Due to the varied climate of devices used today, the web must accommodate to every use case via responsive design. The problem is that images are often forgotten when implementing fluid design, resulting in slower loading of pages and more data sent than necessary. This thesis aims to explore through a literary study which valid standardized responsive image solutions there are using HTML5 and how to implement them. The empirical study of this thesis will be an experiment conducted to find answers on what the benefits are and what you must consider when implementing responsive image solutions. The experiment consists of collecting data from three different websites implementing various degrees of responsive image solutions. We conclude from the literary study that there are currently two valid standardized responsive image solutions in HTML5 in the form of the ‘picture’ element and the ‘srcset’ attribute. We find that the increased size of data stored on disk and the increased complexity of the markup and images must be considered when implementing responsive image solutions. The benefits of the technique will show in terms of decreased loading times of the page and less data sent over the network. We also concluded that a clearer focus and more context was achieved on the images implementing art direction.

Esta dissertação buscou entender como o processo de leitura da linguagem visual pode se expandir, por meio das relações que se estabelecem entre o eu - fotógrafo e o objeto a ser fotografado, através de um curso denominado Oficina de Fotografia. Adotou-se como referencial teórico os estudos de Paulo Freire sobre educação; de Ana Mae Barbosa sobre leitura de imagens; de María Acaso sobre educação visual; e de Joan Fontcuberta sobre a linguagem fotográfica. Para tal, constituiu-se um Grupo Focal com alunos de diferentes idades e profissões que haviam realizado o curso Oficina de Fotografia. Os depoimentos coletados constituem o corpus de análise dessa pesquisa. O método do Grupo Focal diz respeito à abordagem qualitativa de pesquisa, que procura conhecer o que os sujeitos envolvidos em uma determinada situação ou evento pensam a respeito de um tema proposto pelo pesquisador. Os resultados obtidos a partir da análise das entrevistas demonstram que os alunos passaram a ter uma leitura mais criteriosa não só das imagens fotográficas como também da linguagem visual em cenas cotidianas, após realizarem a Oficina de Fotografia. / This dissertation aimed to understand how the process of reading visual language can be expand by the relations that are established between the photographer itself and the object to be photographed in a course called Workshop of Photography. As theoretical reference it was used the studies from Paulo Freire about education; from Ana Mae Barbosa about images reading; from María Acaso about visual education; and from Joan Fontcuberta about photographic language. A Focus Group was organized with subjects from different ages and professions that had already attended the Workshop of Photography. The statements collected constitute the corpus of analysis of this research. The Focus Group is applied in a qualitative research methodology that aims to know what the subjects, involved in a particular situation or event, think about an issue suggested by the researcher. The obtained results based on an analysis from the interviews showed that the subjects, after attending the Workshop of Photography, have achieved a more discerning reading not only of the photographic images, but also from the visual language in everyday life scenes.

Esta dissertação está inserida no âmbito do estudo das imagens em livros didáticos de História. A partir da coleção História em Movimento (2014), pertencente ao Programa Nacional do Livro Didático 2015 (PNLD) para o Ensino Médio, investiguei um registro imagético específico, nesse caso, as reproduções de obras de arte. Busquei compreender de que modo essa coleção articula saberes históricos com saberes artísticos, identificando e refletindo sobre as estratégias pedagógicas e metodológicas utilizadas para tal, bem como sobre as relações entre o ensino de História e o de Artes Visuais. Para tanto, me detive na análise documental do Edital de Convocação para o processo de inscrição e avaliação de obras didáticas para o PNLD 2015 Ensino Médio (01/2013), no exame do Manual do Professor da coleção História em Movimento, na realização e na análise de um levantamento quantitativo dos registros imagéticos e artísticos que compõem a referida coleção e na investigação de uma seção específica – Olho Vivo. A pesquisa demonstra que a presença de registros imagéticos e, particularmente, de reproduções de obras de arte em coleções didáticas é fruto de uma série de fatores ligados a demandas educacionais específicas. No caso da coleção História em Movimento, as reproduções de obras de arte são utilizadas na medida em que tematizam representações de personagens e indivíduos históricos, sendo interpretadas enquanto símbolos e alegorias que possuem significados específicos a serem decifrados. / This dissertation inserts itself in the scope of the study of images in History textbooks. Based on the História em Movimento collection (2014), which belongs to the 2015 National Textbook Program (PNLD) for High Schools, I have investigated a specific imagetic record, in this case, the reproductions of artworks. I have tried to understand how this collection articulates both historic and artistic knowledge, identifying and reflecting on the pedagogical strategies and methodologies used for doing so, as well as the relations between History and Visual Arts teaching. Hence, I have focused on the documental analysis of the convocation public notice for the process of enrollment and evaluation of textbooks for the 2015 National Textbook Program for High School (01/2013), on the examination of the teacher’s manual of the História em Movimento collection, on the realization and analysis of a quantitative survey of the imagetic and artistic records that compose the referred collection and on the investigation of a specific section - Olho Vivo. The research shows that the presence of imagetic records and, particularly, of artwork reproductions in textbook collections is the outcome of a series of factors connected to specific educational demands. In the História em Movimento collection, the artwork reproductions are used as far as they conceptualize representations of characters and historical individuals, being interpreted as symbols and allegories which have specific meanings and need to be deciphered.

In the context of globalization, commercialization, and the increasing presence of Western images (Western models, Western languages, and Western settings) in Chinese society, this study examines how these images are constructed in Chinese magazine advertising. It utilizes quantitative content analysis, facilitated by semiotic analysis to approach issues of race, gender and power reflected in the images of the West. Methodologically, this study sees quantitative content analysis and semiotics as two complementary methods in the study of contemporary visually dominant print advertising. Theories in both social sciences and humanities were reviewed and brought into the analysis of data. Based on advertisements (N=2,882) from a stratified random sample of four months in 2009 in 19 Chinese consumer magazines (men's, women's and general interest), results of the study showed that images of the West are primarily represented by Western models, and White female models in particular. Focusing on advertisements for Chinese products and services, Western models in Chinese magazine advertising were found differently portrayed from their Chinese counterparts, in their frequency, the type of magazines they appear in, product/service categories they feature in, occupational status, the extent of nudity, and relation to product. When Western models were chosen for an ad, they served different roles than Chinese models. Close examination of individual advertisements from a semiotic perspective showed that Western female models were eroticized, while at the same time representing universal beauty; both Western male and female models were signs that signify quality, social status, luxury and enjoyment of good life; China's relation to the West is also complicated by the fact that China sees itself as a collaborator with the West in the economic domain, and at the same time, accepts the power of the West by romanticizing the colonial past. Overall, this study showed that images of the West in the Chinese context are multifaceted: they have different significations in different domains.

Grounding Natural Language Inference on Images Hoa Trong VU July 20, 2018 Abstract Despite the surge of research interest in problems involving linguistic and vi- sual information, exploring multimodal data for Natural Language Inference remains unexplored. Natural Language Inference, regarded as the basic step towards Natural Language Understanding, is extremely challenging due to the natural complexity of human languages. However, we believe this issue can be alleviated by using multimodal data. Given an image and its description, our proposed task is to determined whether a natural language hypothesis contra- dicts, entails or is neutral with regards to the image and its description. To address this problem, we develop a multimodal framework based on the Bilat- eral Multi-perspective Matching framework. Data is collected by mapping the SNLI dataset with the image dataset Flickr30k. The result dataset, made pub- licly available, has more than 565k instances. Experiments on this dataset show that the multimodal model outperforms the state-of-the-art textual model. References 1

[pt] Esta dissertação examina a idéia do bem na República de Platão, nas famosas imagens do Sol (505a-509b), da Linha (509d-511e) e da Caverna (514a- 521b; 531c-535a). Após uma introdução que justifica a importância do tema e traz as hipóteses centrais deste estudo, é descrito o contexto em que é apresentada a idéia do bem, como fundamento da principal proposição do diálogo, o governo dos filósofos. No capítulo seguinte, o terceiro, é analisado o uso de imagens, didaticamente conduzido por Sócrates a fim de convencer seus interlocutores da necessidade deste conhecimento máximo e do modo de conhecê-lo, a dialética que renuncia às imagens. Neste capítulo terceiro, são estudadas, ainda, a metáfora e a analogia, relacionadas à teoria das idéias, a fim de entender o aparente paradoxo de condenar as imagens usando-as. O objetivo do quarto capítulo é entender a analogia do Sol, que representa a função causal do bem e indica a sua condição ontológica e epistemológica, além do ser e da essência. O quinto capítulo estuda a imagem da Linha, a fim de identificar a posição do bem, como princípio não hipotético de tudo, na estrutura geral da realidade, indicando o estado da alma que o conhece, a noesis. O sexto capítulo estuda a alegoria da Caverna, seu aspecto pedagógico e a forma de conhecimento dialético capaz de alcançar a idéia do bem, demonstrando a conseqüência prática, ética e política, deste saber. / [en] This dissertation examines the idea of the good in the Republic of Plato, in the famous images of the sun (505a-509b), the line (509d-511e) and the cave (514a-521b; 531c-535a). After an introduction that justifies the importance of the subject and brings the central hypothesis of this study, it is described the context in which it is presented the idea of the good, as the fundament of the capital proposition of the dialogue, the philosophers’ government. In the following chapter, the third one, it is analyzed the use of images, conducted by Socrates to convince his interlocutors of the necessity of this maximum knowledge and the way of knowing it, the dialects that renounce images. In this third chapter, it is yet studied the metaphor and the analogy related to the theory of ideas, to understand the apparent paradox of condemning images using them. The aim of the fourth chapter is to comprehend the analogy of the sun, which represents the causal function of the good and indicates its ontological and epistemological condition, beyond being and essence. The fifth chapter studies the image of the line, to identify the position of the good, as the non hypothetical first principle of everything, in the general structure of reality, indicating the state of soul that knows it, the noesis. The sixth chapter studies the allegory of the cave, its pedagogical aspect and the form of dialectical knowledge able to achieve the idea of the good, demonstrating the practical consequence, ethical and political, of this knowledge.

Esta dissertação buscou entender como o processo de leitura da linguagem visual pode se expandir, por meio das relações que se estabelecem entre o eu - fotógrafo e o objeto a ser fotografado, através de um curso denominado Oficina de Fotografia. Adotou-se como referencial teórico os estudos de Paulo Freire sobre educação; de Ana Mae Barbosa sobre leitura de imagens; de María Acaso sobre educação visual; e de Joan Fontcuberta sobre a linguagem fotográfica. Para tal, constituiu-se um Grupo Focal com alunos de diferentes idades e profissões que haviam realizado o curso Oficina de Fotografia. Os depoimentos coletados constituem o corpus de análise dessa pesquisa. O método do Grupo Focal diz respeito à abordagem qualitativa de pesquisa, que procura conhecer o que os sujeitos envolvidos em uma determinada situação ou evento pensam a respeito de um tema proposto pelo pesquisador. Os resultados obtidos a partir da análise das entrevistas demonstram que os alunos passaram a ter uma leitura mais criteriosa não só das imagens fotográficas como também da linguagem visual em cenas cotidianas, após realizarem a Oficina de Fotografia. / This dissertation aimed to understand how the process of reading visual language can be expand by the relations that are established between the photographer itself and the object to be photographed in a course called Workshop of Photography. As theoretical reference it was used the studies from Paulo Freire about education; from Ana Mae Barbosa about images reading; from María Acaso about visual education; and from Joan Fontcuberta about photographic language. A Focus Group was organized with subjects from different ages and professions that had already attended the Workshop of Photography. The statements collected constitute the corpus of analysis of this research. The Focus Group is applied in a qualitative research methodology that aims to know what the subjects, involved in a particular situation or event, think about an issue suggested by the researcher. The obtained results based on an analysis from the interviews showed that the subjects, after attending the Workshop of Photography, have achieved a more discerning reading not only of the photographic images, but also from the visual language in everyday life scenes.

Made available in DSpace on 2015-12-10T14:24:03Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-07-31. Added 1 bitstream(s) on 2015-12-10T14:30:13Z : No. of bitstreams: 1 000853211.pdf: 1398658 bytes, checksum: c344bd138e41aaee8de5f99ea0007ff7 (MD5) / O setor florestal tem uma grande importância para a economia brasileira, sendo este, fornecedor de matéria-prima para uso em reflorestamento de áreas degradadas assim como para indústrias como de papel e celulose, carvão, moveleira, entre outras. No entanto para obtenção de bons resultados é necessária a utilização de mudas florestais sadias e livres de doenças. A mancha bacteriana do eucalipto e a mancha causada por Cylindrocladium são duas doenças muito comuns em viveiros e, embora os agentes etiológicos sejam diferentes, seus sintomas são bem parecidos, podendo causar dúvidas em um momento de diagnose. O objetivo do presente trabalho foi desenvolver uma rede neural artificial (RNA) que seja capaz de classificar as duas doenças citadas utilizando imagens digitais. Imagens de ambas doenças foram processadas, utilizando a técnica de limiarização a fim de remover o fundo da imagem, e utilizados seus histogramas para treinamento de uma RNA do tipo perceptron multicamadas com algoritmo backpropagation. Foram realizados dez ensaios com cinco topologias diferentes, sendo elas com 34, 128, 248, 368 e 511 neurônios na camada intermediária. Todas as topologias atingiram em média 95% de acertos, sendo considerada a mais adequada aquela que possuía 256 neurônios. / The forest sector has great importance for the Brazilian economy, whose, is raw materials supplier for reforestation of degraded areas as well as for industries like pulp and paper, coal, furniture, among others. However to achieve good results is necessary to use healthy forest seedlings. Bacterial stain of the eucalyptus and the stain caused by Cylindrocladium are two common diseases in nurseries and although the etiologic agents are different, their symptoms are similar, and may cause doubts in a diagnosis moment. The aim of this study was to develop an artificial neural network (ANN) to classify the two cited diseases using digital images. Images of both diseases were processed using a threshold technique to remove the image background, and their histograms were used to training a multilayer perceptron ANN with backpropagation algorithm. Ten essays with five different topologies ( 34, 128, 248, 368 and 511 neurons in the hidden layer) were carried out. All topologies reached on average 95% of correct classifications. The topology with 256 neurons in hidden layer was considered the most suitable one for this project

Made available in DSpace on 2014-12-02T11:16:54Z (GMT). No. of bitstreams: 0 Previous issue date: 2013-10-07Bitstream added on 2014-12-02T11:20:56Z : No. of bitstreams: 1 000795916.pdf: 3307453 bytes, checksum: f4acdf669e3aaa965609b2d600a6fa5a (MD5) / Nesta dissertação é discutido o desenvolvimento e a implementação de uma técnica de medida de deslocamentos utilizando uma única câmera digital, cujo foco principal é a medição de movimento e vibrações estruturais. A proposta é uma técnica de medição de não contato que tem como base a variação da posição dos pixels de um alvo ou conjunto de alvos (pontos de medições) na sequência de imagens do vídeo utilizado para capturar o movimento do alvo ou conjunto de alvos na cena. Os conceitos e o embasamento teórico são apresentados, envolvendo uma discussão de como se dá a formação e obtenção de imagens digitais, a definição e obtenção dos parâmetros intrínsecos da câmera, bem como as técnicas de processamento de imagens utilizadas para a identificação do alvo na cena. A identificação dos alvos é feita a partir da utilização de várias ferramentas de análise e processamento de imagens, dentre elas, separação de cores, detecção de bordas, thresholding, erosão e dilatação. O movimento do alvo na sequência de imagens é rastreado em termos de pixels e posteriormente é feita a conversão para unidade de medida utilizando a relação unidade de medida/pixel obtida através de um processo de calibração. A avaliação e validação da proposta foram feitas através de testes experimentais, envolvendo tanto medidas estáticas como dinâmicas. Numa etapa final, a técnica foi utilizada para a análise de vibração de um prédio em escala reduzida, mais especificamente, a análise modal com base apenas na resposta da estrutura. Os resultados e análises realizadas mostraram que a metodologia implementada com base na utilização de uma única câmera se mostrou como uma proposta com alto potencial de aplicação como ferramenta de medição de não-contato, relativamente simples e efetiva / In this work is discussed the development and implementation of a technique for measuring of displacements using a single digital camera, whose main focus is the measurement of structural motion and vibrations. The proposal is a non-contact measurement technique, which is based on the variation of the position of the pixels of the target measuring points in the sequence of images of a video used to capture the movement of the target points in the scene. The concepts and theoretical background are presented, involving a discussion concerning to the formation and obtaining of digital images, definition and obtaining of intrinsic parameters of the camera, as well as the image processing techniques used to identify the measuring target points in the scene. The identification of the targets is made from the use of different analysis and image processing tools such as, color separation, edge detection, thresholdding, erosion and dilation used in a complementary way. The movement of the target points in the sequence of image is tracked in terms of pixels and then they are converted in measurement unit using the ratio measurement unit/pixel obtained by a calibration process. The evaluation and validation of the proposal were made through experimental tests, involving both static and dynamic measurements. At a final stage, the technique was used to vibration analyze of a reduced scale building structure, more specifically, the output-only modal analysis of the structure. The results and analyzes showed that the methodology implemented based on the use of a unique camera provided to be a proposal with high application potential as a non-contact measurement tool, relatively simple and effective

Ce mémoire de thèse porte sur l’élaboration de systèmes de reconnaissance d’image qui sont robustes à la variabilité géométrique. La compréhension d’une image est un problème difficile, de par le fait qu’elles sont des projections en deux dimensions d’objets 3D. Par ailleurs, des représentations qui doivent appartenir à la même catégorie, par exemple des objets de la même classe en classification, peuvent être visuellement très différentes. Notre but est de rendre ces systèmes robustes à la juste quantité de déformations, celle-ci étant automatiquement déterminée à partir des données. Nos deux contributions sont les suivantes. Nous montrons tout d’abord comment utiliser des exemples virtuels pour rendre les systèmes de classification d’images robustes et nous proposons ensuite une méthodologie pour apprendre des descripteurs de bas niveau robustes, pour la recherche d’image.Nous étudions tout d’abord les exemples virtuels, en tant que transformations de vrais exemples. En représentant une image en tant que sac de descripteurs transformés, nous montrons que l’augmentation de données, c’est-à-dire le fait de les considérer comme de nouveaux exemples iid, est la meilleure manière de les utiliser, pourvu qu’une étape de vote avec les descripteurs transformés soit opérée lors du test. Du fait que les transformations apportent différents niveaux d’information, peuvent être redondants, voire nuire à la performance, nous pro-posons un nouvel algorithme capable de sélectionner un petit nombre d’entre elles,en maximisant la justesse de classification. Nous montrons par ailleurs comment remplacer de vrais exemples par des virtuels, pour alléger les couts d’annotation.Nous rapportons de bons résultats sur des bancs d’essai de classification.Notre seconde contribution vise à améliorer les descripteurs de régions locales utilisés en recherche d’image, et en particulier nous proposons une alternative au populaire descripteur SIFT. Nous proposons un nouveau descripteur, appelé patch-CKN, appris sans supervision. Nous introduisons un nouvel ensemble de données liant les images et les imagettes, construit à partir de reconstruction3D automatique d’images récupérées sur Internet. Nous définissons une méthode pour tester précisément la performance des descripteurs locaux au niveau de l’imagette et de l’image. Notre approche dépasse SIFT et les autres approches à base d’architectures convolutionnelles sur notre banc d’essai, et d’autres couramment utilisés dans la littérature. / This dissertation focuses on designing image recognition systems which are robust to geometric variability. Image understanding is a difficult problem, as images are two-dimensional projections of 3D objects, and representations that must fall into the same category, for instance objects of the same class in classification can display significant differences. Our goal is to make systems robust to the right amount of deformations, this amount being automatically determined from data. Our contributions are twofolds. We show how to use virtual examples to enforce robustness in image classification systems and we propose a framework to learn robust low-level descriptors for image retrieval. We first focus on virtual examples, as transformation of real ones. One image generates a set of descriptors –one for each transformation– and we show that data augmentation, ie considering them all as iid samples, is the best performing method to use them, provided a voting stage with the transformed descriptors is conducted at test time. Because transformations have various levels of information, can be redundant, and can even be harmful to performance, we propose a new algorithm able to select a set of transformations, while maximizing classification accuracy. We show that a small amount of transformations is enough to considerably improve performance for this task. We also show how virtual examples can replace real ones for a reduced annotation cost. We report good performance on standard fine-grained classification datasets. In a second part, we aim at improving the local region descriptors used in image retrieval and in particular to propose an alternative to the popular SIFT descriptor. We propose new convolutional descriptors, called patch-CKN, which are learned without supervision. We introduce a linked patch- and image-retrieval dataset based on structure from motion of web-crawled images, and design a method to accurately test the performance of local descriptors at patch and image levels. Our approach outperforms both SIFT and all tested approaches with convolutional architectures on our patch and image benchmarks, as well as several styate-of-theart datasets.