A devolução da estética em prótese ocular está diretamente relacionada ao conhecimento da anatomia e morfologia da íris humana, e uma pintura da íris mais fiel será decorrente de uma boa determinação e reprodução dos sinais anatômicos presentes na íris remanescente do paciente. Nesta pesquisa, foi realizado um estudo dos sinais anatômicos presentes em ambas as íris do globo ocular de 112 indivíduos escolhidos aleatoriamente, através de imagens fotográficas digitais. Foram elaboradas tabelas de avaliação dos sinais: lacunas, anéis de tensão,radiis solaris, sulcos radiais, manchas, flocos de neve, nuvens e anel de gordura, verificando a freqüência com que os mesmos aparecem nas íris consideradas de cor escura (castanho, castanho claro e castanho escuro); de cor clara (azul, verde e cinza) e de média (hazel e âmbar). Os dados obtidos foram processados em análise estatística, através do Teste Exato de Fisher e qui quadrado em nível de significância de 5%. Os resultados mostraram que: os sinais anatômicos são menos evidentes nas íris de cor escura, sendo que, as lacunas e os anéis de tensão foram os mais observados e as nuvens não foram observadas. Nas íris de cor clara, os sinais mais observados foram as manchas e os flocos de neve e o anel de gordura não foi observado. As íris de cor média foram as que apresentaram maior variedade de sinais, prevalecendo os anéis de tensão e as lacunas, e os menos freqüentes foram as nuvens e o anel de gordura. O sinal anel de gordura foi observado apenas em indivíduos em idade superior a 50 anos. Por fim, verificou-se também, que as íris direita e esquerda do mesmo indivíduo são similares, mas não iguais, havendo pequenas alterações na quantidade do mesmo sinal presente em ambas. / The devolution of the aesthetics in ocular prosthesis is directly related to the knowledge of the anatomy and morphology of the iris, and a painting of the most fidelity iris will be due to a good determination and reproduction of the present anatomical signs in the patient\'s remaining iris. In this research, the study of the anatomical signs was accomplished in 112 individuals\' human iris, analyzing the right and left irises, through images of digital pictures. Tables of evaluation were created of the signs: tension rings, lacuna, clouds, cholesterol ring, radiis solaris, snow flakes (lymphatic rosary), spots and spastic furrows and later verified the frequency with that the same ones appear in the irises of color blue, green, gray, hazel, amber, brown colors (brown, dark brown and light brown). The sequence of obtained data was processed in statistical analysis, and the results showed that the signs that more they appeared were the lacuna, stress ring and radiis solaris, and the anatomical signs that fewer appeared in the human iris were snow flakes (lymphatic rosary), clouds and cholesterol ring, however the signs of the type cholesterol rings presented the characteristic of being related the patient\'s age (above 50 years). The irises of color amber and hazel were the ones that presented the largest variety of anatomical signs. Already in the irises of brown color (dark brown, brown and light brown), the signs as a whole is less evident. Finally the same individual\'s right and left irises are similar and no same, suffering alteration in the amount of the same sign presented in both.

[pt] O reconhecimento de humanos pela íris é um dos sistemas mais seguros de identificação biométrica e, motivou a construção de um protótipo de identificação humana baseada na estrutura da íris. O sistema construído é composto de um dispositivo de captura de imagens da íris humana e algoritmos para pré- processamento da imagem, para a representação e o reconhecimento. Cada um dos elementos que compõem o protótipo são avaliados a partir de dois bancos de dados de imagens de íris. Os resultados demonstraram que o dispositivo proposto e os modelos apresentados são capazes de realizar o reconhecimento humano através da íris de forma eficiente. / [en] The recognition of human beings for the Iris is one of the safest systems of biometric identification. This motivated the construction of a prototype for identification of human beings based on the structure of the Iris. The constructed system is composed of a device capable to capture images of the Iris and algorithms for image pre - processing, for the representation and recognition each element composing the prototype is evaluated using two data bases of Iris images. The results have demonstrated that the prototype and the presented models are capable to efficiently identify the human based on Iris structure.

Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level. Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image. Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity.Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level. Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image. Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity. / Higher Committee for Education Development in Iraq

This thesis is based on a field study conducted in Uganda in the fall of 2015. The study is analyzing at the process behind the Domestic Violence Act, a law that came in to place in 2010, and try to scrutinize it by using the inclusive democracy theory of Iris Marion Young. In the study numerous interviews with several women’s organizations, as well as representatives for the Ministry of Gender, Labour and Social Development and local police officers are presented.   The study aims at getting a better understanding of what is necessary to create successful laws to prevent violence against women. The main objective is to answer the question how inclusion, or the lack of it, can influence the success of legal norms and laws regarding violence against women. The result of this study shows that inclusion could play a role in a law’s success. However inclusion is not enough, other factors such as allocating enough money in the budget together with educating both the public and the officials that are enforcing the law, are also of great importance for a law’s success. Yet, this study also shows that a greater inclusion could affect these factors in a positive way, however inclusion alone is most likely not sufficient for creating a successful law

.

No description available.

Ingeniero Civil Eléctrico / Los sistemas biométricos corresponden a sistemas que realizan el reconocomiento automático de individuos a través de atributos únicos. Para que un patrón biométrico sea utilizable, debe cumplir ciertas propiedades, entre ellas las más importantes son: universalidad; invarianza en el tiempo; fácil recolección y almacenamiento. Se considera que el iris cumple con todas estas características y ha sido utilizado en sistemas de reconocimiento por más de una década. A medida que ha transcurrido el tiempo, los sistemas basados en reconocimiento de iris han mejorado considerablemente con respecto a su exactitud, sin embargo, aún se utilizan sistemas altamente cooperativos y en particular en condiciones de iluminación dañinas para la salud, por lo que la comunidad de investigadores se ha volcado en torno a relajar las restricciones en las que operan estos sistemas, con el fin de mejorar la experiencia de uso y la calidad de vida de los usufructuarios. En este trabajo se evalúa un sistema de reconocimiento de iris a distancia, en donde se utilizan las características de color y textura del mismo para llevar a cabo la clasificación. El trabajo abarca el estudio de distintos algoritmos de preprocesamiento de imágenes, esquemas de codificación y comparación propuestos durante los últimos años, los cuales se evalúan cada uno por separado y finalmente se utilizan en conjunto, además de describir el método propuesto para computar la desición final del sistema. El algoritmo se evalúa en una base de datos que consta de 1505 imágenes correspondientes a 54 sujetos, tomadas a una distancia aproximada de 1.5 [m] de distancia de la cámara utilizando iluminación de espectro visible. A partir de esta base de datos se seleccionan las mejores imágenes de iris utilizando distintos sistemas de evaluación de enfoque, luego se particiona en una base de datos de entrenamiento de parámetros y otra de validación, y finalmente se muestran los resultados con respecto a la base de datos de validación. Se utilizaron kernels para evaluar el enfoque de las imágenes capturadas, en particular los propuestos por Daugman, Kang y Wei y se observó que el kernel de Daugman resulta ser más efectivo. Se utilizaron distintos métodos de corrección de enfoque, reducción de ruido y ecualización de imágenes combinados para mejorar la tasa de reconocimiento del sistema, y se concluyó que el filtro de Wiener para reducir el ruido otorga los mejores resultados. Se implementó un sistema de codificación generalizada para aumentar arbitrariamente la cantidad de bits que se obtienen por cada pixel de la imagen, concluyendo que el mejor tamaño de codificación fue de 4 bits por pixel luego de reducir el ruido en las imágenes utilizando el filtro de Wiener. Se implementó un mapa de pesos por cada usuario registrado en la base de datos, la cual se encarga de otorgarle mayor peso a los bits que a priori son menos variables entre muestras, lo que aumentó la exactitud promedio desde 17.72% a 26.6% utilizando sólo esta mejora, con 5 imágenes por usuario enrolado. Se evaluaron dos métodos para clasificar el iris de acuerdo a su coloración, diferencia promedio y distancia de difusión, en donde se concluyó que la diferencia promedio otorga mejores resultados, sin embargo esta característica no es lo suficientemente discriminadora como para ser utilizada por sí sola. El resultado final del índice de reconocimiento del sistema al utilizar todas las mejoras antes mencionadas y fusionando las características de textura iridal y color de iris de ambos ojos es del 90.77%, con una tasa de aceptación de impostores del 0% en promedio.

Se estudió el efecto ecotoxicológico, en tres niveles de dureza total de agua, del dodecil benceno sulfonato de sodio (DBSS) y del alquil aril sulfonato de sodio (AASS), ingredientes activos principales de dos detergentes comerciales biodegradables, en el Centro de Producción “El Ingenio”, Huancayo - Perú, empleando la “trucha arco iris” Oncorhynchus mykiss WALBAUM, 1792 como herramienta para la evaluación de riesgo ambiental. Se trabajó en total con 1440 alevines que contaban con 7 días después de haber absorbido el saco vitelino, los que se acondicionaron por una semana hasta la ejecución de los bioensayos, al iniciar los ensayos de toxicidad se realizó una biometría presentando una talla promedio de 26.8 mm y un peso medio de 0.19 g. Se reporta que el valor de la concentración letal media (CL50) a 96 horas de exposición fue en promedio 13.91 mg L-1 y 13.86 mg L-1 para DEISS y AASS respectivamente, no existiendo diferencia significativa entre los porcentajes de mortalidad de los 2 detergentes evaluados en los tres niveles de dureza de agua. El cociente de riesgo (RQ) en ambos detergentes es inferior a 1 considerando baja la probabilidad que ocasionen daño en el ambiente. / --- It has been studied the ecotoxicological effect, on three levels of water total hardness, and dodecil benzene sulphonate sodium (DBSS) and alquil aril sulphonate sodium (AASS), principal active ingredients of two commercial biodegradation detergents, in the Producction Center “El Ingenio”, Huancayo - Perú, using the “rainbow trout Oncorhynchus mykiss WALBAUM, 1792 as tool for evaluation the environmental risk. It had been used 1440 alevines of 7 days after absorbing the vitellum sack, whose it conditioned by a week until the execution of bioassays. At beginning the toxicity tests, it has realized a biometry reaching an average length of 26.8 mm and average weight of 0.19 g. It has reported that value of average lethal concentration (CL50) of 96 exposure hours was 13.91 mg L-1 and 13.86 mg L-1 for DBSS and AASS, respectively; there are not significant differences between the mortality percentage of detergent assessed on three levels of water hardness. The risk quotient (RQ) in both detergents is lower than 1 considering a low probability that they cause damage on the environment.

Theme of the thesis is research and possibility of using GPS system from localization in wireless sensor network. Paper deals with the accuracy and energy consumption of GPS localization. Thesis also solve using of localization in local anchor system. Theoretical part describes IEEE 802.15.4 standard, capability of used nodes and describe ways how to use RSSI value to resolve location

This Master's Thesis is engaged in business management system requirements of International Railway Industry Standard. The goal is to provide reader with an overview of the requirements of Standard, their introduction into the existing business management system, outline some of the issues associated to the implementation and evaluate the benefits of implementation by analyzing presumed costs associated to the certification of system and expected revenues arising from the successful certification. In conclusion it will assess the results of the evaluation of benefits and other factors leading to a final decision whethere it is a better option for the company to undergo the certification process or negatives will exceed and therefore the company will do better by making its effort in another direction.