Search

The Euler number of a binary image is an important topological feature for many image processing, image analysis, pattern recognition, and computer vision applications. This paper proposes a new run-based Euler number computation algorithm. The conventional run-based algorithm processes rows of the given image one-by-one from top to bottom in a single phase. For each row, it finds the runs in the row and records the start and end locations of each run to compute neighbor runs. In contrast, our algorithm calculates the Euler number of an image in two phases. In the first phase, we process odd rows alternately to find runs and only record its end location. In the second phase, we process each of the remaining even rows to find runs and calculate neighboring runs between the current row and the rows immediately above and below using the recorded run data. Using this method, the number of accesses required to compute the Euler number decreases in almost all cases. Analysis of the time complexity and experimental results demonstrate that our algorithm outperforms conventional Euler number computation algorithms. [ABSTRACT FROM AUTHOR]

Static security analysis is an important study carried out in the control centers of electric utilities. Static security assessment (SSA) is the process of determining whether the current operational state is in a secure or emergency (insecure) state. Conventional method of security evaluation involves performing continuous load flow analysis, which is highly time consuming and infeasible for real-time applications. This led to the application of pattern recognition (PR) approach for static security analysis. This paper presents a more efficient design of a PR system suitable for on-line SSA. The feature selection stage in the PR system uses many algorithms to select the optimal feature set. This paper proposes the use of Support Vector Machine (SVM), a recently introduced machine learning tool, in the classifier design stage of PR system. The developed PR system is implemented in IEEE standard test systems for SSA and classification. The performance of SVM classifier is compared with the conventional K-nearest neighbor, method of least squares and neural network classifiers. Simulation results prove that the SVM-PR classifier outperforms other equivalent classifier algorithms, giving high classification accuracy and less misclassification rate. The feasibility of SVM-PR classifier for on-line security assessment process is also presented. [ABSTRACT FROM AUTHOR]

Non-stationary models often capture better spatial variation of real world spatial phenomena than stationary ones. However, the construction of such models can be tedious as it requires modeling both statistical trend and stationary stochastic component. Non-stationary models are an important issue in the recent development of multiple-point geostatistical models. This new modeling paradigm, with its reliance on the training image as the source for spatial statistics or patterns, has had considerable practical appeal. However, the role and construction of the training image in the non-stationary case remains a problematic issue from both a modeling and practical point of view. In this paper, we provide an easy to use, computationally efficient methodology for creating non-stationary multiple-point geostatistical models, for both discrete and continuous variables, based on a distance-based modeling and simulation of patterns. In that regard, the paper builds on pattern-based modeling previously published by the authors, whereby a geostatistical realization is created by laying down patterns as puzzle pieces on the simulation grid, such that the simulated patterns are consistent (in terms of a similarity definition) with any previously simulated ones. In this paper we add the spatial coordinate to the pattern similarity calculation, thereby only borrowing patterns locally from the training image instead of globally. The latter would entail a stationary assumption. Two ways of adding the geographical coordinate are presented, (1) based on a functional that decreases gradually away from the location where the pattern is simulated and (2) based on an automatic segmentation of the training image into stationary regions. Using ample two-dimensional and three-dimensional case studies we study the behavior in terms of spatial and ensemble uncertainty of the generated realizations. [ABSTRACT FROM AUTHOR]

This paper presents a modified TLBO (teaching-learning-based optimization) approach for the local linear radial basis function neural network (LLRBFNN) model to classify multiple power signal disturbances. Cumulative sum average filter has been designed for localization and feature extraction of multiple power signal disturbances. The extracted features are fed as inputs to the modified TLBO-based LLRBFNN for classification. The performance of the proposed modified TLBO-based LLRBFNN model is compared with the conventional model in terms of convergence speed and classification accuracy. Also, an extreme learning machine (ELM) approach is used to optimize the performance of the proposed LLRBFNN and is compared with the TLBO method in classifying the multiple power signal disturbances. The classification results reveal that although the TLBO approach produces slightly better accuracy in comparison with the ELM approach, the latter is much faster in implementation, thus making it suitable for processing large quantum of power signal disturbance data. [ABSTRACT FROM AUTHOR]

The problems in computer vision of finding the global correspondences across a set of images can be formulated as a multiple graph matching problem consisting of pairwise matching problems. In the multiple graph matching problem, matching consistency is as important as matching accuracy for preventing the contrariety among matched results. Unfortunately, since the majority of conventional pairwise matching methods only approximate the original graph matching problem owing to its computational complexity, a framework that separately matches each graph pair could generate inconsistent results in practical environments. In this paper, we propose a novel multiple graph matching method based on the second-order consistency concept, which simultaneously considers the matching information of all possible graph pairs. We reformulate the multiple graph matching problem to encourage second-order consistency and design an iterative optimization framework. In our experiments, the proposed method outperforms the state-of-the-art methods in terms of both consistency and accuracy. [ABSTRACT FROM AUTHOR]

In this paper, we develop a novel feature extractor called recursive locality preserving projection (RLPP). RLPP inherits the advantages of LPP and at the same time makes some improvements. In RLPP, two local weight graphs are constructed. By combining the ideas of LPP and FLDA, a discriminative maximum criterion is proposed to make the local within-class data pairs close and between-class data pairs apart. To further improve the algorithm performance, a simple but effective method is presented to find the statistically uncorrelated discriminative vectors one by one. In this way, each new obtained discriminative vector not only maximizes the discriminative criterion but also contains minimum redundancy. Our experimental results on five databases demonstrate that RLPP is more powerful than the related methods. [ABSTRACT FROM AUTHOR]

This paper deals with plane detection from a monocular image sequence without camera calibration or a priori knowledge about the egomotion. Within a framework of driver assistance applications, it is assumed that the 3D scene is a set of 3D planes. In this paper, the vision process considers obstacles, roads and buildings as planar structures. These planes are detected by exploiting iso-velocity curves after optical flow estimation. A Hough Transform-like frame called c- velocity was designed. This paper explains how this c- velocity, defined by analogy to the v- disparity in stereovision, can represent planes, regardless of their orientation and how this representation facilitates plane extraction. Under a translational camera motion, planar surfaces are transformed into specific parabolas of the c- velocity space. The error and robustness analysis of the proposed technique confirms that this cumulative approach is very efficient for making the detection more robust and coping with optical flow imprecision. Moreover, the results suggest that the concept could be generalized to the detection of other parameterized surfaces than planes. [ABSTRACT FROM AUTHOR]

In this paper, biometric methods for contactless and unrestricted access control for mobile devices are proposed. The major contribution of this paper are palmprint and knuckles feature extraction methods dedicated for the mobile contactless biometrics. We use texture mask-based features for the palmprint. For the knuckles, we use Probabilistic Hough Transform and Speeded Up Robust Features as well as the 3-step classification methodology. We prove the efficiency of the presented methods by reporting promising results. [ABSTRACT FROM AUTHOR]

We consider the problem of semantic segmentation, i.e. assigning each pixel in an image to a set of pre-defined semantic object categories. State-of-the-art semantic segmentation algorithms typically consist of three components: a local appearance model, a local consistency model and a global consistency model. These three components are generally integrated into a unified probabilistic framework. While it enables at training time a joint estimation of the model parameters and while it ensures at test time a globally consistent labeling of the pixels, it also comes at a high computational cost. We propose a simple approach to semantic segmentation where the three components are decoupled (this journal submission is an extended version of the following conference paper: G. Csurka and F. Perronnin, 'A simple high performance approach to semantic segmentation', BMVC, 2008). For the local appearance model, we make use of the Fisher kernel. While this framework was shown to lead to high accuracy for image classification, to our best knowledge this is its first application to the segmentation problem. The semantic segmentation process is then guided by a low-level segmentation which enforces local consistency. Finally, to enforce image-level consistency we use global image classifiers: if an image as a whole is unlikely to contain an object class, then the corresponding class is not considered in the segmentation pipeline. The decoupling of the components makes our system very efficient both at training and test time. An efficient training enables to estimate the model parameters on large quantities of data. Especially, we explain how our system can leverage weakly labeled data, i.e. images for which we do not have pixel-level labels but either object bounding boxes or even only image-level labels. We believe that an important contribution of this paper is to show that even a simple decoupled system can provide state-of-the-art performance on the PASCAL VOC 2007, PASCAL VOC 2008 and MSRC 21 datasets. [ABSTRACT FROM AUTHOR]

The kernel method suffers from the following problem: the computational efficiency of the feature extraction procedure is inversely proportional to the size of the training sample set. In this paper, from a novel viewpoint, we propose a very simple and mathematically tractable method to produce the computationally efficient kernel-method-based feature extraction procedure. We first address the issue that how to make the feature extraction result of the reformulated kernel method well approximate that of the naïve kernel method. We identify these training samples that statistically contribute much to the feature extraction results and exploit them to reformulate the kernel method to produce the computationally efficient kernel-method-based feature extraction procedure. Indeed, the proposed method has the following basic idea: when one training sample has little effect on the feature extraction result and statistically has the high correlation with regard to all the training samples, the feature extraction term associated with this training sample can be removed from the feature extraction procedure. The proposed method has the following advantages: First, it proposes, for the first time, to improve the kernel method through formal and reasonable evaluation on the feature extraction term. Second, the proposed method improves the kernel method at a low extra cost and thus has a much more computationally efficient training phase than most of the previous improvements to the kernel method. The experimental comparison shows that the proposed method performs well in classification problems. This paper also intuitively shows the geometrical relation between the identified training samples and other training samples. [ABSTRACT FROM AUTHOR]

In this paper, an online measurement and error compensation system for curve grinding based on pattern recognition was presented and verified by experiments. The measurement system organization and its principle of operation were introduced in detail. The work piece and grinding wheel image were sampled at certain positions to avoid spark influence. In order to increase system resolution, images were sampled only at local areas of the work piece and grinding wheel. A discrimination technology based on a circular tolerance zone was proposed which can solve the problem of local image edge comparison. For image de-noising, a local threshold algorithm was applied to determine new wavelet coefficients. Furthermore, a two-step edge detection method was used to realize sub-pixel precision. Finally, a series of experiments were carried out to examine the detection precision of the image measurement system and its influencing factors. From experiments, it can be said that the proposed method in this paper is effective, and its detection precision is much better than traditional methods. [ABSTRACT FROM AUTHOR]

Multivariate statistical process control (MSPC) can be applied for condition monitoring (CM) purposes. MSPC is implemented using a variety of techniques including neural networks (NNs). In situations when the number of process attributes is sufficiently large (e.g. 10 or more) concerns can arise with respect to training of NNs for pattern recognition. A classification method known as novelty detection (ND) can provide an effective alternative to conventional NN solutions that suffer from the above problem. Despite its great potential, ND is still unknown to the broad community of manufacturing engineers. This paper successfully demonstrates the ability of ND, using Gaussian mixture models, to notify operators of an end-milling process of the presence of faulty tool conditions. A significant achievement is that ND is used to identify abnormal time-series patterns as opposed to individual vectors of multiple simultaneous measurements related to abnormal conditions. Such patterns are found in windowed streams of signals related to 10 different process features (with an effective problem dimension of 140). The paper also investigates some of the issues related to implementation of ND for pattern recognition in condition monitoring . [ABSTRACT FROM AUTHOR]

Player location is one of the most informative cues for obtaining tactics arrangement and collecting descriptive game statistics. However, state-of-the-art supervised learning-based methods for player localization require a large amount of labeled training data. Hence, the development of automatic systems for player localization becomes indispensable. Volleyball games reach a huge audience base and contain a variety of tactical strategies, necessitating the implementation of systems for inferring tactics and analyzing formations automatically. Therefore, a novel 2D histogram-based player localization method capable of locating players with occlusions is developed and presented in this paper. The proposed system is able to automatically detect the court lines for camera calibration, extract players by calculating both x and y histograms of extracted player masks, and visualize the team formations on real-world court model. The experiments on broadcast volleyball videos demonstrate efficient and effective results against a traditional object segmentation method (connected component analysis) and a supervised learning approach utilizing histogram of oriented gradient features. [ABSTRACT FROM AUTHOR]

Introduction: Dermoscopy is a non-invasive in vivo imaging technique, used in dermatology in feature identification, among pigmented melanocytic neoplasms, from suspicious skin lesions. Often, in the skin exam is possible to ascertain markers, whose identification and proper characterization is difficult, even when it is used a magnifying lens and a source of light. Dermoscopic images are thus a challenging source of a wide range of digital features, frequently with clinical correlation. Among these markers, one of particular interest to diagnosis in skin evaluation is the reticular pattern. Methods: This paper presents a novel approach (avoiding pre-processing, e.g. segmentation and filtering) for reticular pattern detection in dermoscopic images, using texture spectral analysis. The proposed methodology involves a Curvelet Transform procedure to identify features. Results: Feature extraction is applied to identify a set of discriminant characteristics in the reticular pattern, and it is also employed in the automatic classification task. The results obtained are encouraging, presenting Sensitivity and Specificity of 82.35% and 76.79%, respectively. Conclusions: These results highlight the use of automatic classification, in the context of artificial intelligence, within a computer-aided diagnosis strategy, as a strong tool to help the human decision making task in clinical practice. Moreover, the results were obtained using images from three different sources, without previous lesion segmentation, achieving to a rapid, robust and low complexity methodology. These properties boost the presented approach to be easily used in clinical practice as an aid to the diagnostic process. [ABSTRACT FROM AUTHOR]

The discrimination of similar patterns is important because they are the major sources of the classification error. This paper proposes a novel method to improve the discrimination ability of convolutional neural networks (CNNs) by hybrid learning. The proposed method embeds a collection of discriminators as well as a recognizer in a shared CNN. By visualizing contrastive class saliency, we show that learning with embedded discriminators leads the shared CNN to detect and catch the differences among similar classes. Also proposed is a hybrid learning algorithm that learns recognition and discrimination together. The proposed method learns recognition focusing on the differences among similar classes, and thereby improves the discrimination ability of the CNN. Unlike conventional discrimination methods, the proposed method does not require predefined sets of similar classes or additional step to integrate its result with that of the recognizer. In experiments on two handwritten Hangul databases SERI95a and PE92, the proposed method reduced classification error from 2.56 to 2.33, and from 4.04 to 3.66 % respectively. These improvement lead to relative error reduction rates of 8.97 % on SERI95a, and 9.42 % on PE92. Our best results update the state-of-the-art performance which were 4.04 % on SERI95a and 7.08 % on PE92. [ABSTRACT FROM AUTHOR]

The increasing use of wearable actimetry devices in cohort studies can provide a deep and objective insight in physical activity (PA) patterns. For reliable and reproducible pattern recognition, and to minimize the influence of specific device characteristics, there is a need for a generic method to identify relevant PA events in sensor data sets on the basis of comprehensive features such as PA duration and intensity. The objectives of this paper are to present a method to identify universal event detection thresholds for such parameters, and to attempt to find stable meta-clusters of PA behaviour. PA events of 5, 10, 20 and 30 min with low, medium and high intensity thresholds found in literature and intensity deciles were computed for a random sample ( N = 100) of the NHANES 2005-06 accelerometer data set ( N = 7457). On the basis of all combinations of the above, activity events were detected, and parameters mean duration, mean intensity and event regularity were computed. Results were clustered using x-Means clustering and visualized for 5-, 10-, 20-, and 30-min events. Stable clustering results are obtained with intensity thresholds up to the 8th decile and for event durations up to 10 min. Two stable meta-clusters were detected: 'irregularly active' (intensity at 52nd percentile) and 'regularly active' (intensity at 42nd percentile). Distinct generic thresholds could be identified and are proposed. They may prove useful for further investigations of similar actimetry data sets, minimising the influence of specific device characteristics. The results also confirm that distinct PA event patterns - including event regularity - can be identified using wearable sensor devices, especially when regarding low-intensity, short-term activities which do not correspond to current PA recommendations. Further research is necessary to evaluate actual associations between sensor-based PA parameters and health outcome. The author identified generic intensity and duration thresholds for analysing objective PA data from wearable devices. This may contribute to further analyses of PA patterns along with their relations with health outcome parameters. [ABSTRACT FROM AUTHOR]

The paper proposes a new approach to implement common neural network algorithms in the network environment. In our experimental study we have used three different types of neural networks based on Hebb, daline and backpropagation training rules. Our goal was to discover important market (Forex) patterns which repeatedly appear in the market history. Developed classifiers based upon neural networks should effectively look for the key characteristics of the patterns in dynamic data. We focus on reliability of recognition made by the described algorithms with optimized training patterns based on the reduction of the calculation costs. To interpret the data from the analysis we created a basic trading system and trade all recommendations provided by the neural network. [ABSTRACT FROM AUTHOR]

RFID tags are widely used in situations where their counterfeiting or cloning can bring financial rewards. Cloning is a particular problem because it gets round the sophisticated security measures. This paper describes a neural network-based technique for identifying cloned tickets for a public transport system. It is based on modeling passenger behavior. Cardholders' behavioral characteristics in using public transport are modeled with seven neural network model equations, one for each day of the week, and stored in an RFID card. At the time of use, these model equations or characteristics are employed to predict whether the user is the real owner of the card. Therefore, even if the RFID card is cloned, the cloned card cannot be used because a passenger's behavioral characteristics when using public transport are individual and unique, such as the passenger's signature or style of speech. Therefore, the proposed approach provides high security, especially for low-cost RFID tags. [ABSTRACT FROM AUTHOR]

Finger-vein recognition is an increasingly promising biometric identification technology in terms of its high identification accuracy and prominent security performance. The main challenge faced by finger-vein recognition is the low recognition performance caused by segmentation error and local difference. To tackle this challenge, a finger-vein recognition method with modified binary tree (MBT) model is proposed in this paper. MBT model is used to describe the relationship and spatial structure of vein branches quantitatively. Based on the MBT model, four stages including rough selection, model correction, segment matching, and comprehensive judgment are presented to achieve a robust matching for finger-vein. Experiments demonstrate that the proposed method can boost the performance of finger-vein recognition that is degraded by segmentation error and local difference. While maintaining low complexity, the proposed method achieves 0.12 % equal error rate in the introduced dataset with 8,100 finger-vein images from 150 participants, which outperforms the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The dangerous, overgrown vegetation/trees under high voltage transmission lines right-of-ways (ROWs) have caused severe blackouts/flashovers due to interference with power lines which leads to short circuiting among the conductors. Therefore, these dangerous encroachments are monitored periodically along the electrical distribution networks ROWs through visual inspection, or by airborne system. Each of these methods has its own attributes and limitations and have proved to be costly, time consuming and not much accurate. In these circumstances, it is necessary for the electrical utilities to review their vegetation management practices so as to avoid incidents of unintended encroachments. In tropical countries, overgrown vegetation is a common cause for power line failure. This paper proposes an innovative concept of utilizing a single camera for monitoring dangerous vegetation (trees, shrubs and plants, etc.) under transmission lines ROWs. The main focus is on using an imaging device (camera) integrated on each transmission pole to automate inspection for the vegetation encroachments endangering the transmission lines. These cameras are envisioned to be connected wirelessly to each other, forming a series of wireless camera networks that can be monitored remotely. A single camera mounted on power poles acquires images and sends them wirelessly to the base station. At base station, algorithm (software) trained by image processing and pattern recognition techniques is used to identify (height, depth, and width of encroached vegetation, etc.) excess vegetation encroachments within and outside ROWs. The performance evaluation of a real time developed test-bed scenario proves the feasibilities of integrating the method for transmission line maintenance. [ABSTRACT FROM AUTHOR]

Invasive weed optimization (IWO) is a recently published heuristic optimization technique that resembles other evolutionary optimization methods. This paper proposes a new classification technique based on the IWO algorithm, called the invasive weed classification (IWC), to face the problem of pattern classification for multi-class datasets. The aim of the IWC is to find the set of the positions of the class centers that minimize the multi-objective function, i.e., the optimal positions of the class centers. The classification performance is computed as the percentage of misclassified patterns in the testing dataset achieved by the best plants in terms of fitness performance. The performance of the IWC algorithm, both in terms of classification accuracy and training time, is compared with other commonly used classification algorithms. [ABSTRACT FROM AUTHOR]

Collaboration representation-based classification (CRC) was proposed as an alternative approach to the sparse representation method with similar efficiency. The CRC is essentially a competition scheme for the training samples to compete with each other in representing the test sample, and the training class with the minimum representation residual from the test sample wins the competition in the classification. However, the representation error is usually calculated based on the Euclidean distance between a test sample and the weighted sum of all the same-class samples. This paper exploits alternative methods of calculating the representation error in the CRC methods to reduce the representation residual in a more optimal way, so that the sample classes compete with each other in a closer range to represent the test sample. A large number of face recognition experiments on three face image databases show that the CRC methods with optimized presentation residual achieve better performance than the original CRC, and the maximum improvement in classification accuracy is up to 12 %. [ABSTRACT FROM AUTHOR]

In this paper, a new Chinese character recognition (CCR) approach is proposed based on the fuzzy clustering analysis theory. Chinese characters (CCs) have various similar radicals and stroke components, which make it difficult to recognize features in the CCR process. At the same time, the recognition accuracy and the efficiency are lower when the objects to be recognized are complex. In order to solve these problems, a fuzzy clustering analysis method is introduced to enhance the computing efficiency. At first, the CCs including learning samples and testing samples are transformed into binarization templates in the form of matrixes. Then, the minimum distance algorithm is applied to calculate 'distances' between the testing sample templates and the learning sample templates. At last, the character recognition can be achieved by searching the minimum distance from the results. The experiment results of the CCR process can prove the effectiveness and accuracy of the new method. [ABSTRACT FROM AUTHOR]

A volcano can be considered as a dynamical system, and each time series recorded at a volcano can be interpreted as one of its observables. It is therefore theoretically possible to extract, even from a single time series, information about the underlying governing system. This is done through a procedure called 'embedding' that is based on the intuitive statement that the only time series available carries with it information also about the time evolution of other parameters that we are not able to sample or observe. Carrying out this embedding procedure requires estimates of key parameters such as the optimal delay time and a proper embedding dimension. Other independent but often conceptually similar procedures allow decompositions of the time series into components that may in turn be associated to different source processes. The key to the characterization of volcanic regimes is a process of data reduction, aimed at parsing the amount of data into its most useful components which can then facilitate the interpretation of the system. The approaches presented here can be used to conduct such a data reduction phase, and the reduced data stream can be used not only for characterizing different volcanic regimes but also for determining transitions between them, examining their relationship with external or internal events such as tectonic or volcano-tectonic seismic events, looking for precursors of paroxysmal eruptive phases etc. These results can become additional inputs for physical models in order to understand in detail the physical changes that occurred in the volcanic system and their possible consequences. In this paper, the existing literature on this subject will be reviewed and the prospects of future research will be discussed. [ABSTRACT FROM AUTHOR]

This paper develops a supervised discriminant technique, called graph embedding discriminant analysis (GEDA), for dimensionality reduction of high-dimensional data in small sample size problems. GEDA can be seen as a linear approximation of a multimanifold-based learning framework in which nonlocal property is taken into account besides the marginal property and local property. GEDA seeks to find a set of perfect projections that not only can impact the samples of intraclass and maximize the margin of interclass, but also can maximize the nonlocal scatter at the same time. This characteristic makes GEDA more intuitive and more powerful than linear discriminant analysis (LDA) and marginal fisher analysis (MFA). The proposed method is applied to face recognition and is examined on the Yale, ORL and AR face image databases. The experimental results show that GEDA consistently outperforms LDA and MFA when the training sample size per class is small. [ABSTRACT FROM AUTHOR]

Feature selection has been widely discussed as an important preprocessing step in machine learning and data mining. In this paper, a new feature selection evaluation criterion based on low-loss learning vector quantization (LVQ) classification is proposed. Based on the evaluation criterion, a feature selection algorithm that optimizes the hypothesis margin of LVQ classification through minimizing its loss function is presented. Some experiments that are compared with well-known SVM-RFE and Relief are carried out on 4 UCI data sets using Naive Bayes and RBF Network classifier. Experimental results show that new algorithm achieves similar or even higher performance than Relief on all training data and has better or comparable performance than SVM-RFE. [ABSTRACT FROM AUTHOR]

The two-phase test sample representation (TPTSR) scheme was proposed as a useful method for face recognition; however, the sample selection based on sparse representation in the first phase is not necessary. This is because the first phase only plays a role of course search in TPTSR, but the sparse representation method is suitable for fine classification. This paper proves that alternative nearest-neighbor selection criterions with higher efficiency can be used in the first phase of TPTSR without compromising the classification accuracy. Theoretical analysis and experimental results show that the original distance metric based on sparse representation in the first phase of the TPTSR can be approximated with a more straightforward metric while maintaining a comparable classification performance with the original TPTSR. Therefore, the computational load of the TPTSR can be greatly reduced. [ABSTRACT FROM AUTHOR]

The processing of the images simultaneously degraded by blur and affine transformation has become a key task in many applications and many novel methods are designed specifically for it in which the moment-based methods play an important role. However, the existing moment-based methods all resort to non-orthogonal moments invariants which have problem of information redundancy and are sensitive to noise. In this paper, we construct a new set of combined invariants of orthogonal Legendre moments which hold for blur and affine transformation together. The experimental results show that the proposed invariants have better discriminative power and robustness to noise with the comparison to other invariants. [ABSTRACT FROM AUTHOR]

Motivated by the support vector data description, a classical one-class support vector machine, and the twin support vector machine classifier, this paper formulates a twin support vector hypersphere (TSVH) classifier, a novel binary support vector machine (SVM) classifier that determines a pair of hyperspheres by solving two related SVM-type quadratic programming problems, each of which is smaller than that of a conventional SVM, which means that this TSVH is more efficient than the classical SVM. In addition, the TSVH successfully avoids matrix inversion compared with the twin support vector machine, which indicates learning algorithms of the SVM can be easily extended to this TSVH. Computational results on several synthetic as well as benchmark data sets indicate that the proposed TSVH is not only faster, but also obtains better generalization. [ABSTRACT FROM AUTHOR]

This paper proposes a novel traffic event classification approach using event severities at intersections. The proposed system basically learns normal and common traffic flow by clustering vehicle trajectories. Common vehicle routes are generated by implementing trajectory clustering with Continuous Hidden Markov Model. Vehicle abnormality is detected by observing maximum likelihoods of partial vehicle locations and velocities on underlying common route models. The second part of the work is based on extracting the severities of abnormality by deviation measurement using Coefficient of Variances method. By using abnormal event samples, two severity classes are built in order to recognize event severities by Support Vector Machines and k-Nearest Neighborhood algorithms. Experimental results show that the proposed model has high precision with satisfactory incident detection and event severity classification performance. [ABSTRACT FROM AUTHOR]

Sparse Representation Method has been proved to outperform conventional face recognition (FR) methods and is widely applied in recent years. A novel Kernel-based Sparse Representation Method (KBSRM) is proposed in this paper. In order to cope with the possible complex variation of the face images caused by varying facial expression and pose, the KBSRM first uses a kernel-induced distance to determine N nearest neighbors of the testing sample from all the training samples. Then, in the second step, the KBSRM represents the testing sample as a linear combination of the determinate N nearest neighbors and performs the classification by the representation result. It can be inferred that the N nearest training samples selected are closer to the test sample than the rest, so using the N nearest neighbors to represent the testing sample can make the ultimate classification more accurate. A number of FR experiments show that the KBSRM can achieve a better classification result than the algorithm mentioned in Xu et al. (Neural Comput Appl doi:). [ABSTRACT FROM AUTHOR]

Though sparse representation (Wagner et al. in IEEE Trans Pattern Anal Mach Intell 34(2):372-386, , CVPR 597-604, ) can perform very well in face recognition (FR), it still can be improved. To improve the performance of FR, a novel sparse representation method based on virtual samples is proposed in this paper. The proposed method first extends the training samples to form a new training set by adding random noise to them and then performs FR. As the testing samples can be represented better with the new training set, the ultimate classification obtained using the proposed method is more accurate than the classification based on the original training samples. A number of FR experiments show that the classification accuracy obtained using our method is usually 2-5 % greater than that obtained using the method mentioned in Xu and Zhu (Neural Comput Appl, ). [ABSTRACT FROM AUTHOR]

3D image retrieval approach is a challenging problem in the research of content-based image retrieval. In this paper, a novel retrieval approach combined differential geometry and co-occurrence matrix is presented. Firstly, Gaussian curvature and mean curvature are utilized to represent the inherent characteristic of spatial surface, and then we use co-occurrence matrix to store the shape information of 3D images. Secondly, normalization process is applied to the co-occurrence matrix and the invariants independence of the translation, scaling, and rotation transforms are proved. In comparison with the recent methods, experiments indicate a lower computation complexity and a better retrieval rate to 3D images with slight different shape characteristic. [ABSTRACT FROM AUTHOR]

Neural networks have been employed in many medical applications including breast cancer classification. Innovation in diagnostic features of tumors may play a central role in development of new treatment methods for earliest stage of breast cancer detection. This study proposes a new hybrid for breast cancer detection by extending the application of a variation of particle swarm optimization called K-particle swarm optimization (KPSO). In this paper, the centers and variances of radial basis functional neural network are initialized by KPSO and then updated using back propagation. The weights are updated using recursive least square instead of back propagation. The results are compared with some recently developed techniques. It is found that the proposed technique provides more accurate result and better classification as compared to some other techniques. [ABSTRACT FROM AUTHOR]

Atherosclerosis is the leading underlying pathologic process that results in cardiovascular diseases, which represents the main cause of death and disability in the world. The atherosclerotic process is a complex degenerative condition mainly affecting the medium- and large-size arteries, which begins in childhood and may remain unnoticed during decades. The intima-media thickness (IMT) of the common carotid artery (CCA) has emerged as one of the most powerful tool for the evaluation of preclinical atherosclerosis. IMT is measured by means of B-mode ultrasound images, which is a non-invasive and relatively low-cost technique. This paper proposes an effective image segmentation method for the IMT measurement in an automatic way. With this purpose, segmentation is posed as a pattern recognition problem, and a combination of artificial neural networks has been trained to solve this task. In particular, multi-layer perceptrons trained under the scaled conjugate gradient algorithm have been used. The suggested approach is tested on a set of 60 longitudinal ultrasound images of the CCA by comparing the automatic segmentation with four manual tracings. Moreover, the intra- and inter-observer errors have also been assessed. Despite of the simplicity of our approach, several quantitative statistical evaluations have shown its accuracy and robustness. [ABSTRACT FROM AUTHOR]

This paper aims at detecting sleep apnoea–hypopnoea syndrome (SAHS) from single-channel airflow (AF) recordings. The study involves 148 subjects. Our proposal is based on estimating the apnoea–hypopnoea index (AHI) after global analysis of AF, including the investigation of respiratory rate variability (RRV). We exhaustively characterize both AF and RRV by extracting spectral, nonlinear, and statistical features. Then, the fast correlation-based filter is used to select those relevant and non-redundant. Multiple linear regression, multi-layer perceptron (MLP), and radial basis functions are fed with the features to estimate AHI. A conventional approach, based on scoring apnoeas and hypopnoeas, is also assessed for comparison purposes. An MLP model trained with AF and RRV selected features achieved the highest agreement with the true AHI (intra-class correlation coefficient = 0.849). It also showed the highest diagnostic ability, reaching 92.5 % sensitivity, 89.5 % specificity and 91.5 % accuracy. This suggests that AF and RRV can complement each other to estimate AHI and help in SAHS diagnosis. [ABSTRACT FROM AUTHOR]

Fingerprint systems have received a great deal of research and attracted many researchers' effort since they provide a powerful tool for access control and security and for practical applications. A literature review of the techniques used to extract the features of fingerprint as well as recognition techniques is given in this paper. Some of the reviewed research articles have used traditional methods such as recognition techniques, whereas the other articles have used neural networks methods. In addition, fingerprint techniques of enhancement are introduced. [ABSTRACT FROM AUTHOR]

In this paper, we propose a Gabor-based face recognition method. This method fuses multi-resolution Gabor features of face images at the matching score level. The first implementation scheme of this method directly takes the sum of the matching scores of multi-resolution Gabor features of face images as the final matching score. The second implementation scheme first codes the phase of the Gabor feature and then uses a weighted matching score level fusion algorithm to fuse the magnitude and phase of the Gabor feature. A number of experimental results show that the proposed method has a good performance and outperforms conventional Gabor-based face recognition methods that equally treat all the Gabor features and directly fuse them at the feature level. The experimental result also illustrates that in face recognition, the low-resolution representation of the phase of the Gabor feature such as the code of the phase is more discriminative than the phase itself. The codes of our method will be available at http://www.yongxu.org/lunwen.html . [ABSTRACT FROM AUTHOR]

The automation of cost estimation for manufacturing processes is a challenging task in computer-aided manufacturing. In this paper, we introduce a two-step analogy and mathematical approach to estimate the cost of injection molding. In the analogy step, data of molds are partitioned into homogeneous groups based on mold type and mold design. In the prediction step, regression models based upon geometry, topology, and other inherent shape properties are constructed within each group. The variables in the regression models within each group are extracted automatically from one orthographic two-dimensional (2D) image of the injection-molded part. Mean and variance estimates are calculated on a subset of relevant molds so that the risk of an inaccurate bid can be assessed on a subset of relevant molds. [ABSTRACT FROM AUTHOR]

Bad quality spots should be filtered out at early steps in microarray analysis to avoid noisy data. In this paper we implement quality control of individual spots from real microarray images. First of all, we consider the binary classification problem of detecting bad quality spots. We propose the use of ensemble algorithms to perform detection and obtain improved accuracies over previous studies in the literature. Next, we analyze the untackled problem of identifying specific spot defects. One spot may have several faults simultaneously (or none of them) yielding a multi-label classification problem. We propose several extra features in addition to those used for binary classification, and we use three different methods to perform the classification task: five independent binary classifiers, the recent Convex Multi-task Feature Learning (CMFL) algorithm and Convex Multi-task Independent Learning (CMIL). We analyze the Hamming loss and areas under the receiver operating characteristic (ROC) curves to quantify the accuracies of the methods. We find that the three strategies achieve similar results leading to a successful identification of particular defects. Also, using a Random forest-based analysis we show that the newly introduced features are highly relevant for this task. [ABSTRACT FROM AUTHOR]

An important step in root cause analysis is the identification of the time when process first changed. The time when a disturbance first manifested itself into the process is referred to as change point. Identification of the change point could help process engineer to perform root cause analysis effectively. In this paper, an estimator for the change point of a normal process mean using artificial neural network (ANN) is proposed. Five patterns of change namely single step, linear trend, systematic, cyclic, and mixture are studied. Whenever possible, results are compared numerically to the results obtained by other methods proposed by different researchers. First the type of change to be recognized by an ANN-based pattern recognizer is identified and then the change point in the process mean is estimated. Results indicate satisfactory performance for the proposed method that could be used as an effective method for root cause analysis by process engineer. [ABSTRACT FROM AUTHOR]

In the current pandemic, smart technologies such as cognitive computing, artificial intelligence, pattern recognition, chatbot, wearables, and blockchain can sufficiently support the collection, analysis, and processing of medical data for decision making. Particularly, to aid medical professionals in the disease diagnosis process, cognitive computing is helpful by processing massive quantities of data rapidly and generating customized smart recommendations. On the other hand, the present world is facing a pandemic of COVID-19 and an earlier detection process is essential to reduce the mortality rate. Deep learning (DL) models are useful in assisting radiologists to investigate the large quantity of chest X-ray images. However, they require a large amount of training data and it needs to be centralized for processing. Therefore, federated learning (FL) concept can be used to generate a shared model with no use of local data for DL-based COVID-19 detection. In this view, this paper presents a federated deep learning-based COVID-19 (FDL-COVID) detection model on an IoT-enabled edge computing environment. Primarily, the IoT devices capture the patient data, and then the DL model is designed using the SqueezeNet model. The IoT devices upload the encrypted variables into the cloud server which then performs FL on major variables using the SqueezeNet model to produce a global cloud model. Moreover, the glowworm swarm optimization algorithm is utilized to optimally tune the hyperparameters involved in the SqueezeNet architecture. A wide range of experiments were conducted on benchmark CXR dataset, and the outcomes are assessed with respect to different measures . The experimental outcomes pointed out the enhanced performance of the FDL-COVID technique over the other methods. [ABSTRACT FROM AUTHOR]

Breast cancer is the major cause of cancer deaths in women today and it is the most common type of cancer in women. This paper presents some experiments for classifying breast cancer tumor and proposes the use of firefly algorithm (FA) to improve the performance of Local linear wavelet neural network. This work in fact uses FA to optimize the parameters of local linear wavelet neural network. The experiments were conducted on extracted breast cancer data from University of Winconsin Hospital, Madison. The result has been compared with a wide range of classifiers to evaluate its performance. The evaluations show that the proposed approach is very robust, effective and gives better correct classification as compared to other classifiers. [ABSTRACT FROM AUTHOR]

In this paper, we propose a very simple and fast face recognition method and present its potential rationale. This method first selects only the nearest training sample, of the test sample, from every class and then expresses the test sample as a linear combination of all the selected training samples. Using the expression result, the proposed method can classify the testing sample with a high accuracy. The proposed method can classify more accurately than the nearest neighbor classification method (NNCM). The face recognition experiments show that the classification accuracy obtained using our method is usually 2-10% greater than that obtained using NNCM. Moreover, though the proposed method exploits only one training sample per class to perform classification, it might obtain a better performance than the nearest feature space method proposed in Chien and Wu (IEEE Trans Pattern Anal Machine Intell 24:1644-1649, ), which depends on all the training samples to classify the test sample. Our analysis shows that the proposed method achieves this by modifying the neighbor relationships between the test sample and training samples, determined by the Euclidean metric. [ABSTRACT FROM AUTHOR]

The paper proposes an improved quantum associative algorithm with distributed query based on model proposed by Ezhov et al. We introduce two modifications of the query that optimized data retrieval of correct multi-patterns simultaneously for any rate of the number of the recognition pattern on the total patterns. Simulation results are given. [ABSTRACT FROM AUTHOR]

Computational intelligence shows its ability for solving many real-world problems efficiently. Synergism of fuzzy logic, evolutionary computation, and neural network can lead to development of a computational efficient and performance-rich system. In this paper, we propose a new approach for solving the human recognition problem that is the fusion of evolutionary fuzzy clustering and functional modular neural networks (FMNN). Evolutionary searching technique is applied for finding the optimal number of clusters that are generated through fuzzy clustering. The functional modular neural network has been used for recognition process that is evaluated with the help of integration based on combining the outcomes of FMNN. Performance of the proposed technique has been empirically evaluated and analyzed with the help of different parameters. [ABSTRACT FROM AUTHOR]

The recently proposed twin support vector machine (TWSVM) obtains much faster training speed and comparable performance than classical support vector machine. However, it only considers the empirical risk minimization principle, which leads to poor generalization for real-world applications. In this paper, we formulate a robust minimum class variance twin support vector machine (RMCV-TWSVM). RMCV-TWSVM effectively overcomes the shortcoming in TWSVM by introducing a pair of uncertain class variance matrices in its objective functions. As a special case, we present a special type of the uncertain class variance matrices by combining the empirical positive and negative class variance matrices. Computational results on several synthetic as well as benchmark datasets indicate the significant advantages of proposed classifier in both computational time and test accuracy. [ABSTRACT FROM AUTHOR]

A doctor could say that a patient is sick while he/she is healthy or could say that the patient is healthy while he/she is sick, by mistake. So it is important to generate a system that can give a good diagnosis, in this case for abnormal eye movements. An abnormal eye movement is when the patient wants to move the eye to up or down and the eye does not move or the eye moves to other place. In this paper, a method for the pattern recognition is used to provide a better diagnosis for patients related with the abnormal eye movements. The real data of signals of two eye movements (up and down) of patients are obtained using a mindset ms-100 system. A new method that uses one intelligent algorithm for online pattern recognition is proposed. The difference between the proposed method and the previous works is that, in other works, both behaviors (up and down) are trained with one intelligent algorithm, while in this work, up behavior is trained with one intelligent algorithm and down behavior is trained with other intelligent algorithm; it is because the multi-output system can always be decomposed into a collection of single-output systems with the advantage to use different parameters for each one if necessary. The intelligent algorithm used by the proposed method could be any of the following: the adaline network denoted as AN, the multilayer neural network denoted as NN, or the Sugeno fuzzy inference system denoted as SF. So the comparison results of the proposed method using each of the intelligent algorithms for online pattern recognition of two eye movements are presented. [ABSTRACT FROM AUTHOR]

Building patterns are important features that should be preserved in the map generalization process. However, the patterns are not explicitly accessible to automated systems. This paper proposes a framework and several algorithms that automatically recognize building patterns from topographic data, with a focus on collinear and curvilinear alignments. For both patterns two algorithms are developed, which are able to recognize alignment-of-center and alignment-of-side patterns. The presented approach integrates aspects of computational geometry, graph-theoretic concepts and theories of visual perception. Although the individual algorithms for collinear and curvilinear patterns show great potential for each type of the patterns, the recognized patterns are neither complete nor of enough good quality. We therefore advocate the use of a multi-algorithm paradigm, where a mechanism is proposed to combine results from different algorithms to improve the recognition quality. The potential of our method is demonstrated by an application of the framework to several real topographic datasets. The quality of the recognition results are validated in an expert survey. [ABSTRACT FROM AUTHOR]

It is known that long periods of adverse weather have a negative effect on the human cardiovascular system. A number of studies have set a lower limit of around 5 days for the duration of these periods. However, the specific features of the negative dynamics of the main weather characteristics-air temperature and atmospheric pressure-remained open. To address this problem, the present paper proposes a conjunctive method of the theory of pattern recognition. It is shown that this method approaches a globally optimal (in the sense of recognition errors) Neumann critical region and can be used to solve various problems in heliobiology. To illustrate the efficiency of this method, we show that some quickly relaxing short sequences of temperature and pressure time series (the so-called temperature waves and waves of atmospheric pressure changes) increase the risk of cardiovascular diseases and can lead to serious organic lesions (particularly myocardial infarction). It is established that the temperature waves and waves of atmospheric pressure changes increase the average morbidity rate of myocardial infarction by 90% and 110%, respectively. Atmospheric pressure turned out to be a more biotropic factor than air temperature. [ABSTRACT FROM AUTHOR]