Papers by Cinthia Freitas
Anais do X Simpósio Brasileiro de Segurança da Informação e de Sistemas Computacionais (SBSeg 2010)
Atualmente é evidente que o cibercrime é um fenômeno de dimensão global. Espera-se que as provas ... more Atualmente é evidente que o cibercrime é um fenômeno de dimensão global. Espera-se que as provas necessárias sejam encontradas pelos peritos, mas algumas vezes é difícil encontrá-las ou isto se torna um trabalho árduo. Sem evidências o suspeito não pode ser acusado e processado. Em um contexto de negócios, existem muitos tipos de evidências em potencial. Citamse os arquivos de log, e-mails e computadores pessoais. O presente artigo apresenta um mecanismo de agrupamento e classificação das conversações por e-mails. Os resultados experimentais, que atingem taxas de acerto próximas de 98%, demonstraram que é possível usar o rastreamento em relacionamentos por e-mails para a produção de provas digitais.
Proceedings. 17th Brazilian Symposium on Computer Graphics and Image Processing
This paper presents a multiple classifier system applied to the handwritten word recognition (HWR... more This paper presents a multiple classifier system applied to the handwritten word recognition (HWR) problem. The goal is to analyse the influence of different global classifiers taken isolatedly as well as combined in a particular HWR task. The application proposed is the recognition of the Portuguese handwritten names of the months. The strategy takes advantage of the complementary mechanisms of three different classifiers: Conventional Neural Network, Class-Modular Neural Network and Hidden Markov Models, yielding a multiple classifier that is more efficient than either individual technique. The recognition rates obtained vary from 75.9% using the stand alone HMM classifier to 96.0% considering the classifiers combination.
J. Univers. Comput. Sci., 2008
In this paper we provide a study about crime scenes and its features used in criminal investigati... more In this paper we provide a study about crime scenes and its features used in criminal investigations. We argue that the crime scene provides a large set of features that can be used to corroborate the conclusions emitted by the experts. We also propose a set of features to classify the violent crime considering two classes: attack from inside or outside of the scene. The classification stage is based on conventional MLP (Multiple-Layer Perceptron) Neural Network and SVM (Support Vector Machine). The experimental results reveal an error rate of 30.3% (MLP), 22.8% (SVM-linear), and 19.4% (SVM-polynomial) using a database composed of 400 crime scenes. This paper presents an experiment based on a stereoscopic projection that allows to experts analyze and take decisions about the crime scene and its dynamic.
Jovens que escolheram ser empreendedores do meio rural mostram que e possivel lucrar, viver bem e... more Jovens que escolheram ser empreendedores do meio rural mostram que e possivel lucrar, viver bem e ganhar o mundo sem sair da propriedade. Conheca historias de quem esta mudando a cara da agricultura catarinense
Reconhecimento da Organizacao das Nacoes Unidas para Alimentacao e Agricultura da visibilidade in... more Reconhecimento da Organizacao das Nacoes Unidas para Alimentacao e Agricultura da visibilidade internacional ao trabalho realizado em Santa Catarina
Antes de o sol nascer, Antonio Barea toma seu chimarrao e se prepara para enfrentar o dia de trab... more Antes de o sol nascer, Antonio Barea toma seu chimarrao e se prepara para enfrentar o dia de trabalho sabendo que nao precisa mais se preocupar com a conta de energia eletrica. Desde setembro do ano passado, 114 paineis solares com capacidade instalada de 36 quilowatts-pico fornecem energia para a propriedade rural localizada em Planalto Alegre, no Oeste do Estado. Embora o principal impulso para os agricultores investirem na energia solar seja economico, ao aproveitar o sol para produzir energia limpa em casa, eles estao fazendo muito mais para o planeta do que para si. Isso porque o aumento da geracao de energias alternativas alivia a carga do sistema eletrico e reduz a pressao pela construcao de usinas hidreletricas e termoeletricas, por exemplo, que trazem grandes impactos sociais e ambientais.
J. Univers. Comput. Sci., 2008
The contribution of this paper is twofold. First we investigate the use of the confusion matrices... more The contribution of this paper is twofold. First we investigate the use of the confusion matrices in order to get some insight to better define perceptual zoning for character recognition. The features considered in this work are based on concavities/convexities deficiencies, which are obtained by labelling the background pixels of the input image. Four different perceptual zoning (symmetrical and non-symmetrical) are discussed. Experiments show that this mechanism of zoning could be considered as a reasonable alternative to exhaustive search algorithms. The second contribution is a methodology to define metaclasses for the problem of handwritten character recognition. The proposed approach is based on the disagreement among the characters and it uses Euclidean distance computed between the confusion matrices. Through comprehensive experiments we demonstrate that the use of metaclasses can improve the performance of the system.
Proceedings of the International Conference on e-Business, 2008
This paper discusses the security problems on the contracts deals over the Internet and describes... more This paper discusses the security problems on the contracts deals over the Internet and describes a protocol of authenticity that will keep audit trails from the activities during the Web hiring. These audit trails will be stored digitally, either on the side of the provider or on the consumer. It is understood that it is legal obligation of the provider to logs and ensure the integrity of the data related to the operations on the Internet-based commerce. because in situations of dispute can occur a reversal of the burden of proof. Thus, this prototype ensures confidence on the e-contracts, logging relevant information that help in the identification of the parties, using a Plug-in software installed on the e-commerce provider and in the consumer machine, executable in the Web. Moreover, it is important to remind about the security in contracts agreements over the Internet is an essential feature because it allows the consumer a guarantee of contract award, since it maintains the integrity of the document, and also can be presented as evidence to the Judiciary, helping in litigation and satisfying the premises of the legal acceptance of digital documents.
Anais de XXIX Simpósio Brasileiro de Telecomunicações, 2011
Resumo-Neste artigo discute-se uma abordagem de reconhecimento de palavras manuscritas dependente... more Resumo-Neste artigo discute-se uma abordagem de reconhecimento de palavras manuscritas dependente do escritor para léxico livre. Diferentemente das abordagens mais corriqueiras, com escritor independente e léxico dependente, em que o mesmo modelo de aprendizagem de máquina é utilizado para todos os escritores, na abordagem dependente do escritor cada escritor tem o seu próprio modelo. A desvantagem de ter vários modelos é compensada por uma melhora desempenho e a não vinculação a um léxico específico. Os resultados experimentais obtidos com os quatro escritores distintos, cada um com um banco de dados de aproximadamente 4.000 palavras, produziu taxas de reconhecimento de 83% à 93%. Palavras-Chave-Reconhecimento de manuscritos, Reconhecimento de palavras, Dependente do escritor.
Revista Eletrônica de Sistemas de Informação, 2004
Este artigo descreve uma metodologia para seleção de classes de símbolos a partir de classesde gr... more Este artigo descreve uma metodologia para seleção de classes de símbolos a partir de classesde grafemas em um sistema de reconhecimento de palavras manuscritas do extenso de cheques bancáriosbrasileiros baseado em HMM (Hidden Markov Models). Este artigo discute as definições de primitivas,grafemas e símbolos considerando um enfoque Global para o reconhecimento das palavras, o qual evita asegmentação das palavras em letras ou pseudo-letras utilizando HMM. Assim, a entrada para os modelosconsiste em uma descrição da palavra a partir de um alfabeto de símbolos gerados a partir dos grafemasextraídos das imagens das palavras, sendo esta a representação visível para o HMM. Portanto, a idéia éintroduzir uma conceituação de alto nível, tais como primitivas perceptivas (laços, ascendentes,descendentes, concavidades e convexidades) e fornecer um modo de retro-alimentação rápido e informativosobre a informação contida em cada classe de grafema, permitindo uma seleção de classes de símbolos. Oa...
Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition
This work presents a baseline system used to evaluate feature sets for word recognition. The main... more This work presents a baseline system used to evaluate feature sets for word recognition. The main goal is to determine an optimum feature set to represent the handwritten names for the months of the year in Brazilian Portuguese language. Three kinds of features are evaluated: perceptual, directional and topological. The evaluation shows that taken in isolation, the perceptual feature set produces the best results for the lexicon used. These results can be further improved combining the feature sets. The baseline system developed obtains an average recognition rate of 87%. This can be considered a good result considering that no explicit segmentation is performed.
Proceedings of Sixth International Conference on Document Analysis and Recognition, 2001
This paper presents the application of Mutual Information criterion to validate feature sets extr... more This paper presents the application of Mutual Information criterion to validate feature sets extracted from handwritten words in Brazilian legal amounts. The lexicon includes a subset of short words without ascenders /descenders and subsets of words with the same prefix or suffix. These particularities of the Brazilian lexicon show that is necessary to improve the perpetual feature set with complementary geometric features, and also modeling the prefix and suffix of the words. Finally, the experiments show the viability of our approach.
Proceedings. XV Brazilian Symposium on Computer Graphics and Image Processing
This paper evaluates NN and HMM classifiers applied to the handwritten word recognition problem. ... more This paper evaluates NN and HMM classifiers applied to the handwritten word recognition problem. The goal is analyse the individual and combined performance of these classifiers. They are evaluated considering two different combination strategies and the experiments are performed with the same database and similar feature sets. The strategy proposed takes advantage of the different but complementary mechanisms of NN and HMM to obtain a more efficient hybrid classifier. The recognition rates obtained vary from 75.9% using the HMM classifier alone to 90.4% considering the NN and HMM combination.
Lecture Notes in Computer Science, 2011
This paper presents a two-level based character recognition method in which a dynamically selecti... more This paper presents a two-level based character recognition method in which a dynamically selection of the most promising zoning scheme for feature extraction allows us to obtain interesting results for character recognition. The first level consists of a conventional neural network and a look-up-table that is used to suggest the best zoning scheme for a given unknown character. The information provided by the first level drives the second level in the selection of the appropriate feature extraction method and the corresponding class-modular neural network. The experimental protocol has shown significant recognition rates for handwritten characters (from 80.82% to 88.13%).
Handwritten Character Recognition Using Nonsymmetrical Perceptual Zoning
International Journal of Pattern Recognition and Artificial Intelligence, 2007
In this paper we present an alternative strategy to define zoning for handwriting recognition, wh... more In this paper we present an alternative strategy to define zoning for handwriting recognition, which is based on nonsymmetrical perceptual zoning. The idea is to extract some knowledge from the confusion matrices in order to make the zoning process less empirical. The feature set considered in this work is based on concavities/convexities deficiencies, which are obtained by labeling the background pixels of the input image. To better assess the nonsymmetrical zoning we carried out experiments using four different zonings strategies. Experiments show that the nonsymmetrical zoning could be considered as a tool to build more reliable handwriting recognition systems.
2010 IEEE 24th International Conference on Advanced Information Networking and Applications Workshops
2010 IEEE 24th International Conference on Advanced Information Networking and Applications Workshops, 2010
Zoning and metaclasses for character recognition
Proceedings of the 2007 ACM symposium on Applied computing - SAC '07, 2007
Lecture Notes in Computer Science, 2007
We present a methodology to analyze Multiple Classifiers Systems (MCS) performance, using the dis... more We present a methodology to analyze Multiple Classifiers Systems (MCS) performance, using the disagreement concept. The goal is to define an alternative approach to the conventional recognition rate criterion, which usually requires an exhaustive combination search. This approach defines a Distance-based Disagreement (DbD) measure using an Euclidean distance computed between confusion matrices and a soft-correlation rule to indicate the most likely candidates to the best classifiers ensemble. As case study, we apply this strategy to two different handwritten recognition systems. Experimental results indicate that the method proposed can be used as a low-cost alternative to conventional approaches.
Lecture Notes in Computer Science, 2007
In this article we propose a feature extraction procedure based on directional histograms and inv... more In this article we propose a feature extraction procedure based on directional histograms and investigate the application of a nonconventional neural network architecture, applied to the problem of handwritten character recognition. This approach is inspired on some characteristics of the human visual system, as it focus attention on high spatial frequencies and on the recognition of local features. Two architectures were tested and evaluated: a conventional MLP (Multiple Layer Perceptron) and a class-modular MLP. Experiments developed with the Letter database produced a recognition rate of 93.67% for the classmodular MLP. Other set of experiments utilized the IRONOFF database resulting in recognition rates of 89.21% and 80.75% for uppercase and lowercase characters respectively, also with the class-modular MLP.
Brazilian bank check handwritten legal amount recognition
Proceedings 13th Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00878)
Uploads
Papers by Cinthia Freitas