Evolving Behaviors for an Interactive Cube-based Artifact
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Text Entry on Smartwatches: A Systematic Review of Literature
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Interaction with Platform Games Using Smartwatches and Continuous Gesture Recognition: A Case Study
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Disparity Map Adjustment: a Post-Processing Technique
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Virtual and Real Interactions for The Creation of Programming Codes for the C3 Cubes Project
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Stereo matching enhancement by statistical analysis and weighted functions
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Multiresolution Analysis on Searching for People in Gigapixel Images
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Metallographic Specimen Imaging Classification: A Machine Learning Approach
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Disparity refinement through grouping areas and support weighted windows
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Applying association rules to study Bipolar Disorder and Premenstrual Dysphoric Disorder comorbidity
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Scientific computing in Statistical Mechanics: time decay of orientation order in 2D hard disk system
Abstract⎯ The present work verifies the applicability of Bayesian Classifiers over Databases from an energy distribution company. The purpose is finding patterns or profiles into determined energy consumption groups and to estimate the number of insolvent clients. The predictive computational system identifies patterns related to each client historic and projects probable behaviors. Insolvency predictions from a Bayesian Network Augmented Naïve-Bayes (BAN) Classifier are compared to results obtained by a Tree Augmented Naïve-Bayes (TAN) Classifier and a Naïve-Bayes (NB) Classifier, taking into account historical insolvency records. Validity is verified by comparing prediction errors. Conclusions suggest an adequate approach which offers arguments for elaborating effective commercial policies for reducing insolvency.
Stereo vision methods: from development to the evaluation of disparity maps
The challenge in building disparity maps is to determine the best method to calculate consistency among points and use it to approximate the differences between the views. Fortunately, the number of methods to construct disparity maps have increased in the last years. On the other hand, a lot of work needs to be done to evaluate these methods by using different arguments. Besides, it is not clear enough how to build stereo vision algorithms with concepts as reuse and scalability. Thus, we propose a software design architecture to be applied in stereo vision systems. Furthermore, we have implemented disparity methods to perform an evaluation which objective was to determine what cost function better fits in each method analyzed. We conclude that MLMH method with SSD cost function is a good choice to be applied in the scenes we have selected, according to the statistical analysis performed.
Classification of Leaf Diseases in Soybean Culture by Haar Wavelet Transform
Soybean (Glycine max), a plant of great economic importance in Brazil and other countries, is the target of a variety of diseases which attack not only leaf part of plant, but also stem, roots and seed itself. Approximately 40 diseases were identified in country, being caused by fungi, bacteria, viruses and nematodes, negatively affecting production, thus limiting crop yield. The objective of this work is to extract texture characteristics of image using the wavelet transform, to analyze regions of soybean leaves affected by pathogens, making the image decomposition, obtaining approximation coeficients. Data that allow classification of the diseases analyzed. The Haar Wavelet Transform was used in feature extraction process and Euclidean distance in classification process. An exploratory database was used, the results were obtained for a proposal effectiveness, whereas adjustments may improve the method presented.
Automated Evaluation System for Human Pupillary Behavior
Analyze the human pupil behavior is a non-invasive method for evaluating neurological activity. This method contributes to the medical field, because changes in pupillary behavior can be correlated with several health conditions, such as Parkinson’s, Alzheimer’s, autism and diabetes. Analyzing human pupil behavior is simple and low-cost, and may be used as complementary diagnosis. Therefore, this work aims to develop an automated system to evaluate the human pupillary behavior. The solution consists of a portable recording device, a pupillometer; integrated with a recording and evaluation software based on computer vision. The system is able to stimulate, record, measure and extract relevant features of the human pupillary behavior. The results shows that the proposed system is fast and accurate, and can be used as an assessment tool for real and extensive clinical practice and research.
Wrist Player: A Smartwatch Gesture Controller for Smart TVs
Emerging technology on mobile and wearable market, smartwatches have embedded movement sensors whose potential is yet to be fully explored. This paper proposes an interaction method with smart TVs via gestures performed by person’s wrist using a smartwatch. Detailed architecture and implementation for a complete prototype, named Wrist Player, is presented. A user study is also conducted, in order to evaluate the prototype performance and the user’s interest on the proposal. Results show that the method works very well, with participants reporting having a good experience with the prototype. We present our insights on the concept, challenges faced in our research and ideas for future studies.
Rest API Model for Integration with Agnostic Biometric Systems
Several entities are committed to the use of biometrics in the area of security, both in access control, as well as for the identification of investigated and crime prevention. The objective of this work is the creation of a prototype of a tool developed within the approach of Services Web, associated to the development of an API that will aim at the standardization of the services of biometrics, with a view to the exchange of data between biometric applications distinct and independent, with a low level of coupling and with interoperability between systems of different platforms and technologies to perform the task of personal identification.
Method for Text Input with Google Cardboard: An Approach Using Smartwatches and Continuous Gesture Recognition
This work proposes a method that allows the entry of text using the Google Cardboard, for this, uses gestures based on geometric forms and smartwatches. Thus, it is proposed the development of a prototype with the purpose of inserting a letter with no more than two user interactions. Gesture recognition is performed using the incremental recognition algorithm. After recognizing the gestures, they are sent to a classifier Naïve Bayes which is responsible for predicting the letter that will be inserted. The Naïve Bayes classifier was trained with a user gesture base that drew all the letters of the alphabet using only the gestures available in the set presented to them. Using the gesture base and the Naïve Bayes classifier a prototype for smartwatches was developed that works in conjunction with the Google Cardboard. The user performs the interactions on smartwatch and views them on Google Cardboard. This prototype automatically suggests the most likely letters to be entered and allows the user to enter text without having to look at smartwatch thus allowing the user to continue performing his other tasks.
Method for Text Entry in Smartwatches Using Continuous Gesture Recognition
This work proposes a method that allows the entry of text in smartwatches using gestures based on geometric forms. For this it is proposed the development of a prototype capable of inserting a letter with no more than two user interactions. Gesture recognition is performed using the incremental recognition algorithm. A set of gestures with lines and curves were created to be recognized by the incremental recognition algorithm, generated from the reduced equation of the line and the reduced equation of the circumference, respectively. After recognizing the gestures, they are sent to a classifier Naïve Bayes which is responsible for predicting the letter that will be inserted. The Naïve Bayes classifier was trained with a user gesture base that drew all the letters of the alphabet using only the gestures available in the set presented to them. Using the gesture base and the classifier Naïve Bayes a prototype was developed for smartwatches that automatically suggests the most likely letters to be inserted. The prototype was used to perform an experiment, during the experiment the users inserted the five most frequent letters and the five less frequent letters of the English language. The results of the experiment show that the prototype is able to recognize a letter with at most two interactions between the user and the smartwatch. The analysis of the usability and experience test shows that the prototype has generalized potential for use, since it allows the entry of text with up to two interactions and with a 100% hit rate for the most frequent letters and 95.14% for less frequent letters.
Strange Artifacts Detection on chest radiography
Chest radiography is one of recommended imaging test by World Health Organization for childhood pneumonia diagnosis. However, during patient examination is very usual finding artifacts in these images, such as identification labels, fingerprints, shirt buttons, and so forth. Moreover, when these images are digitally scanned, other problems raise such as noise, brightness control and so on. Artifacts can reveal private data and expose patient identification. Furthermore, these artifacts can significantly damage automatic analysis by computer diagnosis aided systems. This works presents an efficient method for artifact identification composed by 3 main stages: histogram based pixel filtering, edge detection with Roberts algorithm and standard deviation spacial filtering. This method has been experimented upon 200 images database and presented about 7ms of time processing per image. Visually inspection was used to error measuring and we achieve 0.98 of precision. As a result of this, the method demonstrate a very promising preprocessing tool.
Recognition of utility poles from Google Street View images
This work has a aim to compare classification stability of productive capacity of forests with traditional methods to artificial neural networks (ANN). In this study, an mount of 30.893 pairs of data from 9.034 permanent portions of a continuos forest inventory (CFI) were used. Data was gathered from clones in Eucaliptus urophylla and Eucaliptus grandis plantations in Bahia south. Location index curves were drawn with guide curve method from dominant highs (Hdom) and estimated by traditional methods (Chapman and Richards model) and with ANN. Multilayer Perceptron network (MLP) were trained with backpropagation algorithm with 70% of data, while 30% of data was used for testing. Classification stability of sites were analysed with Chi-Square test. ANN presented greater results in Hdom estimation when compared to Chapman and Richards model. Also, use of categorical variable brought improvements to estitions. Moreover, ANN increased classification stability of sites when compared to traditional regression model.
Text input in Smartwatches Based Gestures Using Geometric Shape
This paper proposes a method for text input based on gestures to be used in smartwatches using geometric shapes. To make the recognition of gestures, we used the incremental recognition algorithm gestures. a template with straight curves were developed using the reduced equation of the circle. Using this template, thirty users have entered all the letters of the alphabet three times each in three groups totaling ninety inserts for each letter. Gestures entered by users have been used to train a Naive Bayes classifier that calculates the probability of insertion for each letter to from the user-entered gestures. During the development work was also carried out a study of the most frequent letters of the Portuguese language. Another partial result of the project is a prototype in which the user enters all the letters of the alphabet using the template gestures. By the time the user enters a gesture prototype automatically suggests the most frequent letters using the Naïve Bayes classifier.
Parallel CUDA Based Implementation of Gaussian Pyramid Image Reduction
A digital image can be represented in several dimensions. In automatic identification process of buildings, people and others objects, resolution can determine efficiency and efficacy of recognition algorithms. Image dimension reduction is useful to minimize computational effort and avoiding adaptation of object detection algorithms. However, the image reduction process also requires high effort and can spend almost same amount of saved time and effort. This paper proposes a parallel implementation of the Gaussian Pyramid multiresolution approach for image reduction of any image and experimental implementation in CUDA. The implementation was performed in 3 different GPU’s and compared with traditional approach in a regular personal computer. Experiments show significative time processing reduction which means high efficiency and potential for adoption of our parallel solution in a image processing chain.
Depth map production: approaches, challenges and applications
Depth maps have been used in computing to solve many practical problems. Security and surveillance systems, tracking trajectories, 3D scanning, autonomous robotic navigation, cinema, forest inventory and mobility are areas which have used solutions based on production of depth maps. This article aims to discuss the approaches used in these productions, pointing limitations and applications with respect to each of the investigated approaches. This paper provides a comprehensive literature review on depth maps, contextualizes the subject, identifies challenges and indicates three approaches to build them, with discussion of each one. Some key questions guided the work organization: ‘Why do we estimate depth?’ ‘How can we produce it?’ and ‘Where/When can we use it?’ This triad directed our research and a result points to a demanding field of study, but, at the same time, a stimulating work.
Classification of Forest Sites from Estimated Dominant Heights with Artificial Neural Network
This work has a aim to compare classification stability of productive capacity of forests with traditional methods to artificial neural networks (ANN). In this study, an mount of 30.893 pairs of data from 9.034 permanent portions of a continuos forest inventory (CFI) were used. Data was gathered from clones in Eucaliptus urophylla and Eucaliptus grandis plantations in Bahia south. Location index curves were drawn with guide curve method from dominant highs (Hdom) and estimated by traditional methods (Chapman and Richards model) and with ANN. Multilayer Perceptron network (MLP) were trained with backpropagation algorithm with 70% of data, while 30% of data was used for testing. Classification stability of sites were analysed with Chi-Square test. ANN presented greater results in Hdom estimation when compared to Chapman and Richards model. Also, use of categorical variable brought improvements to estitions. Moreover, ANN increased classification stability of sites when compared to traditional regression model.
Artificial Neural Networks for Volume Estimation of Caatinga Tree
Caatinga in Bahia state comprises of transiction ranges which species show a greater size features, distinguishing from shrubby feature which is more common in arid regions. Meanwhile, few information are gathered about that phytophysiognomy, primarily when volume estimation is under dicussion. Therefore, main subject of this paper is to design artificial neural networks (ANN) and evaluate their volume estimation performance of caatinga biome species. Thus, in order to gather total and comercial volume 166 tress where mesured in a caatinga tree area and and many Multilayer Perceptron architectures (MLP) where trained with backpropagation algorithm and compared to volumetric models. Neural Networks presented greater statistics results when compared to linear and non-linear equations and error were grouped in lesser ranges. As neural network as equations shown strong bias to volume overestimation. As a way to improve neural network estimations we conclude more variables migth be included.
BrailleEcran: A Braille Approach to Text Entry on Smartphones
Nowadays, due to touchscreen, mobile devices have become much more dynamic than in the past. However, due the same reason, those devices are less accessible for blind people who depend on physical keyboards to type. In order to allow blind people performing text entry more easily, this research have as goal the development of a method, called BrailleÉcran, for smartphones with no physical keyboard. The method combines an Android application and a kind of screen protector with points and Braille symbols. This paper presents a proposal, results, discussion and future works.
Braille Text Entry on Smartphones: A Systematic Review of the Literature
Nowadays, due to touchscreen, mobile devices have become much more dynamic than in the past. However, due the same reason, those devices are less accessible for blind people. It is expected that in 2018, over half of mobile phone users will have a smartphone, therefore researches about mobile accessibility are very important. So, the aim of this systematic review is to cooperate with new research about methods for braille text entry on smartphones. The systematic searches in 5 databases, resulted in 11 papers that answered the research questions that grounded this work.
Tangible Touch Screen Mobile Device for Visually Impaired People: A Braille Text Entering Approach
Mobile devices with touch screen have brought greater task dynamicity and Interaction. However, the lack of a physical keyboard makes mobile devices hard to be used by visually impaired people. The aim of this work is to develop a method and a interface to support the needs of this audience in text entering. Thereby, the Braille method is used and a Android app is being developed for simulating a virtual keyboard. In order to allow user screen handling and interaction, a plastic film with embossed dot marks is being built. This plastic film has the 6 Braille dots and some others for control and interaction. The interface validation is being performed with support of a visually impaired collaborator, with congenital blindness and Braille literate. In order to improve the interface and the text entering software the volunteer performs a lot of experiments and evaluations. The specialist evaluation shows the proposed method has a great potential. Once this stage is completed usability tests with greater number of volunteers will be performed.
Locomotion Interface with Natural Interaction for Assisting Mobility of Visually Impaired
For individuals who are visually impaired (VIP), navigation requires the construction of a cognitive spatial map of one’s surrounding environment. The proposed system was developed using techniques of Virtual Reality and Natural Interaction to assist the construction of cognitive spatial map and consequently the mobility of visually impaired (VIP). The system allows the modeling of two-dimensional or three-dimensional virtual environments. Is it possible to reproduce any places of interest. The interaction with the virtual environment is based on the recognition of poses, which are interpreted as actions of walking, turning left or right, or groping, and sound responses. Tests with VIPs were performed and the results demonstrated the effectiveness of the system, as well as positive feedback from participants who have developed cognitive map of the navigated environment. Thus, it may be indicated in aid in the work of Orientation and Mobility (OM), and thus contribute to greater social inclusion of those people.
Electric Pole segmentation in Google Street View Images
The Google Street View API offers many relevant information to the area of computer vision as images and their respective depth maps. This informations can be used, e.g., for 3D modeling and pattern recognition. This paper presents an approach based on histograms and edge detection in order to find utility pole from Street View images. Initial results indicate that the method can be used in the pre-processing of the images stage to reduce the area of interest which will be examined in the following steps the pattern recognition system.
A New Approach to Detect Use of Alcohol Through Iris Videos Using Computer Vision
In all modern society the increase in alcohol consumption has caused many problems and the potential harmful effects of alcohol on human health are known. There are some ways to identify alcohol in a person, but they are invasive and embarrassing for people. This work proposes a new non-invasive and simple test to detect use of alcohol through of pupillary reflex analysis. The initial results present rates near 85% in the correct identification using algorithms for pattern recognition, demonstrating the efficacy of the test method.
Mathematical model for reduction of the probable spectrum of skin tones in the RGB color space for application in robotic vision with face detection
Skin detection techniques are widely applied to locate and track the human body for the purpose of further recognition. Some advantages of face detection based on skin color in relation to other techniques are the smallest processing time and the angle of rotation invariance and semi-occlusion of faces. Numerous researches have been done with the aim of increasing the capacity perception of the environment by robotic devices using computer vision. In this article are presented theresults of a survey that investigated the reduction in the amplitude of spectrum of colors in the RGB system for detecting skin with possible application in robotic systems developed for tracking and recognition of people. Applied to 16,777,216 possible combinations of colors in the RGB system the 6 rules considered relevant bythe proposed mathematical model allowed a 98.3657% reduction of the spectrum. Experimental results prove the accuracy method considering only 1.6343% (253,159) of colors as candidates for human skin tones and a high rate of correct classification in tests on images of database FDDB.
Tracking and Recognition of Fist Move on Musical Excerpts at the Piano
This works presents a method for teaching assistance in exercises of piano playing with hand tracking and gesture recognition. A tutor defines exercises to be performed and the system detects and tracks the hands of student and checks the correct perform of exercises. To hand detection and tracking are used coloured markers on the back of hands, captured by a regular webcam and for gesture recognition is used Multidimensional Dynamic Time Warping (MDTW), a n-dimensional version of Dynamic Time Warping (DTW). To verify the performance of the method, we have made videos of exercises, which were filtered for some adjusts and noise reduction. Results shows that the method has been capable to detect and recognize gestures successfully. Although the system needs some improvements, first impression about the system has been satisfactory. Thus, we hope, until the end of this work, implement a real-time version of the method for build a tutor system for piano classes.
Mutation-based compact genetic algorithm for spectroscopy variable selection in determining protein concentration in wheat grain
Wheat is the third most produced grain in the world after maize and rice. Determining the protein concentration in wheat grain is one of the major challenges for measuring its industrial quality. Samples of wheat can be collected using a spectrophotometer device. The challenge is to associate the energy absorbed by the device with the protein concentration in wheat. The device measures hundreds of variable intensities that can be related to the physicochemical properties. The selection of a subset of uncorrelated variables has been shown to be fundamental for establishing correct correlations and reducing prediction error. A new formulation of a compact genetic algorithm that uses only a mutation operator is proposed. The results produced by the proposed approach are compared with traditional techniques for spectroscopy variable selection as successive projection algorithms, partial least square and classical formulations of genetic algorithms. For near-infrared spectral analysis of the protein concentration in wheat, the prediction errors decreased from 0.28 to 0.10 on average, a reduction of 63%.
Human iris segmentation on videos obtained via natural lighting from smartphones
The use of human iris images is considered one of the safest ways to provide biometrical recognition. On iris images obtained with natural lighting, the presence of reflection in its surface difficult the process of segmentation and extraction of the patterns present in its texture. This present work has as objective studying and applying techniques for segmentation of the iris region on videos of human eyes obtained via natural lighting. To do this, methods focusing on improving the location of the internal and external iris contour were used in images obtained on these conditions. Some procedures were also made to the removal of present noise over the iris region, such as eyelids and reflections. The results showed that the search of the internal iris region via estimation of its shape provided high matching rates on the segmentation of this region on the video analysis.
Evaluation of Classifiers to a Childhood Pneumonia Computer-Aided Diagnosis System
This work extends PneumoCAD, a Computer-Aided Diagnosis system for detecting pneumonia in infants using radiographic images, with the aim of improving the system’s accuracy and robustness. We implement and compare five contemporary machine learning classifiers, namely: Naive Bayes, K-Nearest Neighbor (KNN), Support Vector Machines (SVM), Multi-Layer Perceptron (MLP) and Decision Tree, combined with three dimensionality reduction algorithms: Sequential Forward Selection (SFS), Principal Component Analysis (PCA) and Kernel Principal Component Analysis (KPCA). Current results demonstrate that Naive Bayes classifier combined with KPCA produces the best overall results.
Static Analysis Techniques and Tools: A Systematic Mapping Study
The main disadvantage of static analysis tools is their high false positive rates. False positives are errors that either do not exist or do not lead to serious software failures. Thus, the benefits of automated static analysis tools are reduced due to the need for manual interventions to assess true and false positive warnings. This paper presents a systematic mapping study to identify current state-of-the-art static analysis techniques and tools as well as the main approaches that have been developed to mitigate false positives.
Characterization of Techniques and Tools of Visualization Applied to Software Comprehension: A Systematic Mapping
This study aimed to make a characterization about how information visualization has been applied on software comprehension with relation to techniques and tools proposed by literature. Systematic Mapping was adopted as the method to guide the investigation process. The findings, although not being definitive, points direction over important questions such as what kind of artifacts and life-cycle phase have been more considered. It was also investigated how these studies have been evaluated, how they evolved in the main digital libraries over the last decade, and what points deserves further attention, through new research.
Multi Objective Genetic Algorithm for Variable Selection in Multivariate Classification Problem: A Study Case in Biodiesel Adulteration Evaluation
This work aims to propose a multiobjective formulation using genetic algorithms for the problem of variable selection in multivariate calibration. A problem involving the classification of biodiesel samples to detect adulteration through a classifier called linear discriminant analysis is presented. The role of the multiobjective genetic algorithm is to reduce the dimensionality of the original set of variables in order to provide a more robust classification model and consequently better generalization capacity. In particular the work makes an implementation of the fast elitism version of the genetic algorithm of ordering by non-dominance (NSGA-II). A comparison is made between a monoobjective and multiobjective implementation with respect to the model and robustness in relation to the presence of noise. The results show that on average the monoobjective genetic algorithm selects 20 variables and has an error rate of 14% and the multiobjective selects 7 variables and has an error rate of 11%. It was possible to demonstrate that the multiobjective formulation provides classification models with less sensitivity to instrumental noise when compared to the monoobjective formulation.
Texture feature extraction for classification type bone in dental radiographs
The scientific and technological imaging diagnosis advances has been contributing in different fields of health science. The new methods are not restricted to X-rays, but include other sources of energy, allowing the visualization of the various parts of the human body. Maintenance of bone after tooth loss to improve or maintain facial esthetics and improved retention, function, and performance of removable restorations are only some of the advantages for the edentulous patient. The combination of tools available for Computer Science, associated with diagnostic imaging has been shown to be a promising way to aid in the diagnostic steps and planning in dental implant field. Special emphasis has been given to computer vision, with the use of texture analysis methods. This preliminary study aims to verify the results obtained from the application of texture analysis in dental radiographs.
Intelligent Classifier of Banknotes: an approach to Principal Component Analysis and Artificial Neural Networks
This paper proposes a method to assist visually impaired people to identify banknotes of Brazilian Real (BRL). The method uses the combination of the technique of Principal Component Analysis (PCA) and artificial neural networks (ANN). The PCA is used in pre-processing the images to reduce the dimensionality of the data. The coefficients obtained with the PCA is used as input to a classifier based on artificial neural network Multilayer Perceptron (MLP). In several experiments, the success rate was 100%. The results obtained with the method were satisfactory. The proposed method has shown promise. It is hoped in the future to recommend its use in helping blind people.
Development of an Application for Mobile Devices to Aid in Diagnosis and Training of Radiologists in Detection of Pneumonia in Childhood -- STRadio
The Variations and errors in the interpretation of radiological images have been studied by researchers and the interobserver agreement is a necessary component in a diagnosis, especially in suspected cases of childhood pneumonia. Only the interobserver agreement does not ensure certain accuracy in diagnosis, therefore, the agreement needs to be systematically reviewed and assessed. Computational tools carry fundamental roles in studies related to interpretation of radiological images in order to measure and assess the quality of interpretation. In this context, this article presents the development of a computational tool for mobile devices for analysis/interpretation studies of variations (intra and inter-observer), as well as assist in the education/training of the readings of chest radiographs for pneumonia in childhood.
An augmented reality approach to building virtual scenes
This article aims to present an alternative approach to building virtual sets, using augmented reality techniques. It starts with an introduction of the main proposal, addressing the themes of RPG (Role Playing Game) and augmented reality, and how these can be related through the use of markers. Followed by a review of studies related to augmented reality in the context of current computing. Finally we present the application of principles developed to meet the proposal and a description of the tests performed. To finalize alternative applications are described for augmented reality approach presented.
Aid System for Locomotion to the Person Blind using Poses Recognition
abstract: This paper proposes a method to aid blind persons in skill development to walk in unfamiliar environments. The method aims to use the human ability in construct of spatial cognitive maps. Thus, we developed a system that uses the Virtual Reality and Kinect. Two-dimensional and three-dimensional virtual environments modeling are allowed. It is possible to reproduce real environments such as schools, universities and other places of interest. The interaction with the virtual environment works with pose recognition. Recognized poses are interpreted as an action such as left and right turn or walk. Experiments were carried out to verify the pose recognition algorithms and to identify the optimal parameters. The results were satisfactory and demonstrated that the proposed method can be quite promising. Thus, it can be indicated as an aid to skill development of locomotion in real environments.
Virtual environment manipulated by recognition of poses using kinect: A study to help blind locomotion in unfamiliar surroundings
This paper proposes a method to aid blind persons in skill development to walk in unfamiliar environments. The method aims to use the human ability in construct of spatial cognitive maps. Thus, we developed a system that uses the Virtual Reality and Kinect R. Two-dimensional and three-dimensional virtual environments modeling are allowed. It is possible to reproduce real environments such as schools, universities and other places of interest. The interaction with the virtual environment works with pose recognition. Recognized poses are interpreted as an action such as left and right turn or walk. Experiments were carried out to verify the pose recognition algorithms and to identify the optimal parameters. The results were satisfactory and demonstrated that the proposed method can be quite promising. Thus, it can be indicated as an aid to skill development of locomotion in real environments.
Recursive diameter prediction for calculating merchantable volume of eucalyptus clones using Multilayer Perceptron
A very common problem in forestry is the realization of the forest inventory. The forest inventory is very important because it allows the trading of mediumand long-term timber to be extracted. On completion , the inventory is necessary to measure different diameters and total height to calculate their volumes. However, due to the high number of trees and their heights, these measurements are an extremely time consuming and expensive. In this work, a new approach to predict recursively diameters of eucalyptus trees by means of Multilayer Perceptron artificial neural networks is presented. By taking only three diameter measures at the base of the tree, diameters are predicted recursively until they reach the value of 4 cm, with no previous knowledge of total tree height. The training was conducted with only 10% of the total trees planted site, and the remaining 90% of total trees were used for testing. The Smalian method was used with the predicted diameters to calculate merchantable tree volumes. To check the performance of the model, all experiments were compared with the least square polynomial approximator and the diameters and volumes estimates with both methods were compared with the actual values measured. The performance of the proposed model was satisfactory when predicted diameters and volumes are compared to actual ones.
Comparative performance analysis of machine learning classifiers in detection of childhood pneumonia using chest radiographs
This work extends PneumoCAD, a Computer-Aided Diagnosis system for detecting pneumonia in infants using radiographic images, with the aim of improving the system’s accuracy and robustness. We implement and compare three contemporary machine learning classifiers, namely: Naive Bayes, K-Nearest Neighbor (KNN), and Support Vector Machines (SVM). Results of our experiments demonstrate that the SVM classifier produces the best overall results.
Experimentation Package for Evaluation of Problems Applied to the Software Project Subject Using PBL
This paper presents an experimentation package that compares the traditional and problem-based learning (PBL) approaches in the Software Project academic subject. The package was applied in a controlled experiment in a Computer Science class of a higher education institution, having message-oriented middleware as a case study. The case study enabled us to validate the experimentation package and to collect initial data to investigate the advantages and disadvantages of PBL against traditional learning. Even though the statistical analysis failed to show differences between the two approaches in view of the data collected, student’s answers to a questionnaire enable us to verify how PBL may be used to increase their motivation and interest in the subject.
RODES 2.0: Educational Softwre for Linear Programming Learning
This paper presents an educational free and multiplatform software, designed for collaborating in teaching Operation Research course, commonly tougth in undergraduate programs in Brazilian universities. RHODES 2.0 aid teaching Simplex Method, providing functionality to problem algebric formulation, graphic presentation in 2D/3D and Simplex tableaus, setting up database options with example problems, among others. Evaluations performed by the authors shows its usefulness as an helpful tool for improving the teaching-learning process inside or outside of classroom.
Recursive Diameter Prediction using Neural Networks to calculate comercial volume of Eucalyptus Clones
This work made predictions of tree diameters of eucalyptus species with artificial neural networks Multiple Layers Perceptron. Recursively estimates are these diameters up to the value of 4 cm using only three steps diameter at the base and without the need knowledge of the total height of the tree. The training was conducted with only 10% of the total planted site. The performance of the proposed model was satisfactory when compared with the diameters predicted actual diameters.
Recursive diameter prediction for calculating merchantable volume of Eucalyptus clones without previous knowledge of total tree height using artificial neural networks
In this work, diameters of Eucalyptus trees are predicted by means of Multilayer Perceptron and Radial Basis Function artificial neural networks. By taking only three diameter measures at the base of the tree, diameters are predicted recursively until they reach the value of minimum merchantable diameter, with no previous knowledge of total tree height. It was considered the diameter top of 4 cm outside bark as minimum merchantable diameter. The training was conducted with only 10% of the trees from the total planted site. The Smalian method utilizes the predicted diameters to calculate merchantable tree volumes. The performance of the proposed model was satisfactory when predicted diameters and volumes are compared to actual ones.
Recursive Diameter Predition with Multilayer Peceptron for comercial volume calculation of Eucalyptus Clones
A method of predicting tree diameters of eucalyptus clones with Artificial neural networks Multiplied Perceptron (MLP) was presented in this work. The objective was to optimize the time and cost of carrying out the forest inventory process. The predictions made with the model used 3 diameters of the base of the tree and the other diameters were estimated one by one in a recursive way until reaching the minimum commercial diameter of 4 cm. The Smalian method was used to calculate the commercial volumes of the trees. MLP training was performed with only 10% of the total trees in the planted site and were tested with the remaining 90% of the total trees. To evaluate the performance of the model, the diameters were compared with the actual measured values. The proposed model was satisfactory in the tests performed, not presenting significant differences when compared to the diameters estimated with the actual values of the trees.
RODES: Educational Softwre for Linear Programming Learning
This paper discusses a virtual learning environment - called RHODES - which provide an intuitive graphical user interface (GUI) in order to aid learning process (and teaching) of Linear Programming, subject commonly tought in Operations Research courses of undergraduate programs.
Recursive diameter prediction and volume calculation of eucalyptus trees using Multilayer Perceptron Networks
A major challenge in forest management is the ability to quickly and accurately predict bole volume of standing trees. This study presents a new model that uses Multilayer Perceptron (MLP) artificial neural networks for predicting tree diameters values. The model requires three diameter measures at the base of the tree, and recursively predicts other diameter measures. The predicted diameters allow for calculating tree volume using the Smalian method. The performance of the proposed model was satisfactory when compared with data obtained from tree scaling and volume equations.
Multilayer Percepton for Recursive Diameter Prediction of Eucalyptus Clones
A major challenge in the forest area is the prediction of the volume of forest stands quickly and accurately without the need to fell trees. The purpose of this work is a model that uses artificial neural network Multilayer Perceptron (MLP) for predicting tree diameters. This model requires three measurements of diameter of the base of the tree and recursively provides other measures of diameters. The performance of the proposed model was satisfactory when compared with diameters obtained from the cubed.
Virtual Reality as support in the treatment of the fear of traveling by train
A phobia is an unreasonable sense of fear, anxiety, persistent, intense and unrealistic, a feeling that arises when the phobic situation is faced with the cause of his phobia. One way to treat phobias is the use of computers to simulate these situations. And, in computing, the area that provides better reproduce real life situations is Virtual Reality. The objective this work is to create a virtual environment that allows the user to experience being in a train journey. Using this system professionals can to assess and treat theirs pacients with the experience generated.
Colaborative Virtual Environment to help on Teaching/Learning Processs in Algorithms and Computer Programming Courses
In this paper we present the development of a Virtual Collaborative Environment to assist the Teaching-Learning process in the Algorithms and Computer Programming disciplines. The main purpose of the work is to build a distributed system that allows several users to construct or understand the operation of the various phases of Algorithm (Structured Portuguese).
Multilayer Percepton for Recursive Diameter Prediction of Eucalyptus Clones
A major challenge in forest management is the ability to quickly and accurately predict bole volume of standing trees. This study presents a model that uses Multilayer Perceptron (MLP) artificial neural networks for predicting tree volume. The model requires three diameter measures at the base of the tree, and recursively predicts other diameter measures. The performance of the proposed model was satisfactory when compared with data obtained from tree scaling.
Virtual Reality on games for Teaching The Four Basic Math Operations
The process of teaching and learning has taken a great time of specialists in order to make it more effective and attractive. Several initiatives are being worked. A special feature is the use of new technologies such as computing. In this context, the teaching of basic mathematical operations (addition, subtraction, multiplication and division) has not undergone major changes in their process of education in recent years, even as it’s a simple and basic content. And therefore it is an investigation if new teaching models have positive influence on learning. Thus the computing and Virtual Reality (VR) may be of great help, as it is defined as the most advanced user interface with the computer until now available. This work aims at the creation of virtual environments, in the form of educational game to help the process of teaching and learning the four basic operations of mathematics. This work is justified on the basis of literature that shows a lack of educational software using Virtual Reality.
Virtual Reality in Demonstration of Field Soccer Tactics
Every soccer coach always looks for the best strategy to surprise the opposing team, that is, a good tactical system of game always aiming the best positioning of his team with actions to surprise at any moment the opponent, be it a play in the middle of the field, on the sides, faults rehearsed or until then in charge of corners. All these actions are defined as the tactical scheme imposed by the trainer. Given the various characteristics and specificities and the great amount of tactical training that exists, there is a difficulty in transferring these formations to the athletes. This difficulty is directly related to the methods used for this attempt. Blackboard or Whiteboard, are used and do not always allow athletes to understand with clarity tactical formations and possible unfolding. In this way the Virtual Reality can be the ideal instrument in this process with visual and interactive characteristics.
Virtual Reality for Viewing Legacy Systems Integration with Workflow
This paper presents a methodology for use of Workflow systems as technology of integration of legacy systems. Using case study the university petitions student’s process, and, XPDL (XML Process Definition Language) as modeling language and the tool SHARK as Workflow Engine, tests was implemented and it had shown the effectiveness of the developed prototype validating the methodology proposal. However, this evaluation revealed a negative aspect: visualization information. This was resolved with the use of Virtual Reality in this visualization.
Teaching Electric Circuits in High School with Virtual Reality
Recent studies have revealed major difficulties in learning the contents of electric circuits. One of the factors contributing to this problem there is that most methods for the teaching of this content aren’t attractive and use the traditional tools that don’t show the facts really surrounding this phenomenon. The use of simulators can improve and one of the areas that can create this kind of interaction is the Virtual Reality. With this necessity in the electric circuits teaching and a great resource in the area of Virtual Reality, it’s proposed to implement the use of Virtual Reality as a complementary tool to the teaching of electric circuits. This tool can be described as a threedimensional interactive virtual environment that simulates the work and analysis of simple electrical circuits and aims to minimize the problem of difficulty of learning in teaching them.
