocolorimetry. From the TG and DTG curves was observed that for the physically treated starches, the values of weight loss for starch in the presence of folic acid iron in the first and third stages decreased and the second increased as increased moisture. The DSC, modified starches, the temperatures at which the event occurred and the enthalpy of gelatinization decreased and the higher the moisture treatment was lower enthalpy. All starches showed bimodal distribution and irregularities in the micro-images of AFM with changes in average diameter and roughness, after physical modification. For all starches DRX peaks did not undergo pronounced displacement, however there was a decrease in crystallinity after modification. All starches showed high paste temperature and low peak viscosity. Was determined absence iron concentration to starch without folic acid and iron and little change for the presence of starch with these nutrients. With respect to color analysis for all samples studied, there was little variation in color between them, with causing a trend mainly to yellow after modification. / Cereais possuem importante papel na alimentação humana. O trigo é um dos cereais mais cultivados e consumidos no mundo sendo utilizado na elaboração de enorme diversidade de produtos. A fortificação da farinha de trigo com ferro e ácido fólico é obrigatória no Brasil de acordo com a legislação vigente. O amido é um polissacarídeo de reserva vegetal constituído basicamente por amilose e amilopectina. É uma matéria prima abundante e renovável sendo a principal fonte de carboidrato na alimentação humana. Amidos na forma nativa possuem características limitadas para uso industrial por isso torna-se importante sua modificação visando adquirir melhor estabilidade, propriedades reológicas, de pasta, textura de géis, retenção de água entre outros. A modificação química, ainda muito utilizada, não é considerada uma forma tão segura de modificação, por isso a modificação física é bem vista pela indústria alimentícia. A modificação por umidade e temperatura controladas (HMT) em amido é considerada um método natural e é conduzida com teor de umidade de até 30% e temperatura máxima de 121 ºC. Foram estudados amidos trigo nativos e modificados, com presença e ausência de ácido fólico e ferro, por HMT a 121°C por um período de 1 hora. Os efeitos do HMT foram estudados utilizando as seguintes técnicas: termogravimetria e termogravimetria derivada (TG/DTG), calorimetria exploratória diferencial (DSC), difração de raios X (DRX), microscopia de força atômica (NC-AFM), análise viscoamilográfica rápida (RVA), espectrometria de absorção atômica com chama (FEAA) e refletância fotocolorimetria. A partir das curvas TG e DTG, observou-se que para os amidos tratados fisicamente, os valores de perda de massa para o amido em presença de ácido fólico e ferro, nas primeiras e terceiras etapas diminuíram e da segunda aumentaram, conforme aumento da umidade. Após a análise de DSC, os amidos modificados, as temperaturas em que ocorreu o evento e o valor da entalpia de gelatinização diminuíram e quanto maior a umidade de tratamento menor foi à entalpia. Os amidos deste estudo apresentaram distribuição bimodal e irregularidades nas microimagens de AFM, com alterações nos diâmetros médios e rugosidade, após a modificação física. Para todos os amidos os picos de DRX não sofreram deslocamento acentuado, porem houve redução da cristalinidade após a modificação. Todos os amidos apresentaram alta temperatura de pasta e baixo pico de viscosidade. Foi determinada ausência de ferro para o amido sem ácido fólico e ferro e pouca alteração para o amido com presença desses nutrientes. Com relação a análise de cor, para todas as amostras estudadas, houve pouca variação de cor entre elas, com tendência ao amarelamento após a modificação.

Yes / In the health care systems, a high security level is required to protect extremely sensitive patient records. The goal is to provide a secure access to the right records at the right time with high patient privacy. As the most accurate biometric system, the iris recognition can play a significant role in healthcare applications for accurate patient identification. In this paper, the corner stone towards building a fast and robust iris recognition system for healthcare applications is addressed, which is known as iris localization. Iris localization is an essential step for efficient iris recognition systems. The presence of extraneous features such as eyelashes, eyelids, pupil and reflection spots make the correct iris localization challenging. In this paper, an efficient and automatic method is presented for the inner and outer iris boundary localization. The inner pupil boundary is detected after eliminating specular reflections using a combination of thresholding and morphological operations. Then, the outer iris boundary is detected using the modified Circular Hough transform. An efficient preprocessing procedure is proposed to enhance the iris boundary by applying 2D Gaussian filter and Histogram equalization processes. In addition, the pupil’s parameters (e.g. radius and center coordinates) are employed to reduce the search time of the Hough transform by discarding the unnecessary edge points within the iris region. Finally, a robust and fast eyelids detection algorithm is developed which employs an anisotropic diffusion filter with Radon transform to fit the upper and lower eyelids boundaries. The performance of the proposed method is tested on two databases: CASIA Version 1.0 and SDUMLA-HMT iris database. The Experimental results demonstrate the efficiency of the proposed method. Moreover, a comparative study with other established methods is also carried out.

La chromatine est essentielle au maintien de l’intégrité du génome, mais, ironiquement, constitue l’obstacle principal à la transcription des gènes. Plusieurs mécanismes ont été développés par la cellule pour pallier ce problème, dont l’acétylation des histones composant les nucléosomes. Cette acétylation, catalysée par des histones acétyl transférases (HATs), permet de réduire la force de l’interaction entre les nucléosomes et l’ADN, ce qui permet à la machinerie transcriptionnelle de faire son travail. Toutefois, on ne peut laisser la chromatine dans cet état permissif sans conséquence néfaste. Les histone déacétylases (HDACs) catalysent le clivage du groupement acétyle pour permettre à la chromatine de retrouver une conformation compacte. Cette thèse se penche sur la caractérisation de la fonction et du mécanisme de recrutement des complexes HDACs Rpd3S et Set3C. Le complexe Rpd3S est recruté aux régions transcrites par une interaction avec le domaine C-terminal hyperphosphorylé de Rpb1, une sous-unité de l’ARN polymérase II. Toutefois, le facteur d’élongation DSIF joue un rôle dans la régulation de cette association en limitant le recrutement de Rpd3S aux régions transcrites. L’activité HDAC de Rpd3S, quant à elle, dépend de la méthylation du résidu H3K36 par l’histone méthyltransférase Set2. La fonction du complexe Set3C n’est pas clairement définie. Ce complexe est recruté à la plupart de ses cibles par l’interaction entre le domaine PHD de Set3 et le résidu H3K4 di- ou triméthylé. Un mécanisme indépendant de cette méthylation, possiblement le même que pour Rpd3S, régit toutefois l’association de Set3C aux régions codantes des gènes les plus transcrits. La majorité de ces résultats ont été obtenus par la technique d’immunoprécipitation de la chromatine couplée aux biopuces (ChIP-chip). Le protocole technique et le design expérimental de ce type d’expérience fera aussi l’objet d’une discussion approfondie. / Chromatin is essential for the maintenance of genomic integrity but, ironically, is also the main barrier to gene transcription. Many mechanisms, such as histone acetylation, have evolved to overcome this problem. Histone acetylation, catalyzed by histone acetyltransferases (HATs), weakens the internucleosomal and nucleosome-DNA interactions, thus permitting the transcriptional machinery access to its template. However, this permissive chromatin state also allows for opportunistic DNA binding events. Histone deacetylases (HDACs) help restore a compact chromatin structure by catalyzing the removal of acetyl moieties from histones. This thesis focuses on the characterization of the function and of the recruitment mechanism of HDAC complexes Rpd3S and Set3C. The Rpd3S complex is recruited to actively transcribed coding regions through interactions with the hyperphosphorylated C-terminal domain of Rpb1, a subunit of RNA polymerase II, with the DSIF elongation factor playing a role in limiting this recruitment. However, the HDAC activity of Rpd3S depends on H3K36 methylation, which is catalyzed by the Set2 histone methyltransferase. The Set3C complex’ function is still not clearly defined. It is recruited to most of its targets through the interaction between the Set3 PHD domain and di- or trimethylated H3K4. However, Set3C recruitment to genes displaying high RNA polymerase II occupancy is independent of H3K4 methylation. The mechanism by which Set3C is recruited to this gene subset is under investigation. These results have mostly been obtained through chromatin immunoprecipitation coupled to tiling microarrays (ChIP-chip). The protocol and experimental design challenges inherent to this technique will also be discussed in depth.

La chromatine est essentielle au maintien de l’intégrité du génome, mais, ironiquement, constitue l’obstacle principal à la transcription des gènes. Plusieurs mécanismes ont été développés par la cellule pour pallier ce problème, dont l’acétylation des histones composant les nucléosomes. Cette acétylation, catalysée par des histones acétyl transférases (HATs), permet de réduire la force de l’interaction entre les nucléosomes et l’ADN, ce qui permet à la machinerie transcriptionnelle de faire son travail. Toutefois, on ne peut laisser la chromatine dans cet état permissif sans conséquence néfaste. Les histone déacétylases (HDACs) catalysent le clivage du groupement acétyle pour permettre à la chromatine de retrouver une conformation compacte. Cette thèse se penche sur la caractérisation de la fonction et du mécanisme de recrutement des complexes HDACs Rpd3S et Set3C. Le complexe Rpd3S est recruté aux régions transcrites par une interaction avec le domaine C-terminal hyperphosphorylé de Rpb1, une sous-unité de l’ARN polymérase II. Toutefois, le facteur d’élongation DSIF joue un rôle dans la régulation de cette association en limitant le recrutement de Rpd3S aux régions transcrites. L’activité HDAC de Rpd3S, quant à elle, dépend de la méthylation du résidu H3K36 par l’histone méthyltransférase Set2. La fonction du complexe Set3C n’est pas clairement définie. Ce complexe est recruté à la plupart de ses cibles par l’interaction entre le domaine PHD de Set3 et le résidu H3K4 di- ou triméthylé. Un mécanisme indépendant de cette méthylation, possiblement le même que pour Rpd3S, régit toutefois l’association de Set3C aux régions codantes des gènes les plus transcrits. La majorité de ces résultats ont été obtenus par la technique d’immunoprécipitation de la chromatine couplée aux biopuces (ChIP-chip). Le protocole technique et le design expérimental de ce type d’expérience fera aussi l’objet d’une discussion approfondie. / Chromatin is essential for the maintenance of genomic integrity but, ironically, is also the main barrier to gene transcription. Many mechanisms, such as histone acetylation, have evolved to overcome this problem. Histone acetylation, catalyzed by histone acetyltransferases (HATs), weakens the internucleosomal and nucleosome-DNA interactions, thus permitting the transcriptional machinery access to its template. However, this permissive chromatin state also allows for opportunistic DNA binding events. Histone deacetylases (HDACs) help restore a compact chromatin structure by catalyzing the removal of acetyl moieties from histones. This thesis focuses on the characterization of the function and of the recruitment mechanism of HDAC complexes Rpd3S and Set3C. The Rpd3S complex is recruited to actively transcribed coding regions through interactions with the hyperphosphorylated C-terminal domain of Rpb1, a subunit of RNA polymerase II, with the DSIF elongation factor playing a role in limiting this recruitment. However, the HDAC activity of Rpd3S depends on H3K36 methylation, which is catalyzed by the Set2 histone methyltransferase. The Set3C complex’ function is still not clearly defined. It is recruited to most of its targets through the interaction between the Set3 PHD domain and di- or trimethylated H3K4. However, Set3C recruitment to genes displaying high RNA polymerase II occupancy is independent of H3K4 methylation. The mechanism by which Set3C is recruited to this gene subset is under investigation. These results have mostly been obtained through chromatin immunoprecipitation coupled to tiling microarrays (ChIP-chip). The protocol and experimental design challenges inherent to this technique will also be discussed in depth.

Ce travail s’inscrit dans le cadre du développement d’une médecine davantage personnalisée et préventive, pour laquelle la fusion d’informations multi modale et de différentes représentations d'une même modalité sont nécessaires afin d'aboutir à une quantification fiable des images médicales en oncologie. Dans cette étude nous présentons deux applications de traitement et d'analyse des images médicales: le débruitage des images TEP et la détermination des volumes anatomo-fonctionnels des tumeurs en imagerie multi modale TEP/TDM. Pour le débruitage des images TEP, nous avons mis en place une approche intitulée "WCD" permettant de bénéficier des caractéristiques complémentaires de la transformée en ondelettes et la transformée en Curvelets afin de mieux représenter les structures isotropiques et anisotropiques dans ces images, ce qui permet de réduire le bruit tout en minimisant les pertes d'informations utiles dans les images TEP. En ce qui concerne la deuxième application, nous avons proposé une méthode de segmentationTEP/TDM intitulée "WCHMT" permettant d'exploiter la spécificité des arbres de Markov caché de prendre en compte les dépendances statistiques entre l’ensemble des données. Ce modèle permet de gérer simultanément les propriétés complémentaires de l’imagerie fonctionnelle et l’imagerie morphologique dans un cadre unifié où les données sont représentées dans le domaine des Contourlets. Le débruitage en TEP a abouti à une hausse significative du rapport signal sur-bruit (SNR) en garantissant la moindre variation de l'intensité et du contraste local. Quant à la segmentation multimodale TEP/TDM, elle a démontré une bonne précision lors de la détermination du volume tumoral en terme du coefficient de Dice (DSC) avec le meilleur compromis entre la sensibilité (SE) et la valeur prédictive positive (PPV) par rapport à la vérité terrain. / This thesis is a part of the development of more personalized and preventive medicine, for which a fusion of multi modal information and diverse representations of the same modality is needed in order to get accurate and reliable quantification of medical images in oncology. In this study we present two applications for image processing analysis: PET denoising and multimodal PET/CT tumor segmentation. The PET filtering approach called "WCD" take benefit from the complementary features of the wavelet and Curvelets transforms in order to better represent isotropic and anisotropic structures in PET images. This algorithm allows the reduction of the noise while minimizing the loss of useful information in PET images. The PET/CT tumor segmentation application is performed through a Markov model as a probabilistic quadtree graph namely a Hidden Markov Tree (HMT).Our motivation for using such a model is to provide fast computation, improved robustness and an effective interpretational framework for image analysis on oncology. Thanks to two efficient aspects (multi observation and multi resolution), when dealing with Hidden Markov Tree (HMT), we exploit joint statistical dependencies between hidden states to handle the whole data stack. This model called "WCHMT" take advantage of the high resolution of the anatomic imaging (CT) and the high contrast of the functional imaging (PET). The denoising approach led to the best trade-off between denoising quality and structure preservation with the least quantitative bias in absolute intensity recovery. PET/CT segmentation's results performed with WCHMT method has proven a reliable segmentation when providing high Dice Similarity Coeffcient (DSC) with the best trade-off between sensitivity (SE) and positive predictive value (PPV).

No description available.