2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2018
Electroencephalography (EEG) signals may be severely affected by noise originated from various so... more Electroencephalography (EEG) signals may be severely affected by noise originated from various sources due to their low amplitude nature, specially if they are collected from scalp sensors. Several methods have been proposed for EEG denoising in order to facilitate diagnosis and communication in brain-computer interfaces, but such algorithms often have high complexity. This work presents a denoising approach based on deep learning using a deep convolutional autoencoder, which should reduce the effort of projecting denoising filters. Experiments were performed using two types of noise, originated from eye blink and from jaw clenching. Performance was evaluated with peak signal-to-noise ratio (PSNR) and the results showed that all confidence intervals for the proposed approach were superior to those obtained by the baseline bandpass traditional filtering method. Best average PSNR results for eye blink were obtained for Cz channels with $(20.3\pm 2.6)\mathrm{d}\mathrm{B}$ versus $(14.3\pm 2.4)\mathrm{d}\mathrm{B}$. For jaw clenching, best average PSNR results were obtained for Fz channels with $(21.7\pm 3.1)\mathrm{d}\mathrm{B}$ versus $(13.9\pm 2.6)\mathrm{d}\mathrm{B}$. The proposed approach seems to open a promising scope of research for noise filtering in EEG.
In emotion recognition using EEG, it is not generally agreed upon how much time an EEG signal seq... more In emotion recognition using EEG, it is not generally agreed upon how much time an EEG signal sequence must have in order to maximize precision and recall rates. To the best of our knowledge, there is not a systematic evaluation of effects on classif er performance related to EEG signal durations. The human factors related to attention decreasing and tiredness increasing have imposed diff culties to create EEG datasets containing a rich variation of signal samples. This paper proposes an experimental evaluation of three different EEG datasets (DEAP, MAHNOB, and STEED) each one mainly characterized by short, intermediate and long signal (or stimulus) durations. Statistical evaluation pointed out that for an EEG dataset to be well-suited for emotion recognition it should have two main characteristics: emotion stimulus data should be publicly available and evaluated by worldwide volunteers, and media stimulus should have duration long enough to affect the subjects. Our statistical analysis revealed that, at least for the considered datasets, signals with duration longer than 60 seconds allow better classif cation results. This work did not analyse the impact to humans of longer stimulus media.
Resumo Este artigo apresenta uma tradução de partes do capítulo 16 do Tripura Rahasya seguidas de... more Resumo Este artigo apresenta uma tradução de partes do capítulo 16 do Tripura Rahasya seguidas de discussões e análises tendo como base conceitos apresentados no Yoga Sutra de Patanjali. O capítulo 16 do Tripura Rahasya trata dentre outros fatores do conceito de indizível na tradição indiana e de métodos para alcançar o estado em que é possível acessar o indizível, ou aquilo que não pode ser explicado por palavras. Argumentamos que o Rāja-Yoga de Patañjali é uma metodologia adequada para experimentar esse estado. Abstract This paper presents a translation of parts of Chapter 16 of Tripura Rahasya followed by discussion and analysis based on concepts presented in the Patanjali's Yoga Sutra. Chapter 16 of Tripura Rahasya focuses on, among other factors, the concept of unutterable in the Indian tradition and methods to reach the state where you can access the unsayable, or what can not be explained by words. We argue that the Patanjali's Rāja-Yoga is an appropriate methodology to experience this state.
Recent advances in deep learning methodologies are enabling the construction of more accurate cla... more Recent advances in deep learning methodologies are enabling the construction of more accurate classifiers. However, existing labeled face datasets are limited in size, which prevents CNN models from reaching their full generalization capabilities. A variety of techniques to generate new training samples based on data augmentation have been proposed, but the great majority is limited to very simple transformations. The approach proposed in this paper takes into account intrinsic information about human faces in order to generate an augmented dataset that is used to train a CNN, by creating photo-realistic smooth face variations based on Active Appearance Models optimized for human faces. An experimental evaluation taking CNN models trained with original and augmented versions of the MORPH face dataset allowed an increase of 10 in the F-Score and yielded Receiver Operating Characteristic curves that outperformed state-of-the-art work in the literature.
obtained for all other uses, in any current or future media, including reprinting/republishing th... more obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
This paper deals with two problems: (1) the selection of a set of music features in order to achi... more This paper deals with two problems: (1) the selection of a set of music features in order to achieve high genre classification accuracies; (2) the absence of a representative music dataset of regional brazilian music. In this paper, we propose a set of features to classify genres of music. The features proposed were obtained by a methodical selection of important features used in the literature of Music Information Retrieval (MIR) and Music Emotion Recognition (MER). Besides, we propose a new music dataset called BMD (Brazilian Music Dataset) 1 , containing 120 songs labeled in 7 musical genres: Forró, Rock, Repente, MPB(M ´ usica Popular Brasileira-Brazilian Popular Music), Brega, Sertanejo and Disco. An important characteristic of this new dataset compared with others, is the presence of three popular genres in Brazil Northeast region: Repente, Brega and a characteristic genre similar to MPB, which we also call as MPB. We evaluated our proposed features on both datasets: GTZAN and BMD. The proposed approach achieved average accuracy (after 30 runs of 5-fold-cross-validations) of 79.7% for GTZAN and 86.11% for the BMD. Another important contribution of this work is random repetition of cross-validation executions. Most of the papers performs only a single n-fold cross-validation. We criticize that practice and propose, at least, 30 random executions to compute the average accuracy.
Neste artigo, é apresentada uma análise quantitativa de como a qualidade das interações entre pro... more Neste artigo, é apresentada uma análise quantitativa de como a qualidade das interações entre professores e alunos afetam o desempenho acadêmico. Foram analisadas respostas de 131 alunos a um questionário contendo 14 perguntas. Três tipos de análises foram realizadas: análise descritiva, testes de hipótese e classificação dos alunos em desempenho positivo e desempenho negativo de acordo com as respostas dos questionários submetidas a uma máquina de vetores de suporte (SVM). A partir da análise descritiva verificou-se que um fator muito importante para o desempenho acadêmico dos alunos é o quanto eles procuram os professores em horário de antedimento. Embora exista uma vasta literatura com pesquisas qualitativas sobre como o relacionamento professor-aluno afeta o desempenho dos alunos, os testes de hipótese realizados nesta pesquisa apontaram que os dados coletados não apresentam evidências suficientes para rejeitar a hipótese de que não há diferenças significativas entre os desempenhos de alunos com bom e mau relacionamento com os professores. Por fim, foi possível obter acurácia média de classificação de 60,46% treinando uma SVM utilizando respostas de apenas 10 questões.
After the publication of “Emotional Intelligence” by Daniel Goleman, much interest has been demon... more After the publication of “Emotional Intelligence” by Daniel Goleman, much interest has been demonstrated by education researchers regarding the student qualities responsible for academic success. In this paper, we propose an online software environment to aid the implementation of a methodology focused on three important aspects of student performance: psychological well-being, physical well-being, and studying techniques. Psychological well-being techniques used in our methodology aim to improve attention and emotional skills, physical well-being techniques aim to create good health habits, and studying techniques aim to help students personal time management and teach students how to approach their course materials in order to learn faster and with more quality.
RITA: Revista de Informática Teórica e Aplicada, Nov 2014
Among all cancer types, breast cancer is the one with the second highest incidence rate for women... more Among all cancer types, breast cancer is the one with the second highest incidence rate for women. Mammography is the most used method for breast cancer detection, as it reveals abnormalities such as masses, calcifications, asymmetries and architectural distortions. In this paper, we propose a classification method for breast cancer that has been tested for six different cancer types: CALC, CIRC, SPIC, MISC, ARCH, ASYM. The proposed approach is composed of a SVM classifier trained with LBP features. The MIAS image database was used in the experiments and ROC curves were generated. To the best of our knowledge, our approach is the first to handle those six different cancer types using the same technique. One important result of the proposed approach is that it was tested over six different breast cancer types proving to be generic enough to obtain high classification results in all cases.
19th Iberoamerican Congress on Pattern Recognition - CIARP 2014
An influential framework for object detection in digital images, especially faces, is the one pro... more An influential framework for object detection in digital images, especially faces, is the one proposed by Viola and Jones [1][2]. A
drawback of that framework is the large amount of time needed to train the underlying cascade classifiers. In this paper, we propose a novel hybrid approach for parallelizing the Viola and Jones cascade classifier training framework. The approach is based on message passing among computers and multi-threading in the processor cores, hence its hybrid characteristic. In contrast to related works, which dealt with only parts the original framework, in this paper we addressed the problem of parallelizing the complete framework. Besides, the set of weak classifiers
obtained by our parallel approach is identical to the set of weak classifiers obtained in a serial version. An experimental evaluation carried out within the domain of face detection and focusing on speedup and scalability measures has shown the improvements of the proposed hybrid approach over a serial implementation of the original framework.
Anais do Encontro Nacional de Educação, Ciência e Tecnologia UEPB, 2012
Este artigo apresenta um estudo da influência da iluminação na classificação de faces humanas em ... more Este artigo apresenta um estudo da influência da iluminação na classificação de faces humanas em imagens digitais. Foram usadas as imagens da base The Yale Face Database B
que é composta de 5850 imagens de faces adquiridas sobre condições de iluminação variadas sistematicamente. As seguintes características foram extraídas das imagens: valores dos pixels, histogramas dos valores dos pixels, componentes principais extraídos por meio de PCA e LBP. Essas
características foram utilizadas para treinar e testar os seguintes classificadores: máquinas de vetores de suporte e redes neurais artificiais. Além disso, a extração das características foi realizada global e localmente e métodos básicos de combinação de
classificadores foram aplicados. Os resultados experimentais comprovaram a superioridade da utilização de redes neurais com características LBP extraídas localmente para a tarefa de classificação de imagens de faces com variações de condições de iluminação.
Revista de Tecnologia da Informação e Comunicação, 2013
This paper presents a statistical analysis of methods for face image classification when submitte... more This paper presents a statistical analysis of methods for face image classification when submitted to strong illumination variations. Two classifiers were analyzed: support vector machines and artificial neural networks. Those classifiers were trained using four type of features: pixel intensity values, histogram of pixel intensity values, histogram of local binary patterns, and principal components obtained by principal component analysis. The mentioned features were globally and locally extracted. The F-score of classifications results were computed and evaluated by analysis of variance (ANOVA). The ANOVA demonstrated the statistical significance of results, and allowed the ranking of approaches by F-score values.
This work is concerned with the proposition and empirical evaluation of a new feature extraction ... more This work is concerned with the proposition and empirical evaluation of a new feature extraction approach that combines two existing image descriptors, Integral Histograms and Local Binary Patterns (LBP), to achieve a representation that exhibits relevant properties to object detection tasks (such as face detection): fast constant time processing, rotation, and scale invariance. This novel approach is called the Integral Local Binary Patterns (INTLBP), which is based on an existing method for calculating Integral Histograms from LBP images. This paper empirically demonstrates the properties of INTLBP in a scenario of texture extraction for face/non-face classification. Experiments have shown that the new representation added robustness to scale variations in the test images - the proposed approach achieved a mean classification rate 92% higher than the standard Rotation Invariant LBP approach, when testing over images with scales different from the ones used for training. Moreover, the INTLBP dramatically reduced the required processing time when searching patterns in a face detection task
In this work, we present a system based on a Neural Network classifier for eye detection in human... more In this work, we present a system based on a Neural Network classifier for eye detection in human face images. This classifer works on eye candidate regions extracted from a face image and represented by a reduced number of features, selected by Principal Component Analysis. The regions are determined considering that in an image window containing the eye, the grey level distribution will generally assume a pattern of adjacent light-dark-light horizontal and vertical stripes, corresponding to the eyelid, pupil and eyelid, respectively. For training, validation and testing, a database was built with a total of 4,400 images. Experimental results have shown that the proposed approach correctly detects more eyes than any of two existing systems (Rowley-Baluja-Kanade and Machine Perception Toolbox), for eye location error tolerances from 0 to 5 pixels. Considering an error tolerance of 9 pixels, the correct detection rate achieved was above 90%
2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2018
Electroencephalography (EEG) signals may be severely affected by noise originated from various so... more Electroencephalography (EEG) signals may be severely affected by noise originated from various sources due to their low amplitude nature, specially if they are collected from scalp sensors. Several methods have been proposed for EEG denoising in order to facilitate diagnosis and communication in brain-computer interfaces, but such algorithms often have high complexity. This work presents a denoising approach based on deep learning using a deep convolutional autoencoder, which should reduce the effort of projecting denoising filters. Experiments were performed using two types of noise, originated from eye blink and from jaw clenching. Performance was evaluated with peak signal-to-noise ratio (PSNR) and the results showed that all confidence intervals for the proposed approach were superior to those obtained by the baseline bandpass traditional filtering method. Best average PSNR results for eye blink were obtained for Cz channels with $(20.3\pm 2.6)\mathrm{d}\mathrm{B}$ versus $(14.3\pm 2.4)\mathrm{d}\mathrm{B}$. For jaw clenching, best average PSNR results were obtained for Fz channels with $(21.7\pm 3.1)\mathrm{d}\mathrm{B}$ versus $(13.9\pm 2.6)\mathrm{d}\mathrm{B}$. The proposed approach seems to open a promising scope of research for noise filtering in EEG.
In emotion recognition using EEG, it is not generally agreed upon how much time an EEG signal seq... more In emotion recognition using EEG, it is not generally agreed upon how much time an EEG signal sequence must have in order to maximize precision and recall rates. To the best of our knowledge, there is not a systematic evaluation of effects on classif er performance related to EEG signal durations. The human factors related to attention decreasing and tiredness increasing have imposed diff culties to create EEG datasets containing a rich variation of signal samples. This paper proposes an experimental evaluation of three different EEG datasets (DEAP, MAHNOB, and STEED) each one mainly characterized by short, intermediate and long signal (or stimulus) durations. Statistical evaluation pointed out that for an EEG dataset to be well-suited for emotion recognition it should have two main characteristics: emotion stimulus data should be publicly available and evaluated by worldwide volunteers, and media stimulus should have duration long enough to affect the subjects. Our statistical analysis revealed that, at least for the considered datasets, signals with duration longer than 60 seconds allow better classif cation results. This work did not analyse the impact to humans of longer stimulus media.
Resumo Este artigo apresenta uma tradução de partes do capítulo 16 do Tripura Rahasya seguidas de... more Resumo Este artigo apresenta uma tradução de partes do capítulo 16 do Tripura Rahasya seguidas de discussões e análises tendo como base conceitos apresentados no Yoga Sutra de Patanjali. O capítulo 16 do Tripura Rahasya trata dentre outros fatores do conceito de indizível na tradição indiana e de métodos para alcançar o estado em que é possível acessar o indizível, ou aquilo que não pode ser explicado por palavras. Argumentamos que o Rāja-Yoga de Patañjali é uma metodologia adequada para experimentar esse estado. Abstract This paper presents a translation of parts of Chapter 16 of Tripura Rahasya followed by discussion and analysis based on concepts presented in the Patanjali's Yoga Sutra. Chapter 16 of Tripura Rahasya focuses on, among other factors, the concept of unutterable in the Indian tradition and methods to reach the state where you can access the unsayable, or what can not be explained by words. We argue that the Patanjali's Rāja-Yoga is an appropriate methodology to experience this state.
Recent advances in deep learning methodologies are enabling the construction of more accurate cla... more Recent advances in deep learning methodologies are enabling the construction of more accurate classifiers. However, existing labeled face datasets are limited in size, which prevents CNN models from reaching their full generalization capabilities. A variety of techniques to generate new training samples based on data augmentation have been proposed, but the great majority is limited to very simple transformations. The approach proposed in this paper takes into account intrinsic information about human faces in order to generate an augmented dataset that is used to train a CNN, by creating photo-realistic smooth face variations based on Active Appearance Models optimized for human faces. An experimental evaluation taking CNN models trained with original and augmented versions of the MORPH face dataset allowed an increase of 10 in the F-Score and yielded Receiver Operating Characteristic curves that outperformed state-of-the-art work in the literature.
obtained for all other uses, in any current or future media, including reprinting/republishing th... more obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
This paper deals with two problems: (1) the selection of a set of music features in order to achi... more This paper deals with two problems: (1) the selection of a set of music features in order to achieve high genre classification accuracies; (2) the absence of a representative music dataset of regional brazilian music. In this paper, we propose a set of features to classify genres of music. The features proposed were obtained by a methodical selection of important features used in the literature of Music Information Retrieval (MIR) and Music Emotion Recognition (MER). Besides, we propose a new music dataset called BMD (Brazilian Music Dataset) 1 , containing 120 songs labeled in 7 musical genres: Forró, Rock, Repente, MPB(M ´ usica Popular Brasileira-Brazilian Popular Music), Brega, Sertanejo and Disco. An important characteristic of this new dataset compared with others, is the presence of three popular genres in Brazil Northeast region: Repente, Brega and a characteristic genre similar to MPB, which we also call as MPB. We evaluated our proposed features on both datasets: GTZAN and BMD. The proposed approach achieved average accuracy (after 30 runs of 5-fold-cross-validations) of 79.7% for GTZAN and 86.11% for the BMD. Another important contribution of this work is random repetition of cross-validation executions. Most of the papers performs only a single n-fold cross-validation. We criticize that practice and propose, at least, 30 random executions to compute the average accuracy.
Neste artigo, é apresentada uma análise quantitativa de como a qualidade das interações entre pro... more Neste artigo, é apresentada uma análise quantitativa de como a qualidade das interações entre professores e alunos afetam o desempenho acadêmico. Foram analisadas respostas de 131 alunos a um questionário contendo 14 perguntas. Três tipos de análises foram realizadas: análise descritiva, testes de hipótese e classificação dos alunos em desempenho positivo e desempenho negativo de acordo com as respostas dos questionários submetidas a uma máquina de vetores de suporte (SVM). A partir da análise descritiva verificou-se que um fator muito importante para o desempenho acadêmico dos alunos é o quanto eles procuram os professores em horário de antedimento. Embora exista uma vasta literatura com pesquisas qualitativas sobre como o relacionamento professor-aluno afeta o desempenho dos alunos, os testes de hipótese realizados nesta pesquisa apontaram que os dados coletados não apresentam evidências suficientes para rejeitar a hipótese de que não há diferenças significativas entre os desempenhos de alunos com bom e mau relacionamento com os professores. Por fim, foi possível obter acurácia média de classificação de 60,46% treinando uma SVM utilizando respostas de apenas 10 questões.
After the publication of “Emotional Intelligence” by Daniel Goleman, much interest has been demon... more After the publication of “Emotional Intelligence” by Daniel Goleman, much interest has been demonstrated by education researchers regarding the student qualities responsible for academic success. In this paper, we propose an online software environment to aid the implementation of a methodology focused on three important aspects of student performance: psychological well-being, physical well-being, and studying techniques. Psychological well-being techniques used in our methodology aim to improve attention and emotional skills, physical well-being techniques aim to create good health habits, and studying techniques aim to help students personal time management and teach students how to approach their course materials in order to learn faster and with more quality.
RITA: Revista de Informática Teórica e Aplicada, Nov 2014
Among all cancer types, breast cancer is the one with the second highest incidence rate for women... more Among all cancer types, breast cancer is the one with the second highest incidence rate for women. Mammography is the most used method for breast cancer detection, as it reveals abnormalities such as masses, calcifications, asymmetries and architectural distortions. In this paper, we propose a classification method for breast cancer that has been tested for six different cancer types: CALC, CIRC, SPIC, MISC, ARCH, ASYM. The proposed approach is composed of a SVM classifier trained with LBP features. The MIAS image database was used in the experiments and ROC curves were generated. To the best of our knowledge, our approach is the first to handle those six different cancer types using the same technique. One important result of the proposed approach is that it was tested over six different breast cancer types proving to be generic enough to obtain high classification results in all cases.
19th Iberoamerican Congress on Pattern Recognition - CIARP 2014
An influential framework for object detection in digital images, especially faces, is the one pro... more An influential framework for object detection in digital images, especially faces, is the one proposed by Viola and Jones [1][2]. A
drawback of that framework is the large amount of time needed to train the underlying cascade classifiers. In this paper, we propose a novel hybrid approach for parallelizing the Viola and Jones cascade classifier training framework. The approach is based on message passing among computers and multi-threading in the processor cores, hence its hybrid characteristic. In contrast to related works, which dealt with only parts the original framework, in this paper we addressed the problem of parallelizing the complete framework. Besides, the set of weak classifiers
obtained by our parallel approach is identical to the set of weak classifiers obtained in a serial version. An experimental evaluation carried out within the domain of face detection and focusing on speedup and scalability measures has shown the improvements of the proposed hybrid approach over a serial implementation of the original framework.
Anais do Encontro Nacional de Educação, Ciência e Tecnologia UEPB, 2012
Este artigo apresenta um estudo da influência da iluminação na classificação de faces humanas em ... more Este artigo apresenta um estudo da influência da iluminação na classificação de faces humanas em imagens digitais. Foram usadas as imagens da base The Yale Face Database B
que é composta de 5850 imagens de faces adquiridas sobre condições de iluminação variadas sistematicamente. As seguintes características foram extraídas das imagens: valores dos pixels, histogramas dos valores dos pixels, componentes principais extraídos por meio de PCA e LBP. Essas
características foram utilizadas para treinar e testar os seguintes classificadores: máquinas de vetores de suporte e redes neurais artificiais. Além disso, a extração das características foi realizada global e localmente e métodos básicos de combinação de
classificadores foram aplicados. Os resultados experimentais comprovaram a superioridade da utilização de redes neurais com características LBP extraídas localmente para a tarefa de classificação de imagens de faces com variações de condições de iluminação.
Revista de Tecnologia da Informação e Comunicação, 2013
This paper presents a statistical analysis of methods for face image classification when submitte... more This paper presents a statistical analysis of methods for face image classification when submitted to strong illumination variations. Two classifiers were analyzed: support vector machines and artificial neural networks. Those classifiers were trained using four type of features: pixel intensity values, histogram of pixel intensity values, histogram of local binary patterns, and principal components obtained by principal component analysis. The mentioned features were globally and locally extracted. The F-score of classifications results were computed and evaluated by analysis of variance (ANOVA). The ANOVA demonstrated the statistical significance of results, and allowed the ranking of approaches by F-score values.
This work is concerned with the proposition and empirical evaluation of a new feature extraction ... more This work is concerned with the proposition and empirical evaluation of a new feature extraction approach that combines two existing image descriptors, Integral Histograms and Local Binary Patterns (LBP), to achieve a representation that exhibits relevant properties to object detection tasks (such as face detection): fast constant time processing, rotation, and scale invariance. This novel approach is called the Integral Local Binary Patterns (INTLBP), which is based on an existing method for calculating Integral Histograms from LBP images. This paper empirically demonstrates the properties of INTLBP in a scenario of texture extraction for face/non-face classification. Experiments have shown that the new representation added robustness to scale variations in the test images - the proposed approach achieved a mean classification rate 92% higher than the standard Rotation Invariant LBP approach, when testing over images with scales different from the ones used for training. Moreover, the INTLBP dramatically reduced the required processing time when searching patterns in a face detection task
In this work, we present a system based on a Neural Network classifier for eye detection in human... more In this work, we present a system based on a Neural Network classifier for eye detection in human face images. This classifer works on eye candidate regions extracted from a face image and represented by a reduced number of features, selected by Principal Component Analysis. The regions are determined considering that in an image window containing the eye, the grey level distribution will generally assume a pattern of adjacent light-dark-light horizontal and vertical stripes, corresponding to the eyelid, pupil and eyelid, respectively. For training, validation and testing, a database was built with a total of 4,400 images. Experimental results have shown that the proposed approach correctly detects more eyes than any of two existing systems (Rowley-Baluja-Kanade and Machine Perception Toolbox), for eye location error tolerances from 0 to 5 pixels. Considering an error tolerance of 9 pixels, the correct detection rate achieved was above 90%
Segundo é descrito no Markandeya Purana (DUTT, 1896) e apresentado por Feuerstein (1998), a espos... more Segundo é descrito no Markandeya Purana (DUTT, 1896) e apresentado por Feuerstein (1998), a esposa do brâmane Kausika era bastante dedicada a seu marido. Ao levá-lo doente nas costas, na escuridão da noite, tropeçou em um asceta que rogou-lhe praga para que ao nascer do dia eles morressem. A mulher aflita, pediu que o sol não nascesse. Por ser uma esposa casta, suas preces foram atendidas. Os devas ficaram preocupados pois os rituais realizados ao nascer do sol não poderiam mais ocorrer. Brahma sugeriu que apenas outra casta esposa poderia restaurar a ordem natural e que Anusuya fosse agradada para que rogasse a volta do nascer do sol. Perguntaram-lhe o que ela queria. Um de seus pedidos foi conceber como filhos Brahma, Vishnu e Shiva. Assim, o sol voltou a nascer e Vishnu nasceu como Dattatreya filho de Anusuya.
Aplicação de técnicas de inteligência emocional e meditação para a melhoria do desempenho acadêmi... more Aplicação de técnicas de inteligência emocional e meditação para a melhoria do desempenho acadêmico de estudantes.
Apesar de alguns estudiosos (Daniel Dennet e John Searle, por exemplo) afirmarem que a consciênci... more Apesar de alguns estudiosos (Daniel Dennet e John Searle, por exemplo) afirmarem que a consciência é apenas um fenômeno resultante da fisiologia do cérebro humano, alguns experimentos realizados no contexto da Física apresentam indícios de que a consciência humana é capaz de interferir na realidade. O físico Tom Campbell desenvolveu uma teoria (My Big TOE) que descreve, dentre outros fatores, os motivos pelos quais nossa realidade é virtual e como poderíamos desenvolver nossa consciência para vivermos melhor nessa realidade. Deste modo, podemos afirmar que a ciência ocidental está convergindo para algo que é conhecido há milênios pelos sábios do oriente: podemos entender melhor a realidade em que vivemos pelo desenvolvimento de nossas habilidades mentais, especificamente a consciência. É neste ponto que esta palestra entra em intersecção com o Yoga. Pois o Yoga, e suas variações, é a tecnologia por excelência para o treinamento de nossas habilidades de experienciar a realidade.
Uma pergunta fundamental que grande parte dos estudantes deve se fazer é: qual o ingrediente fund... more Uma pergunta fundamental que grande parte dos estudantes deve se fazer é: qual o ingrediente fundamental para o sucesso acadêmico ou profissional? Um estudo realizado na Nova Zelândia acompanhou detalhadamente um grupo de 1000 crianças desde o nascimento até a idade de 32 anos,
buscando uma resposta para ao questionamento anterior. O texto abaixo é uma tradução livre do trecho do artigo que menciona algumas conclusões dos autores: A capacidade de autoconsciência prevê não apenas notas melhores, como também um bom ajuste emocional, melhores habilidades interpessoais, sensação de segurança e adaptabilidade.
Então, a próxima pergunta que deveria ser feita é: como desenvolver a autoconsciência? Durante milênios, têm sido desenvolvidas técnicas relacionadas originalmente com o Yoga e posteriormente com o Budismo para desenvolver a mente. De modo geral estas técnicas tem sido chamadas no ocidente de meditação e tem sido popularizadas na mídia como atenção plena (mindfulness). Originalmente desenvolvido com objetivos de libertação espiritual, atualmente o Yoga tem sido aplicado a diversas áreas da vida humana, dentre elas: saúde física, saúde mental e aumento do desempenho pessoal. Ao contrário do que a mídia apresenta, o Yoga não se restringe a alongamentos e
contorcionismos. Pelo contrário, o principal conjunto de técnicas empregados pelo Yoga, conhecidos como
Raja Yoga, são estritamente mentais e visam, dentre outras coisas, desenvolver ao máximo a capacidade de concentração e foco do praticante. Vários pesquisadores na área de educação tem desenvolvido trabalhos aplicando técnicas de meditação e verificando o impacto que a prática de tais técnicas causa na vida das pessoas. Alguns estudos também demonstram que algumas técnicas de meditação
podem auxiliar o desenvolvimento da empatia nos professores. Professores com maiores habilidades de
empatia interagem melhor com seus alunos e podem auxiliá-los com mais qualidade.
Uploads
Papers by eanes pereira
drawback of that framework is the large amount of time needed to train the underlying cascade classifiers. In this paper, we propose a novel hybrid approach for parallelizing the Viola and Jones cascade classifier training framework. The approach is based on message passing among computers and multi-threading in the processor cores, hence its hybrid characteristic. In contrast to related works, which dealt with only parts the original framework, in this paper we addressed the problem of parallelizing the complete framework. Besides, the set of weak classifiers
obtained by our parallel approach is identical to the set of weak classifiers obtained in a serial version. An experimental evaluation carried out within the domain of face detection and focusing on speedup and scalability measures has shown the improvements of the proposed hybrid approach over a serial implementation of the original framework.
que é composta de 5850 imagens de faces adquiridas sobre condições de iluminação variadas sistematicamente. As seguintes características foram extraídas das imagens: valores dos pixels, histogramas dos valores dos pixels, componentes principais extraídos por meio de PCA e LBP. Essas
características foram utilizadas para treinar e testar os seguintes classificadores: máquinas de vetores de suporte e redes neurais artificiais. Além disso, a extração das características foi realizada global e localmente e métodos básicos de combinação de
classificadores foram aplicados. Os resultados experimentais comprovaram a superioridade da utilização de redes neurais com características LBP extraídas localmente para a tarefa de classificação de imagens de faces com variações de condições de iluminação.
drawback of that framework is the large amount of time needed to train the underlying cascade classifiers. In this paper, we propose a novel hybrid approach for parallelizing the Viola and Jones cascade classifier training framework. The approach is based on message passing among computers and multi-threading in the processor cores, hence its hybrid characteristic. In contrast to related works, which dealt with only parts the original framework, in this paper we addressed the problem of parallelizing the complete framework. Besides, the set of weak classifiers
obtained by our parallel approach is identical to the set of weak classifiers obtained in a serial version. An experimental evaluation carried out within the domain of face detection and focusing on speedup and scalability measures has shown the improvements of the proposed hybrid approach over a serial implementation of the original framework.
que é composta de 5850 imagens de faces adquiridas sobre condições de iluminação variadas sistematicamente. As seguintes características foram extraídas das imagens: valores dos pixels, histogramas dos valores dos pixels, componentes principais extraídos por meio de PCA e LBP. Essas
características foram utilizadas para treinar e testar os seguintes classificadores: máquinas de vetores de suporte e redes neurais artificiais. Além disso, a extração das características foi realizada global e localmente e métodos básicos de combinação de
classificadores foram aplicados. Os resultados experimentais comprovaram a superioridade da utilização de redes neurais com características LBP extraídas localmente para a tarefa de classificação de imagens de faces com variações de condições de iluminação.
buscando uma resposta para ao questionamento anterior. O texto abaixo é uma tradução livre do trecho do artigo que menciona algumas conclusões dos autores: A capacidade de autoconsciência prevê não apenas notas melhores, como também um bom ajuste emocional, melhores habilidades interpessoais, sensação de segurança e adaptabilidade.
Então, a próxima pergunta que deveria ser feita é: como desenvolver a autoconsciência? Durante milênios, têm sido desenvolvidas técnicas relacionadas originalmente com o Yoga e posteriormente com o Budismo para desenvolver a mente. De modo geral estas técnicas tem sido chamadas no ocidente de meditação e tem sido popularizadas na mídia como atenção plena (mindfulness). Originalmente desenvolvido com objetivos de libertação espiritual, atualmente o Yoga tem sido aplicado a diversas áreas da vida humana, dentre elas: saúde física, saúde mental e aumento do desempenho pessoal. Ao contrário do que a mídia apresenta, o Yoga não se restringe a alongamentos e
contorcionismos. Pelo contrário, o principal conjunto de técnicas empregados pelo Yoga, conhecidos como
Raja Yoga, são estritamente mentais e visam, dentre outras coisas, desenvolver ao máximo a capacidade de concentração e foco do praticante. Vários pesquisadores na área de educação tem desenvolvido trabalhos aplicando técnicas de meditação e verificando o impacto que a prática de tais técnicas causa na vida das pessoas. Alguns estudos também demonstram que algumas técnicas de meditação
podem auxiliar o desenvolvimento da empatia nos professores. Professores com maiores habilidades de
empatia interagem melhor com seus alunos e podem auxiliá-los com mais qualidade.