The design of a machine which reads unconstrained words still remains an unsolved problem. For example, automatic interpretation of handwritten documents by a computer is still under research. Most systems attempt to segment words into letters and read words one character at a time. However, segmenting handwritten words is very difficult. So to avoid this words are treated as a whole. This research investigates a number of features computed from whole words for the recognition of handwritten words in particular. Arabic text classification and recognition is a complicated process compared to Latin and Chinese text recognition systems. This is due to the nature cursiveness of Arabic text. The work presented in this thesis is proposed for word based recognition of handwritten Arabic scripts. This work is divided into three main stages to provide a recognition system. The first stage is the pre-processing, which applies efficient pre-processing methods which are essential for automatic recognition of handwritten documents. In this stage, techniques for detecting baseline and segmenting words in handwritten Arabic text are presented. Then connected components are extracted, and distances between different components are analyzed. The statistical distribution of these distances is then obtained to determine an optimal threshold for word segmentation. The second stage is feature extraction. This stage makes use of the normalized images to extract features that are essential in recognizing the images. Various method of feature extraction are implemented and examined. The third and final stage is the classification. Various classifiers are used for classification such as K nearest neighbour classifier (k-NN), neural network classifier (NN), Hidden Markov models (HMMs), and the Dynamic Bayesian Network (DBN). To test this concept, the particular pattern recognition problem studied is the classification of 32492 words using ii the IFN/ENIT database. The results were promising and very encouraging in terms of improved baseline detection and word segmentation for further recognition. Moreover, several feature subsets were examined and a best recognition performance of 81.5% is achieved.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

As the use of technology for personal, professional, and learning purposes increases, more and more assessments are transitioning from a traditional paper-based testing format to a computer-based one. During this transition, some assessments are being offered in both paper and computer formats in order to accommodate examinees and testing center capabilities. Scores on the paper-based test are often intended to be directly comparable to the computer-based scores, but such claims of comparability are often unsupported by research specific to that assessment. Not only should the scores be examined for differences, but the thought processes used by raters while scoring those assessments should also be studied to better understand why raters might score response modes differently. Previous comparability literature can be informative, but more contemporary, test-specific research is needed in order to completely support the direct comparability of scores. The goal of this thesis was to form a more complete understanding of why analytic scores on a writing assessment might differ, if at all, between handwritten and typed responses. A representative sample of responses to the writing composition portion of a large-scale high school equivalency assessment were used. Six trained raters analytically scored approximately six-hundred examinee responses each. Half of those responses were typed, and the other half were the transcribed handwritten duplicates. Multiple methods were used to examine why differences between response modes might exist. A MANOVA framework was applied to examine score differences between response modes, and the systematic analyses of think-alouds and interviews were used to explore differences in rater cognition. The results of these analyses indicated that response mode was of no practical significance, meaning that domain scores were not notably dependent on whether or not a response was presented as typed or handwritten. Raters, on the other hand, had a more substantial effect on scores. Comments from the think-alouds and interviews suggest that, while the scores were not affected by response mode, raters tended to consider certain aspects of typed responses differently than handwritten responses. For example, raters treated typographical errors differently from other conventional errors when scoring typed responses, but not while scoring the handwritten duplicates. Raters also indicated that they preferred scoring typed responses over handwritten ones, but felt they could overcome their personal preferences to score both response modes similarly. Empirical investigations on the comparability of scores, combined with the analysis of raters’ thought processes, helped to provide a more evidence-based answer to the question of why scores might differ between response modes. Such information could be useful for test developers when making decisions regarding what mode options to offer and how to best train raters to score such assessments. The design of this study itself could be useful for testing organizations and future research endeavors, as it could be used as a guide for exploring score differences and the human-based reasons behind them.

Given the accessibility of media devices available to us today and utilising van Leeuwen's concept of inscription and synthesis as a guide, this thesis explores the practice of re-presenting a domestic material object, the Croxley Recipe Book, into digital media. Driven by a creative practice research method, but also utilising materiality, digital storytelling practices and modality as important conceptual frames, this project was fundamentally experimental in nature. A materiality-framed content analysis, interpreted through cultural analysis, initially unraveled some of the cookbook's significance and contextualised it within a particular time of New Zealand's cultural history. Through the expressive and anecdotal practice of digital storytelling the cookbook's significance was further negotiated, especially as the material book was engaged with through the affective and experiential digital medium of moving-image. A total of six digital film works were created on an accompanying DVD, each of which represents some of the cookbook's significance but approached through different representational strategies. The Croxley Recipe Book Archive Film and Pav. Bakin' with Mark are archival documentaries, while Pav is more expressive and aligned with the digital storytelling form. Spinning Yarns and Tall Tales, a film essay, engages and reflects with the multiple processes and trajectories of the project, while Extras and The Creative Process Journal demonstrate the emergent nature of the research. The written thesis discusses the emergent nature of the research process and justifies the conceptual underpinning of the research.

Pagrindinis darbo tikslas yra pritaikant dirbtinio intelekto algoritmus sukurti ranka rašyto teksto atpažinimo įrankį. Siekiant šio tikslo buvo apžvelgti dirbtinio intelekto atpažinimo metodai, atlikta teksto atpažinimo algoritmų analizė. Remiantis analizės rezultatais, sukurta ranka rašyto teksto atpažinimo programa, kurioje teksto segmentavimo operacija atliekama „lašelio aptikimo“ algoritmu. Teksto atpažinimo operacijai atlikti naudojamas bitų masyvų analizės algoritmas. / The aim of this work is to create an application for handwritten text recognition using artificial intelligence algorithms. For this purpose a number of recognition methods based on artificial intelligence were reviewed. Based on the review information an application was created for the purpose of recognizing handwritten text. The text segmentation was implemented using a blob detection algorithm. Text recognition was performed using bit array analysis algorithm. During the implementation and testing stage the main problem areas of such application were identified.

Marginalijos – svarbus, informatyvus knygos elementas, atskleidžiantis ne tik marginalijų ypatybes, sukūrimo laiką, vietą, bet ir jų kūrėjo visai kitą asmenybės pusę. Bakalauro darbe aptariama marginalijų samprata, tyrimų kryptys ir vienas iš sudėtingiausių aspektų – marginalijų klasifikacija. Darbe analizuojamos Ž. Mikšio ŠU bibliotekai dovanotų knygų marginalijos, kurios sudaro tokias grupes: paprastosios arba bendrosios marginalijos, taisančiosios marginalijos ir papildoma informacija. Pagal marginalijų kūrėjo ryšį su tekstu, jo pateikiamią vertinimą šios grupės suskirstytos į smulkesnius pogrupius. Atlikus tyrimą, paaiškėjo, kad didžiąją marginalijų dalį sudaro paprastosios marginalijos, turinčios neigiamą reikšmę. Ž. Mikšio kritiški įrašai knygose atskleižia jo požiūrį į Lietuvoje leidžiamas knygas, jų kokybę, leidėjus. Taisančiųjų marginalijų ir papildomos informacijos pagrindinė funkcija – pagalba skaitytojui. Prieduose pateikiamas šaltinių sąrašas, kurį sudaro 297 knygos, iš kurių buvo išrinktos marginalijos. Taip pat prieduose pridėti įdomesni, informatyvesni Ž. Mikšio marginalijų pavyzdžiai. / Marginalia is an important and informative element of the book. What is more, it reveals not only the characteristics of marginalia, the creation of time, place, but also includes a completely different personality side of the creator. These Bachelor's thesis discusses about the concept of marginalia, research directions, and also focuses on one of the most challenging aspects – classification of marginalia. Moreover, this paper aims to provide analysis of marginalia, which are selected from the donated Ž. Mikšys books to Šiauliai university library. In this case they consist of the following groups: common marginalia, emendation marginalia and additional information. According to marginalia creator’s connection with the text and his presented evaluation, it has been decided to divide this group into smaller sub-groups. The investigation has shown that the majority of the marginality consist of ordinary marginality, which have negative meanings. Criticized Ž. Mikšys records in the books reveal his approach regarding the books published in Lithuania, their quality and publishers. The main feature of emendation marginalia and additional information – the aid to a reader. In the annexes is given a list of sources which indicates 279 books, and all marginalia have been selected from the presented books. It is important to mention that annexes include more interesting and informative examples of Ž. Mikšys marginalia as well.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The author presents novel algorithms to design unconstrained handwriting recognition systems organized in three parts: In Part One, novel algorithms are presented for processing of Arabic text prior to recognition. Algorithms are described to convert a thinned image of a stroke to a straight line approximation. Novel heuristic algorithms and novel theorems are presented to determine start and end vertices of an off-line image of a stroke. A straight line approximation of an off-line stroke is converted to a one-dimensional representation by a novel algorithm which aims to recover the original sequence of writing. The resulting ordering of the stroke segments is a suitable preprocessed representation for subsequent handwriting recognition algorithms as it helps to segment the stroke. The algorithm was tested against one data set of isolated handwritten characters and another data set of cursive handwriting, each provided by 20 subjects, and has been 91.9% and 91.8% successful for these two data sets, respectively. In Part Two, an entirely novel fuzzy set-sequential machine character recognition system is presented. Fuzzy sequential machines are defined to work as recognizers of handwritten strokes. An algorithm to obtain a deterministic fuzzy sequential machine from a stroke representation, that is capable of recognizing that stroke and its variants, is presented. An algorithm is developed to merge two fuzzy machines into one machine. The learning algorithm is a combination of many described algorithms. The system was tested against isolated handwritten characters provided by 20 subjects resulting in 95.8% recognition rate which is encouraging and shows that the system is highly flexible in dealing with shape and size variations. In Part Three, also an entirely novel text recognition system, capable of recognizing off-line handwritten Arabic cursive text having a high variability is presented. This system is an extension of the above recognition system. Tokens are extracted from a onedimensional representation of a stroke. Fuzzy sequential machines are defined to work as recognizers of tokens. It is shown how to obtain a deterministic fuzzy sequential machine from a token representation that is capable'of recognizing that token and its variants. An algorithm for token learning is presented. The tokens of a stroke are re-combined to meaningful strings of tokens. Algorithms to recognize and learn token strings are described. The. recognition stage uses algorithms of the learning stage. The process of extracting the best set of basic shapes which represent the best set of token strings that constitute an unknown stroke is described. A method is developed to extract lines from pages of handwritten text, arrange main strokes of extracted lines in the same order as they were written, and present secondary strokes to main strokes. Presented secondary strokes are combined with basic shapes to obtain the final characters by formulating and solving assignment problems for this purpose. Some secondary strokes which remain unassigned are individually manipulated. The system was tested against the handwritings of 20 subjects yielding overall subword and character recognition rates of 55.4% and 51.1%, respectively.

abstract: Handwritten documents have gained popularity in various domains including education and business. A key task in analyzing a complex document is to distinguish between various content types such as text, math, graphics, tables and so on. For example, one such aspect could be a region on the document with a mathematical expression; in this case, the label would be math. This differentiation facilitates the performance of specific recognition tasks depending on the content type. We hypothesize that the recognition accuracy of the subsequent tasks such as textual, math, and shape recognition will increase, further leading to a better analysis of the document. Content detection on handwritten documents assigns a particular class to a homogeneous portion of the document. To complete this task, a set of handwritten solutions was digitally collected from middle school students located in two different geographical regions in 2017 and 2018. This research discusses the methods to collect, pre-process and detect content type in the collected handwritten documents. A total of 4049 documents were extracted in the form of image, and json format; and were labelled using an object labelling software with tags being text, math, diagram, cross out, table, graph, tick mark, arrow, and doodle. The labelled images were fed to the Tensorflow’s object detection API to learn a neural network model. We show our results from two neural networks models, Faster Region-based Convolutional Neural Network (Faster R-CNN) and Single Shot detection model (SSD). / Dissertation/Thesis / Masters Thesis Computer Science 2018

Made available in DSpace on 2014-06-12T15:58:13Z (GMT). No. of bitstreams: 2 arquivo3310_1.pdf: 8155353 bytes, checksum: 2f4dcd5adb2b0b1a23c40bf343b36b34 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2011 / Faculdade de Amparo à Ciência e Tecnologia do Estado de Pernambuco / Ensemble of Classifiers (EoC) é uma nova alternative para alcançar altas taxas de reconhecimento em sistemas de reconhecimento de padrões. O uso de ensemble é motivado pelo fato de que classificadores diferentes conseguem reconhecer padrões diferentes, portanto, eles são complementares. Neste trabalho, as metodologias de EoC são exploradas com o intuito de melhorar a taxa de reconhecimento em diferentes problemas. Primeiramente o problema do reconhecimento de caracteres é abordado. Este trabalho propõe uma nova metodologia que utiliza múltiplas técnicas de extração de características, cada uma utilizando uma abordagem diferente (bordas, gradiente, projeções). Cada técnica é vista como um sub-problema possuindo seu próprio classificador. As saídas deste classificador são utilizadas como entrada para um novo classificador que é treinado para fazer a combinação (fusão) dos resultados. Experimentos realizados demonstram que a proposta apresentou o melhor resultado na literatura pra problemas tanto de reconhecimento de dígitos como para o reconhecimento de letras. A segunda parte da dissertação trata da seleção dinâmica de classificadores (DCS). Esta estratégia é motivada pelo fato que nem todo classificador pertencente ao ensemble é um especialista para todo padrão de teste. A seleção dinâmica tenta selecionar apenas os classificadores que possuem melhor desempenho em uma dada região próxima ao padrão de entrada para classificar o padrão de entrada. É feito um estudo sobre o comportamento das técnicas de DCS demonstrando que elas são limitadas pela qualidade da região em volta do padrão de entrada. Baseada nesta análise, duas técnicas para seleção dinâmica de classificadores são propostas. A primeira utiliza filtros para redução de ruídos próximos do padrão de testes. A segunda é uma nova proposta que visa extrair diferentes tipos de informação, a partir do comportamento dos classificadores, e utiliza estas informações para decidir se um classificador deve ser selecionado ou não. Experimentos conduzidos em diversos problemas de reconhecimento de padrões demonstram que as técnicas propostas apresentam um aumento de performance significante