The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
SVM classifiers with Half Against Half (HAH) architecture are reported to be the fastest classifier amongst other SVM classification architectures reported in literature. An attempt is made to enhance the speed of HAH SVM classifier and is named as Fast HAH (F-HAH) classifier. The performance of proposed F-HAH classifier is evaluated using speaker dependent and multi-speaker dependent isolated digits...
In recent years, the design of classification algorithms, with the aid of information combination methods, has received a considerable attention. In machine vision, in order to overcome the high inter-class variations between the classes of image, various feature descriptors have been designed to be robust to these inter-class variations. However, no single feature can be robust to these variations...
A texture descriptor based on a set of indices of degrees of local approximating polynomials is proposed in this paper. An image is split into non-overlapping patches, reshaped into one-dimensional source vectors and convolved with the polynomial approximation kernels of various degrees p. As a result, a set of approximations is obtained. For each element of the source vector, these approximations...
In order to diagnose fault source effectively, this paper proposed a novel fault pattern recognition method called dynamic independent component based sparse kernel classifier (DICSKC). In the proposed method, fault pattern recognition is viewed as a classification problem and kernel trick is applied to construct nonlinear classifier for each fault scene. To improve classification performance, dynamic...
The objective of this study is to investigate different pattern classification paradigms in the automatically understanding and characterizing driver behaviors. With features extracted from a driving posture dataset consisting of grasping the steering wheel, operating the shift lever, eating a cake and talking on a cellular phone, created at Southeast University, holdout and cross-validation experiments...
Using the Mallat fast algorithm with sym5 wavelet, the pulse waves of 20 heroin druggers and 20 healthy normal subjects are decomposed into two levels. The squared distances from the third and tenth scale coefficients in the second-level decomposition of every pulse wave to the global mean value are used to form a feature vector. The extracted feature vectors have good separable characteristics in...
Camera motion classification is an important research topic in video content analysis and retrieval. In this paper, we propose a nonparametric classification method of camera motion, which employs support vector machine to learn motion vector file and then train classifiers to categorize camera motion. Libsvm is used as support vector machine tool to validate the method and experiments demonstrate...
This paper presents the support vector machine (SVM) for classification of the quality grade of knitted yarns. The SVM, Kernel Fisher Discriminant Analysis (KFDA), back promulgation neural network (BPNN), and radial basis function neural network (RBFNN) are comparatively investigated in 94 classified knitted yarns from different mills in four-dimensional space, four methods are employed on IRIS and...
Support Vector Machine (SVM) is one of the state-of-the-art tools for linear and nonlinear pattern classification. One of the design issues in SVM classifier is reducing the number of support vectors without compromising the classification accuracy. In this paper, a novel technique which requires only a subset of the support vectors is proposed. The subset is obtained by including only those support...
This paper evaluates supervised and unsupervised adaptive schemes applied to online support vector machine (SVM) that classifies BCI data. Online SVM processes fresh samples as they come and update existing support vectors without referring to pervious samples. It is shown that the performance of online SVM is similar to that of the standard SVM, and both supervised and unsupervised schemes improve...
As an important kernel function in support vector machines (SVM), Gaussian kernel (GK) is widely used in pattern recognition and artificial intelligence. However, the fact that Gaussian kernel could not distinguish the importance of data features is not conforming to the practical situation. According to the deficiency of Gaussian kernel, weighted Gaussian kernel with multiple widths (WGKMW) is proposed...
A new functional model for burst firing in the dorsal thalamus is proposed where thalamocortical pattern recognition systems, based on kernel machine principles, are connected by burst signaling. The systems include input trapping in the dorsal thalamus, cortical learning state memory and processing in the thalamic reticular nucleus. Misclassified events are captured as training examples in the waking...
A novel algorithm for multi-class support vector machines (SVMs) is proposed in this paper. The tree constructed in our algorithm consists of a series of two-class SVMs. Considering both separability and balance, in each iteration multi-class patterns are divided into two sets according to the distances between pairwise classes and the number of patterns in each class. This algorithm can well treat...
In structural pattern recognition, a major drawback of graph based representation is the lack of algorithmic tools. To overcome this lack, we embed graphs in vector spaces by means of prototype selection and graph edit distance, thus making them available to all algorithms of statistical pattern recognition that operate on feature vectors. In previous work a similar procedure was applied. However,...
An automated method that detects early cancerous specimens based on image analysis is described. After acquisition and noise reduction, the microscope images are segmented into individual cell nucleus, from which the feature vectors of nucleus are calculated. The dimensionality of the feature vectors is then reduced using a method combing F-Score and random forest algorithms. The types of the cell...
Support vector machine (SVM) has become a popular tool in the area of pattern recognition, combining support vector machines with other theories has been proposed as a new direction to improve classification performance. This paper applies fuzzy theory to support vector machines for classification. In the first phase, a fuzzy support vector machine is proposed for the classification of real-world...
Support vector machine, a universal method for learning from data, gains its development based on statistical learning theory. It shows many advantages in solving nonlinearly small sample and high dimensional problems of pattern recognition. Only a part of samples or support vectors (SVs) plays an important role in the final decision function. But SVs could not be obtained in advance until a quadratic...
This paper deals with automatic modulation classification of communication signals. A new scheme of automatic modulation classification using wavelet analysis and wavelet support vector machine (WSVM) is proposed. Further, a new way of training for wavelet features is carried out to adapt to signals which are non-stable and varied in a wide range of signal-to-noise rates (SNR). Through such training,...
We present a novel way of interpreting and modifying the outputs of the support vector machine classifiers. The geometrical interpretation of the SVM outputs, as distance of the patterns from the hyperylane, gives us a way to calculate posterior probability, i.e. construct a rejection threshold for vectors belonging to one of the classes. We illustrate the results by analyzing three classification...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.