Sharif Digital Repository

In this paper, a new feature set for audio classification is presented and evaluated based on sinusoidal modeling of audio signals. Variance of the birth-death frequencies in sinusoidal model of signal, as a measure of harmony, is used and compared to typical features as the input into an audio classifier. The performance of this sinusoidal model feature is evaluated through classification of audio to speech and music using both the GMM and the SVM classifiers. Classification results show that the proposed feature is quite successful in speech/music classification. Experimental comparisons with popular features for audio classification, such as HZCRR and LSTER, are presented and discussed.... 

Naive Bayes (NB) classifier ease of use along with its remarkable performance has led many researchers to extend the scope of its applications to real-world domains by relaxing the conditional independence assumption of features given the information about the class variable. However, fulfilling this objective, most of the generalizations, cut their own way through compromising the model's simplicity, make more complex classifiers with a substantial deviation from the original one. Multi Independent Latent Component Naive Bayes Classifier (MILC-NB) leverages a set of latent variables to preserve the overall structure of naive Bayes classifier while rectifying its major restriction. Each... 

Naive Bayes (NB) classifier ease of use along with its remarkable performance has led many researchers to extend the scope of its applications to real-world domains by relaxing the conditional independence assumption of features given the information about the class variable. However, fulfilling this objective, most of the generalizations, cut their own way through compromising the model's simplicity, make more complex classifiers with a substantial deviation from the original one. Multi Independent Latent Component Naive Bayes Classifier (MILC-NB) leverages a set of latent variables to preserve the overall structure of naive Bayes classifier while rectifying its major restriction. Each... 

In this paper, the use of simulated annealing (SA) metaheuristic for constructing a fuzzy classification system is presented. In several previous investigations, the capability of fuzzy systems to solve different kinds of problems has been demonstrated. Simulated annealing based fuzzy classification system (SAFCS), hybridizes the learning capability of SA metaheuristic with the approximate reasoning method of fuzzy systems. The objective of this paper is to illustrate the ability of SA to develop an accurate fuzzy classifier. The use of SA in classification is an attempt to effectively explore and exploit the large search space usually associated with classification problems, and find the... 

Recent years have seen an increasing trend in developing 3D action recognition methods. However, despite the advances, existing models still suffer from some major drawbacks including the lack of any provision for recognizing action sequences with some missing frames. This significantly hampers the applicability of these methods for online scenarios, where only an initial part of sequences are already provided. In this paper, we introduce a novel sequence-To-sequence representation-based algorithm in which a query sample is characterized using a collaborative frame representation of all the training sequences. This way, an optimal classifier is tailored for the existing frames of each query... 

Identification of voltage and current disturbances is an important task in power system monitoring and protection. In this paper, a new algorithm for online characterization of a wide range of voltage disturbances based on two dimensional wavelet transformation is proposed. This algorithm is more complicated than algorithms based on one dimensional wavelet transformation, but it's more precise and is useful for steady state disturbances, transients with slow variations and transients with rapid changes. After each five cycles, a matrix is formed based on the last fourteen cycles, in a way that the voltage signal in one cycle forms one row of the matrix. Then, the resulted image is decomposed... 

In this paper Long Term Memory (LTM) process during leftward and rightward orientation recalling have been analyzed using Magnetoencephalogram (MEG) signals. This paper presents a novel criterion for decision making using union subspace based classifier. The proposed method involves the Eigenvalues from Singular Value Decomposition (SVD) of each subspace not only to select basis for each subspace but also to weight the decision making criterion to discriminate two classes. The proposed method has provided orientation detection from recalling signal with 6.75 percent increase in classification accuracy compared to better results on this data  

Integrity verification or forgery detection of an image is a difficult procedure, since the forgeries use various transformations to create an altered image. Pixel mapping transforms, such as contrast enhancement, histogram equalisation, gamma correction and so on, are the most popular methods to improve the objective property of an altered image. In addition, fabricators add Gaussian noise to the altered image in order to remove the statistical traces produced because of pixel mapping transforms. A new method is introduced to detect and classify four various categories including original, contrast modified, histogram-equalised and noisy images. In the proposed method, the absolute value of... 

In this paper, we present a new method to solve tracking problem. The proposed method combines sparse representation and motion estimation to track an object. Recently. sparse representation has gained much attention in signal processing and computer vision. Sparse representation can be used as a classifier but has high time complexity. Here, we utilize motion information in order to reduce this computation time by not calculating sparse codes for all the frames. Experimental results demonstrates that the achieved result are accurate enough and have much less computation time than using just a sparse classifier  

In this paper, Kullback-Leibler Distance (KLD) is employed to measure the dissimilarity between marginal statistical features of contourlet transform to fingerprint identification. Conourlet transform is a non separable two dimensional transform which can well capture the geometry of edges in the images which convey important information for the human visual system (HVS). Here, marginal statistics of each transform subband are modeled by a Generalized Gaussian Density (GGD) model and the GGD parameters-α and β- are granted as the extracted features from the corresponding subbands and the fingerprint recognition is done based on k-NN classifier employing Kullback-Leibler Distance (KLD)... 

This study presents an intelligent fault classification method for identification of transformer winding fault through transfer function (TF) analysis. For this analysis support vector machine (SVM) is used. The required data for training and testing of SVM are obtained by measurement on two groups of transformers (one is a classic 20 kV transformer and the other is a model transformer) under intact condition and under different fault conditions (axial displacement, radial deformation, disc space variation and short circuit of winding). Two different features extracted from the measured TFs are then used as the inputs to SVM classifier for fault classification. The accuracy of proposed... 

The voluminous of the e-mails are spam. Several algorithms are represented for spam detection based on batch learning. In this paper, a new algorithm based on incremental learning is introduced. The algorithm composes new knowledge from new training data with previous knowledge by combining classifiers based on weighted majority voting. The experiment results show that the proposed algorithm outperforms other related incremental algorithms and non-incremental algorithms  

An appropriate pre-processing algorithm in classification is not only of great importance with respect to classifier choice, but also would be more crucial. In this paper, a pre-processing step is proposed in order to increase accuracy of classification. The aim of this approach is finding a transformation matrix causes classes to be more discriminable by transforming data into the new space and consequently, increases the classification accuracy. This transformation matrix is computed through two methods based on linear discrimination. In the first method, we use class independent LDA to increase classification accuracy by finding a transformation that maximizes the between-class scatter...