- Image Processing, Machine Learning, Pattern Recognition, Social Media, Computer Networks, Human Computer Interaction, and 52 moreWeb 2.0, Computer Vision, Natural Language Processing, Digital Signal Processing, Fuzzy Logic Control, Character Recognition, Computer Science, Video Analysis, Person recognition, Engineering, Physics, Information Technology, Artificial Intelligence, Data Mining, Statistical machine learning, Computational Intelligence, Classification (Machine Learning), Speech Recognition, Artificial Neural Networks, Speaker Recognition, Medical Image Processing, Signal and Image Processing, Compressive Sensing, Signal Processing, Optimization, Neural Networks, Biometrics, Firefly Algorithm, Cuckoo Search, Object Recognition (Computer Vision), Probability and Mathematical Statistics, Swarm Intelligence, Deep Learning, Decision Tree, PDE (Mathematics), Image segmentation, Convex Optimization, Numerical Methods, Scientific Computing (Computational Science), Metaheuristics (Operations Research), Model Selection, Natural Computing, Filtering, Algorithm, Philosophy of Science, Image Watermarking, Nature-Inspired Computing, Medical Imaging, Statistics, EEG Signal Processing, Algorithms, and Computer Graphicsedit
Research Interests:
ABSTRACT One of the most important features of biological life in all levels is its astounding diversity. In this work we study the well-known game ``Life'' due to Conway analysing the statistics of cluster population,... more
ABSTRACT One of the most important features of biological life in all levels is its astounding diversity. In this work we study the well-known game ``Life'' due to Conway analysing the statistics of cluster population, N(t), and cluster diversity, D(t). We have performed simulations on ``Life'' for dimensions d = 1 and 2 starting with an uncorrelated distribution of live and dead sites at t = 0. For d = 2 we study the effect of different neighbourhood relations in identifying and counting clusters. An interesting scaling relation connecting the maxima of N(t) and D(t) is found.
Research Interests:
Research Interests:
Research Interests:
ABSTRACT We present a combined features approach for speaker segmentation task. This approach utilizes different acoustic features extracted from audio stream. The Bayesian Information Criterion (BIC) is used for each acoustic feature as... more
ABSTRACT We present a combined features approach for speaker segmentation task. This approach utilizes different acoustic features extracted from audio stream. The Bayesian Information Criterion (BIC) is used for each acoustic feature as a distance measure to verify the merging of two audio segments. An Artificial Neural Network (ANN) combines the time index from each ?BIC with the highest value, and estimates the change point. In the experiments, a data set containing examples with several speakers is used to compare our approach with the Chen and Gopalakrishnan's window-growing-based approach, using different acoustic features sets. The results show an improvement in both the Miss Detection Rate (MDR) and the False Alarm Rate (FAR) compared to the window-growing-based approach.
ABSTRACT Pedestrian detection is a very promising area in computer vision, since it enables interesting and a variety of applications such as car assistance, surveillance systems and robot vision. During the last years, a variety of new... more
ABSTRACT Pedestrian detection is a very promising area in computer vision, since it enables interesting and a variety of applications such as car assistance, surveillance systems and robot vision. During the last years, a variety of new techniques were proposed which greatly improved the detection rates. However, the performance of such systems rapidly deteriorates when pedestrians are under occlusion. This paper analyze how the detection rates of HOG, HOG-LBP, and two new combinations, HOG-LTP and HOG-LMEBP, are affected when occlusion area are progressively added to pedestrian images. Using the INRIA dataset, occlusions were synthetically generated by merging different sizes of non-pedestrian images from different directions. We show that detection of pedestrian under occlusion can be improved by simply combining features.
ABSTRACT Neighborhood coding is a binary image representation method that has been performing successfully for a variety of applications such as features extraction for image recognition, shape descriptor, and image compression. Despite... more
ABSTRACT Neighborhood coding is a binary image representation method that has been performing successfully for a variety of applications such as features extraction for image recognition, shape descriptor, and image compression. Despite the success of this coding method, the representation lacked a formal notation. Here, we proposed a formal mathematical notation to represent any binary image into a neighborhood coding scheme. Using this representation we also introduce the concept of neighborhood operations, a procedure akin to mathematical morphology, however having a lower computational cost.
Research Interests:
Research Interests:
Crumpled surfaces (CS) obtained from random and irreversible compactification of aluminium foils are low-density fractal structures which become increasingly easy to deform as their size increases. The authors study the deformation of... more
Crumpled surfaces (CS) obtained from random and irreversible compactification of aluminium foils are low-density fractal structures which become increasingly easy to deform as their size increases. The authors study the deformation of these objects when ...
Research Interests:
Geometrical and statistical properties of non-equilibrium crumpled surfaces (CS) and crumpled wires (CW) are investigated and compared. The relationship between the geodesic distance x and the Pythagorean distance r in CS and CW and their... more
Geometrical and statistical properties of non-equilibrium crumpled surfaces (CS) and crumpled wires (CW) are investigated and compared. The relationship between the geodesic distance x and the Pythagorean distance r in CS and CW and their dependence on the linear (uncrumpled) size L is studied. Among other results the authors show that the moments of the probability distribution P(x, r) for
Research Interests:
ABSTRACT The authors study how the three-dimensional 'air' or Pythagorean distance r(Q, Q') between two points Q and Q' on a non-equilibrium crumpled fractal surface (CS), with the topology of the plane,... more
ABSTRACT The authors study how the three-dimensional 'air' or Pythagorean distance r(Q, Q') between two points Q and Q' on a non-equilibrium crumpled fractal surface (CS), with the topology of the plane, transforms in the internal or geodesic distance x(Q, Q')-with probability P(x, r)-after the unfolding of the CS on a plane. The probability distribution P(x, r) governing this process is examined for the first time. Among other results they find that (1) the width of P(x, r) 'diverges' for r near the ensemble average radius R of the CS and (2) (x) approximately r13.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
We report the occurrence of critical probabilities associated with the maximum of diversity and the maximum number of fragments (clusters) on a two-dimensional square lattice. Some scaling relations of these two variables are observed in... more
We report the occurrence of critical probabilities associated with the maximum of diversity and the maximum number of fragments (clusters) on a two-dimensional square lattice. Some scaling relations of these two variables are observed in accordance with work on fragmentation processes.
Research Interests:
Diversity of cluster size has been used as a measurement of complexity in several systems that generate a statistical distribution of clusters. Using Monte Carlo simulations, we present a statistical analysis of the cluster size diversity... more
Diversity of cluster size has been used as a measurement of complexity in several systems that generate a statistical distribution of clusters. Using Monte Carlo simulations, we present a statistical analysis of the cluster size diversity and the number of clusters generated on randomly occupied lattices for the Euclidean dimensions 1 to 6. A tuning effect for diversity of cluster size and critical probabilities associated with the maximum diversity and the maximum number of clusters are reported. The probability of maximum diversity is related to the percolation threshold and several scaling relations between the variables measured are reported. The statistics of cluster size diversity has important consequences in the statistical description of the Universe as a complex system.
Research Interests:
Research Interests:
Research Interests:
We present an algorithm to identify and count different lattice animals (LA's) in the site-percolation model. This algorithm allows a definition of clusters based on the distinction of cluster shapes, in contrast with the well-known... more
We present an algorithm to identify and count different lattice animals (LA's) in the site-percolation model. This algorithm allows a definition of clusters based on the distinction of cluster shapes, in contrast with the well-known Hoshen-Kopelman algorithm, in which the clusters are differentiated by their sizes. It consists in coding each unit cell of a cluster according to the nearest neighbors (NN) and ordering the codes in a proper sequence. In this manner, a LA is represented by a specific code sequence. In addition, with some modification the algorithm is capable of differentiating between fixed and free LA's. The enhanced Hoshen-Kopelman algorithm [J. Hoshen, M. W. Berry, and K. S. Minser, Phys. Rev. E 56, 1455 (1997)] is used to compose the set of NN code sequences of each cluster. Using Monte Carlo simulations on planar square lattices up to 2000×2000, we apply this algorithm to the percolation model. We calculate the cluster diversity and cluster entropy of the system, which leads to the determination of probabilities associated with the maximum of these functions. We show that these critical probabilities are associated with the percolation transition and with the complexity of the system.
Research Interests:
Using Monte Carlo simulations, we report the behaviour of the total number of clusters, the cluster size diversity and the lattice animals (LA) diversity on randomly occupied square lattices. The critical probability associated with the... more
Using Monte Carlo simulations, we report the behaviour of the total number of clusters, the cluster size diversity and the lattice animals (LA) diversity on randomly occupied square lattices. The critical probability associated with the maximum of these variables is determined in comparison with the percolation probability pc. Our simulations indicate that pc and the critical probability of the maximum
Research Interests:
ABSTRACT
Research Interests:
Classical feature extraction techniques, like PCA and LDA, do not deal properly with multimodal problems. Such techniques create projections that do not preserve the multimodal structure of the original data distribution. Locality... more
Classical feature extraction techniques, like PCA and LDA, do not deal properly with multimodal problems. Such techniques create projections that do not preserve the multimodal structure of the original data distribution. Locality Preserving Projection (LPP) is a feature extraction technique which looks for a transformation matrix that minimizes the changes into the structure of the data after the transformation. This local structure is captured by the affinity matrix. However, there many ways to calculate this affinity matrix. The main aim of this paper is to evaluate the influence of different affinity matrices over the LPP accuracy. The experiments showed that the correct choice of the affinity matrix can lead to a performance gain. Among the analyzed affinity matrices, Local Scaling and Nearest Neighbor reached the best results.
Research Interests:
Research Interests:
Instance-based learning algorithms typically suffer influences of dissimilarity functions. The problem is frequently related to the Nearest Neighbor rules of these algorithms. This paper will introduce a new dissimilarity measure, called... more
Instance-based learning algorithms typically suffer influences of dissimilarity functions. The problem is frequently related to the Nearest Neighbor rules of these algorithms. This paper will introduce a new dissimilarity measure, called Heterogeneous Centered Difference Measure, which is tested over many known databases. The results are compared with other distance functions.
Research Interests:
Research Interests:
Face verification systems reach good performance on ideal environmental conditions. Conversely, they are very sensitive to non-controlled environments. This work proposes the class-Modular Image Principal Component Analysis (cMIMPCA)... more
Face verification systems reach good performance on ideal environmental conditions. Conversely, they are very sensitive to non-controlled environments. This work proposes the class-Modular Image Principal Component Analysis (cMIMPCA) algorithm for face verification. It extracts local and global information of the user faces aiming to reduce the effects caused by illumination, facial expression and head pose changes. Experimental results performed over three well-known face databases showed that cMIMPCA obtains promising results for the face verification task.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
A social network is composed by communities of individuals or organizations that are connected by a common interest. Online social networking sites like Twitter, Facebook and Orkut are among the most visited sites in the Internet.... more
A social network is composed by communities of individuals or organizations that are connected by a common interest. Online social networking sites like Twitter, Facebook and Orkut are among the most visited sites in the Internet. Presently, there is a great interest in trying to understand the complexities of this type of network from both theoretical and applied point of view. The understanding of these social network graphs is important to improve the current social network systems, and also to develop new applications. Here, we propose a friend recommendation system for social network based on the topology of the network graphs. The topology of network that connects a user to his friends is examined and a local social network called Oro-Aro is used in the experiments. We developed an algorithm that analyses the sub-graph composed by a user and all the others connected people separately by three degree of separation. However, only users separated by two degree of separation are candidates to be suggested as a friend. The algorithm uses the patterns defined by their connections to find those users who have similar behavior as the root user. The recommendation mechanism was developed based on the characterization and analyses of the network formed by the user's friends and friends-of-friends (FOF).
Research Interests:
Research Interests: Image Processing, Informatics, Data Compression, Image Recognition (Computer Vision), Image and Video Compression, and 24 moreImage Compression (Engineering), Image segmentation, Image compression (Computer Science), Digital Image Processing, Image coding, Image compression, Image Reconstruction, Matching, Classification, Image recognition, Template matching, Shape, Shape Analysis, Speech Signal Processing, Compression, Vectors, String Matching, RUn length encoding, Shape Recognition, Shape Descriptor, Handwritten Character Recognition, Huffman Coding, Lossless Compression, and Binary Image
In soccer, scoring goals is a fundamental objective which depends on many conditions and constraints. Considering the RoboCup soccer 2D-simulator, this paper presents a data mining-based decision system to identify the best time and... more
In soccer, scoring goals is a fundamental objective which depends on many conditions and constraints. Considering the RoboCup soccer 2D-simulator, this paper presents a data mining-based decision system to identify the best time and direction to kick the ball towards the goal to maximize the overall chances of scoring during a simulated soccer match. Following the CRISP-DM methodology, data for modeling were extracted from matches of major international tournaments (10691 kicks), knowledge about soccer was embedded via transformation of variables and a Multilayer Perceptron was used to estimate the scoring chance. Experimental performance assessment to compare this approach against previous LDA-based approach was conducted from 100 matches. Several statistical metrics were used to analyze the performance of the system and the results showed an increase of 7.7% in the number of kicks, producing an overall increase of 78% in the number of goals scored.
Research Interests:
Pupil segmentation is usually the first step used for searching iris regions. Iris localization is an extremely important procedure in iris biometrics systems, since the correct segmentation of inner and outer boundaries is critical to... more
Pupil segmentation is usually the first step used for searching iris regions. Iris localization is an extremely important procedure in iris biometrics systems, since the correct segmentation of inner and outer boundaries is critical to achieve high recognition rates. An iris localization method based on a spring force-driven iterative scheme, called Pulling & Pushing have been proposed by He et al. 2006. Here, we propose a pupil segmentation procedure that combines Pulling & Pushing and Active Contour Models, overcoming and improving the results of the previous method. We also developed a new strategy to identify and fill reflection points that appear inside the pupil. We tested our method in MMU1 and Casia V1 and V3 iris databases, obtaining accurate results.
Research Interests:
Research Interests:
Research Interests:
In this paper, the face verification problem is addressed. A neural network with autoassociation memory and receptive fields based architecture is proposed. It is called AAPNet (AutoAssociative Pyramidal Neural Network). The proposed... more
In this paper, the face verification problem is addressed. A neural network with autoassociation memory and receptive fields based architecture is proposed. It is called AAPNet (AutoAssociative Pyramidal Neural Network). The proposed neural network integrates feature extraction and image reconstruction in the same structure. For a given recognition task, at least one instance of the AAPNet must be trained for each known class. Thus, the AAPNet outputs how similar is a given probe image to its class. The AAPNet is applied in a face verification task using thumbnail-sized faces and achieves better results when compared to state-of-the-art models.
Research Interests:
Research Interests:
ABSTRACT We introduce a fingerprint spoof detection technique based on MLP and SVM that combines several features. The proposed technique is evaluated on two scenarios: (i) when an impostor can perform consecutive attempts to be... more
ABSTRACT We introduce a fingerprint spoof detection technique based on MLP and SVM that combines several features. The proposed technique is evaluated on two scenarios: (i) when an impostor can perform consecutive attempts to be considered authentic; and, (ii) when the system deals with fingerprints from elderly people. In order to analyze these scenarios, a database was developed. The results show that the proposed combination of features increases the system performance in at least 33.56% and that the average error increases as more attempts for acceptance are allowed. The SVM classifier presents better performance in almost all the tested configurations. However, MLP is more accurate with biometrics from elderly people.
Research Interests:
Research Interests:
Research Interests:
ABSTRACT We present a neighbourhood vector representation as shape parameter for binary images. This method is based on the pixel neighbourhood relation. Each pixel is transformed into a vector, V = (n, e, s, w), where each element of the... more
ABSTRACT We present a neighbourhood vector representation as shape parameter for binary images. This method is based on the pixel neighbourhood relation. Each pixel is transformed into a vector, V = (n, e, s, w), where each element of the vector represents the total number of neighbour pixels in the respective direction, north, east, south, west. A binary object is represented by a set of neighbourhood vectors (NV), in which the information of the shape structure is retained. The k-means and fuzzy c-means clustering methods are used to reduce the total amount of NV and the probability distribution of the reduced NV is used to characterize a class of image. We applied this method for handwritten character recognition, using neural network as classifiers. The results show that the shape parameter can be used as a general method of feature extraction for problems in image processing and pattern recognition. In addition, we present an application of this representation scheme for the neighbourhood image operator.