nam anh - Academia.edu (original) (raw)
Papers by nam anh
2018 1st International Conference on Multimedia Analysis and Pattern Recognition (MAPR), 2018
A method for analytical computing of finding objects of interest in images has been developed wit... more A method for analytical computing of finding objects of interest in images has been developed with multivariate normal distribution to improve the visual capability. Key requirement is the ability of significantly shifting attention to image region that is texture based in general case of real images. The visual attention evaluation is mainly involved for initial task of most visual applications including segmentation, gaze tracking and image re-targeting. To enhance the accuracy of saliency detection, we have to analyze the salient distinction of textured region by combining several techniques. As an initial step, the multivariate filters are designed for estimating local texture feature that is rotation invariant. Significant distinction of patches is then calculated to describe the possible interest regions. The final morphological operations bring fixation of objects of interest. On a test set which consists of ten thousands of images in several themes, the method provides a precision of 92%, recall of 83% and F-measure of 86%.
Advances in Intelligent Systems and Computing, 2018
This paper introduces a geometric aesthetic approach for the analysis of visual attention to extr... more This paper introduces a geometric aesthetic approach for the analysis of visual attention to extract regions of interest from images. Modulation awareness, such as that perceived by visual features, can be represented by attractive proportions of visual objects. Together with supporting techniques such as similarity estimation and lighting condition manipulation, the aesthetic geometry-based analysis can be implemented to form refined attentive shifting observed from image scenes. In this paper, we propose robust kernels which comply with the golden ratio for analysis of aesthetic attractiveness which can raise visual awareness. Properties and relations of points and regions are evaluated by the corresponding kernels for images scenes. We also establish robust likelihood reasoning for the kernels with respect to human aesthetic attraction. The experimental results with a benchmark show the efficiency of the proposed method for identifying region of visual interest in images.
Studies in Computational Intelligence, 2020
Visualizing musical sound for content expression is very efficient application which allows prese... more Visualizing musical sound for content expression is very efficient application which allows presenting music in all its various facets. This article explores the significance of musical emotion anticipating image emotion features. This is a novel representation of the music data to show how the emotion features can add value to a set of existing sound aspects. The musical emotions were then represented by a filter with support of the Gaussian distribution to be used as a color balance filter diversified in term of musical features. With this filter, a color adjustment model can use RGB color system to modify color channels to produce color transform for image regions which are associated with the original musical emotions. As the transform filter is based on music emotions, the image has its color changed adaptively by the emotions. The visualizing solution is then performed in experiments with a music database and image dataset to evaluate the performance. The experiments show the productive visual effect of emotion taking place in the music database with a wide range of instruments and styles and should be of interest for applications of mapping the music and the visual data.
Frontiers in Intelligent Computing: Theory and Applications, 2019
The article explores how class imbalance representation in the probabilistic models is evolved to... more The article explores how class imbalance representation in the probabilistic models is evolved to achieve robust and competent method for balancing distributions of learning data. We tackle the challenging question of imbalance in data level by combining neighborhood relation of samples and the condition of class imbalance in an adaptive model of the k-Nearest Neighbors algorithm. Neighborhood relation with samples in opposite imbalance classes is studied giving expected distribution for minority class samples. It is used to regulate labels of neighbor samples of majority class realizing the classification performance, persistent with the given imbalanced benchmark datasets. Experiments show that our method is capable to accommodate the variation of data types. We thus conclude that machine learning techniques that support aspects of generic classification for data continue to enforce specific profound aspects like class imbalance.
Journal of Advanced Computational Intelligence and Intelligent Informatics, 2020
Based on the results of artificial samples generated in the minority class and through the label ... more Based on the results of artificial samples generated in the minority class and through the label regulation of the neighbor samples of the majority class, the precision of the classification prediction for imbalanced learning has clearly been enhanced. This article presents a unified solution combining learning factors to improve the learning performance. The proposed method solves this imbalance through a feature selection incorporating the generation of artificial samples and label regulation. A probabilistic representation is used for all aspects of learning: class, sample, and feature. A Bayesian inference is applied to the learning model to interpret the imbalance occurring in the training data and to describe solutions for recovering the balance. We show that the generation of artificial samples is sample based approach and label regulation is class based approach. We discovered that feature selection achieves surprisingly good results when combined with a sample- or class-bas...
International Journal of Natural Computing Research, 2017
This article describes how facial expression detection and adjustment in complex psychological as... more This article describes how facial expression detection and adjustment in complex psychological aspects of vision is central to a number of visual and cognitive computing applications. This article presents an algorithm for automatically estimating happiness expression of face images whose demographic aspects like race, gender and eye direction are changeable. The method is also broadening for alteration of level of happiness expression for face images. A schema of the weighted modification is proposed for enhancement of happiness expression. The authors employ a robust face representation which combines the color patch similarity and the self-resemblance of image patches. A large set of face images with appearance of the properties is learned in a statistical model for interpreting the facial expression of happiness. The authors will show the experiments of such a model using face features for learning by SVM and analyze the performance.
2015 Seventh International Conference on Knowledge and Systems Engineering (KSE), 2015
Detection of regions with high visual attention from image has various applications including adv... more Detection of regions with high visual attention from image has various applications including advertising design where ads are often associated with relevant semantic visual information. The salient regions in the image/video have to be identified in a consistent way, even if original objects or background are texture scene. This is achieved by solving combinatorial problem of down-sampling that searches for the optimal texture region map. The complexity of this solution makes it impractical. The problem becomes easy by a new approach for saliency detection. It is based on the spatial attention model that evaluates divergence of a given local region from its surrounding where objects and background can be texture scene. Our proposed solution is based on an adaptive version of the bilateral filter that searches for the divergence of a pixel with its local neighbors. The contributions of this work are new divergence estimation function which reduces potential global search into a simple local filter, and efficient convex-hull algorithm for creating saliency map. Experimental results show that the solution can deal with texture during analysis of visual attention, and saliency detection's performance is improved.
International Journal of Computer Applications, 2015
2014 Fourth International Conference of Emerging Applications of Information Technology, 2014
Proceedings 1996 Asia-Pacific Software Engineering Conference
This paper reports on a collaborative project to pilot the use of formal methods in the developme... more This paper reports on a collaborative project to pilot the use of formal methods in the development of safetyrelated somare. Using the SVRC's Cogito methodology, stafs from CSC Australia undertook: formal specijication; validation of the specijication by mathematical consistency checks; hazard analysis; and validation of the speciJication against the safety requirements. Part of the design was modelled formally and verijied.
International Journal of Computer Applications, 2014
International Journal of Computer Applications, 2014
In fields such as demosaicking, texture removal, dynamic range compression, and photo enhancement... more In fields such as demosaicking, texture removal, dynamic range compression, and photo enhancement many imaging modalities operate with images corrupted by different noise models. Bilateral filter and non-local mean filter are often applied for deduction of noise. This paper presents a new adaptive bilateral filter model to reconstruct edges by choosing neighborhood with non-local mean concept. The method yields considerable gain reduction of noise and keep edges better than original method. Basing in visual inspection, the new method considered as effective even in case of mixed noise.
Proceedings 1996 Asia-Pacific Software Engineering Conference
There are many examples of the use of the technique of domain analysis for modelling software sys... more There are many examples of the use of the technique of domain analysis for modelling software systems in the initial stages of their development, although the case studies chosen are often of small systems or of small parts of large systems. We show that the techniques can be as readily applied to very large domains and we show how a
2006 International Conference onResearch, Innovation and Vision for the Future
Constructivism is a learning theory that states that people learn best when they actively constru... more Constructivism is a learning theory that states that people learn best when they actively construct their own knowledge. Adaptability is the ability to adapt learning ex- periences to different kinds of learners. A significant number of ICT-based constructivist learning systems and ICT-based adaptive learning systems have been proposed in recent years. A critical problem related to the design and use
American Journal of Physiology-Lung Cellular and Molecular Physiology, 2006
Respiratory syncytial virus (RSV) is a prominent cause of airway morbidity in children under 1 yr... more Respiratory syncytial virus (RSV) is a prominent cause of airway morbidity in children under 1 yr of age. It is assumed that host factors influence the severity of the disease presentation and thus the need for hospitalization. As a first step toward the identification of the underlying genes involved, this study was undertaken to establish whether inbred mouse strains differ in susceptibility to pneumonia virus of mice (PVM), the murine counterpart of RSV, which has been shown to accurately mimic the RSV disease of children. With this purpose in mind, double-chamber plethysmography and carbon monoxide uptake data were collected daily for 7 days after inoculation of PVM in six inbred strains of mice. In parallel, histological examinations and lung viral titration were carried out from day 5 to day 7 after inoculation. Pulmonary structure/function values reflected the success of viral replication in the lungs and revealed a pattern of continuous variation, with resistant, intermediat...
2018 1st International Conference on Multimedia Analysis and Pattern Recognition (MAPR), 2018
A method for analytical computing of finding objects of interest in images has been developed wit... more A method for analytical computing of finding objects of interest in images has been developed with multivariate normal distribution to improve the visual capability. Key requirement is the ability of significantly shifting attention to image region that is texture based in general case of real images. The visual attention evaluation is mainly involved for initial task of most visual applications including segmentation, gaze tracking and image re-targeting. To enhance the accuracy of saliency detection, we have to analyze the salient distinction of textured region by combining several techniques. As an initial step, the multivariate filters are designed for estimating local texture feature that is rotation invariant. Significant distinction of patches is then calculated to describe the possible interest regions. The final morphological operations bring fixation of objects of interest. On a test set which consists of ten thousands of images in several themes, the method provides a precision of 92%, recall of 83% and F-measure of 86%.
Advances in Intelligent Systems and Computing, 2018
This paper introduces a geometric aesthetic approach for the analysis of visual attention to extr... more This paper introduces a geometric aesthetic approach for the analysis of visual attention to extract regions of interest from images. Modulation awareness, such as that perceived by visual features, can be represented by attractive proportions of visual objects. Together with supporting techniques such as similarity estimation and lighting condition manipulation, the aesthetic geometry-based analysis can be implemented to form refined attentive shifting observed from image scenes. In this paper, we propose robust kernels which comply with the golden ratio for analysis of aesthetic attractiveness which can raise visual awareness. Properties and relations of points and regions are evaluated by the corresponding kernels for images scenes. We also establish robust likelihood reasoning for the kernels with respect to human aesthetic attraction. The experimental results with a benchmark show the efficiency of the proposed method for identifying region of visual interest in images.
Studies in Computational Intelligence, 2020
Visualizing musical sound for content expression is very efficient application which allows prese... more Visualizing musical sound for content expression is very efficient application which allows presenting music in all its various facets. This article explores the significance of musical emotion anticipating image emotion features. This is a novel representation of the music data to show how the emotion features can add value to a set of existing sound aspects. The musical emotions were then represented by a filter with support of the Gaussian distribution to be used as a color balance filter diversified in term of musical features. With this filter, a color adjustment model can use RGB color system to modify color channels to produce color transform for image regions which are associated with the original musical emotions. As the transform filter is based on music emotions, the image has its color changed adaptively by the emotions. The visualizing solution is then performed in experiments with a music database and image dataset to evaluate the performance. The experiments show the productive visual effect of emotion taking place in the music database with a wide range of instruments and styles and should be of interest for applications of mapping the music and the visual data.
Frontiers in Intelligent Computing: Theory and Applications, 2019
The article explores how class imbalance representation in the probabilistic models is evolved to... more The article explores how class imbalance representation in the probabilistic models is evolved to achieve robust and competent method for balancing distributions of learning data. We tackle the challenging question of imbalance in data level by combining neighborhood relation of samples and the condition of class imbalance in an adaptive model of the k-Nearest Neighbors algorithm. Neighborhood relation with samples in opposite imbalance classes is studied giving expected distribution for minority class samples. It is used to regulate labels of neighbor samples of majority class realizing the classification performance, persistent with the given imbalanced benchmark datasets. Experiments show that our method is capable to accommodate the variation of data types. We thus conclude that machine learning techniques that support aspects of generic classification for data continue to enforce specific profound aspects like class imbalance.
Journal of Advanced Computational Intelligence and Intelligent Informatics, 2020
Based on the results of artificial samples generated in the minority class and through the label ... more Based on the results of artificial samples generated in the minority class and through the label regulation of the neighbor samples of the majority class, the precision of the classification prediction for imbalanced learning has clearly been enhanced. This article presents a unified solution combining learning factors to improve the learning performance. The proposed method solves this imbalance through a feature selection incorporating the generation of artificial samples and label regulation. A probabilistic representation is used for all aspects of learning: class, sample, and feature. A Bayesian inference is applied to the learning model to interpret the imbalance occurring in the training data and to describe solutions for recovering the balance. We show that the generation of artificial samples is sample based approach and label regulation is class based approach. We discovered that feature selection achieves surprisingly good results when combined with a sample- or class-bas...
International Journal of Natural Computing Research, 2017
This article describes how facial expression detection and adjustment in complex psychological as... more This article describes how facial expression detection and adjustment in complex psychological aspects of vision is central to a number of visual and cognitive computing applications. This article presents an algorithm for automatically estimating happiness expression of face images whose demographic aspects like race, gender and eye direction are changeable. The method is also broadening for alteration of level of happiness expression for face images. A schema of the weighted modification is proposed for enhancement of happiness expression. The authors employ a robust face representation which combines the color patch similarity and the self-resemblance of image patches. A large set of face images with appearance of the properties is learned in a statistical model for interpreting the facial expression of happiness. The authors will show the experiments of such a model using face features for learning by SVM and analyze the performance.
2015 Seventh International Conference on Knowledge and Systems Engineering (KSE), 2015
Detection of regions with high visual attention from image has various applications including adv... more Detection of regions with high visual attention from image has various applications including advertising design where ads are often associated with relevant semantic visual information. The salient regions in the image/video have to be identified in a consistent way, even if original objects or background are texture scene. This is achieved by solving combinatorial problem of down-sampling that searches for the optimal texture region map. The complexity of this solution makes it impractical. The problem becomes easy by a new approach for saliency detection. It is based on the spatial attention model that evaluates divergence of a given local region from its surrounding where objects and background can be texture scene. Our proposed solution is based on an adaptive version of the bilateral filter that searches for the divergence of a pixel with its local neighbors. The contributions of this work are new divergence estimation function which reduces potential global search into a simple local filter, and efficient convex-hull algorithm for creating saliency map. Experimental results show that the solution can deal with texture during analysis of visual attention, and saliency detection's performance is improved.
International Journal of Computer Applications, 2015
2014 Fourth International Conference of Emerging Applications of Information Technology, 2014
Proceedings 1996 Asia-Pacific Software Engineering Conference
This paper reports on a collaborative project to pilot the use of formal methods in the developme... more This paper reports on a collaborative project to pilot the use of formal methods in the development of safetyrelated somare. Using the SVRC's Cogito methodology, stafs from CSC Australia undertook: formal specijication; validation of the specijication by mathematical consistency checks; hazard analysis; and validation of the speciJication against the safety requirements. Part of the design was modelled formally and verijied.
International Journal of Computer Applications, 2014
International Journal of Computer Applications, 2014
In fields such as demosaicking, texture removal, dynamic range compression, and photo enhancement... more In fields such as demosaicking, texture removal, dynamic range compression, and photo enhancement many imaging modalities operate with images corrupted by different noise models. Bilateral filter and non-local mean filter are often applied for deduction of noise. This paper presents a new adaptive bilateral filter model to reconstruct edges by choosing neighborhood with non-local mean concept. The method yields considerable gain reduction of noise and keep edges better than original method. Basing in visual inspection, the new method considered as effective even in case of mixed noise.
Proceedings 1996 Asia-Pacific Software Engineering Conference
There are many examples of the use of the technique of domain analysis for modelling software sys... more There are many examples of the use of the technique of domain analysis for modelling software systems in the initial stages of their development, although the case studies chosen are often of small systems or of small parts of large systems. We show that the techniques can be as readily applied to very large domains and we show how a
2006 International Conference onResearch, Innovation and Vision for the Future
Constructivism is a learning theory that states that people learn best when they actively constru... more Constructivism is a learning theory that states that people learn best when they actively construct their own knowledge. Adaptability is the ability to adapt learning ex- periences to different kinds of learners. A significant number of ICT-based constructivist learning systems and ICT-based adaptive learning systems have been proposed in recent years. A critical problem related to the design and use
American Journal of Physiology-Lung Cellular and Molecular Physiology, 2006
Respiratory syncytial virus (RSV) is a prominent cause of airway morbidity in children under 1 yr... more Respiratory syncytial virus (RSV) is a prominent cause of airway morbidity in children under 1 yr of age. It is assumed that host factors influence the severity of the disease presentation and thus the need for hospitalization. As a first step toward the identification of the underlying genes involved, this study was undertaken to establish whether inbred mouse strains differ in susceptibility to pneumonia virus of mice (PVM), the murine counterpart of RSV, which has been shown to accurately mimic the RSV disease of children. With this purpose in mind, double-chamber plethysmography and carbon monoxide uptake data were collected daily for 7 days after inoculation of PVM in six inbred strains of mice. In parallel, histological examinations and lung viral titration were carried out from day 5 to day 7 after inoculation. Pulmonary structure/function values reflected the success of viral replication in the lungs and revealed a pattern of continuous variation, with resistant, intermediat...