Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Voiced/Unvoiced Classification and Pitch Period Detection Algorithm Based on Wavelet Transform

Hu Ying,Chen Ning,

电子与信息学报 , 2008,
Abstract: This paper proposes a robust pitch period detection method based on wavelet transformation. A Voiced Regions Detection(VRD)algorithm based on wavelet transform and Teager energy operator is proposed firstly. Then an algorithm based on spatial correlation function for estimating pitch frequency only in voiced regions is presented. Experiments show that this algorithm has a better robustness and more precision compared with the classical wavelet-based methods and auto-correlated function (ACF).
Voiced/unvoiced Decision Based on Recurrence Quantification Analysis

Yan Run-qiang,Zhu Yi-sheng,

电子与信息学报 , 2007,
Abstract: Voiced/unvoiced decision is an important component in speech signal processing. In this paper, different topological structures in Recurrence Plots (RPs) are described for the different physical models of speech production. By statistically analyzing the determinism and the normalized maximal length of diagonal structures acquired from Recurrence Quantification Analysis (RQA), a flexible and efficient decision framework is proposed. Comparing with some traditional methods, the proposed algorithm has lower wrong decision rate. The method provides a new way for feature extraction and speech recognition.
Automatic Classification for Various Images Collections Using Two Stages Clustering Method  [PDF]
Wan Hyun Cho, In Seop Na, Jun Yong Choi, Tae Hoon Lee
Open Journal of Applied Sciences (OJAppS) , 2013, DOI: 10.4236/ojapps.2013.31B010
Abstract: In this paper, we propose an automatic classification for various images collections using two stage clustering method. Here, we have used global and local image features. First, we review about various types of feature vector that is suita-ble to represent local and global properties of images, and similarity measures that can be represented an affinity be-tween these images. Second, we consider a clustering method for image collection. Here, we first build a coarser clus-tering by partitioning various images into several clusters using the flexible Mean shift algorithm and K-mean cluster-ing algorithm. Second, we construct dense clustering of images collection by optimizing a Gaussian Dirichlet process mixture model taking initial clusters as given coarser clustering. Finally, we have conducted the comparative experi-ments between our method and existing methods on various images datasets. Our approach has significant advantage over existing techniques. Besides integrating temporal and image content information, our approach can cluster auto-matically photographs without some assumption about number of clusters or requiring a priori information about initial clusters and it can also generalize better to different image collections.
A method for improved clustering and classification of microscopy images using quantitative co-localization coefficients
Vasanth R Singan, Kenan Handzic, Kathleen M Curran, Jeremy C Simpson
BMC Research Notes , 2012, DOI: 10.1186/1756-0500-5-281
Abstract: We have developed a novel method for combining two existing image analysis approaches, which results in highly efficient and accurate discrimination of proteins with seemingly similar distributions. We have combined image texture-based analysis with quantitative co-localization coefficients, a method that has traditionally only been used to study the spatial overlap between two populations of molecules. Here we describe and present a novel application for quantitative co-localization, as applied to the study of Rab family small GTP binding proteins localizing to the endomembrane system of cultured cells.We show how quantitative co-localization can be used alongside texture feature analysis, resulting in improved clustering of microscopy images. The use of co-localization as an additional clustering parameter is non-biased and highly applicable to high-throughput image data sets.The distribution of proteins to specific subcellular structures in eukaryotic cells allows distinct functions to be performed in parallel. Accurate determination of protein localization is therefore an essential step towards understanding cell function [1]. A variety of methods to automatically annotate subcellular localization have been described [2], primarily using supervised classification methods based on standard subcellular localization profiles [3]. One important image analysis technique for the analysis of large-scale cell-based data is texture-based analysis [4]. Of particular note are the algorithms developed by Haralick, which take account of pixel intensity information in localized areas of an image [5]. Texture-based analyses are a very powerful method to discriminate localization patterns, and as such have been implemented in various commercial and open-source image analysis solutions [6]. Despite the proven application of texture-based methods in the analysis of a variety of cell-based assays, their application in the discrimination of subtle, yet important, localization diffe
International Journal of Engineering Science and Technology , 2011,
Abstract: The distinctive feature of wavelet transforms applications it is used in speech signals. Some problem are produced in speech signals their synthesis, analysis compression and classification. A method being evaluated uses wavelets for speech analysis and synthesis distinguishing between voiced and unvoiced speech, determining pitch, and methods for choosing optimum wavelets for speech compression are discussed. This comparative perception results that are obtained by listening to the synthesized speech using both scalar and vector quantized wavelet parameters are reported in this paper.
Active SVM multi-class classification method based on AP clustering

ZHANG Jian-peng,CHEN Fu-cai,

计算机应用研究 , 2012,
Abstract: For the shortcomings of active learning algorithm existing in multi-class classifier application, such as low accuracy, slow speed and so on, this paper presented an improved active learning algorithm and its application to multi-class SVM. It presented a novel optimization method of training samples with affinity propagation AP clustering algorithm and active learning algorithm for multi-class SVM classification problem. This method choose the most beneficial N new samples added to the training samples for learning in order to spend less marked cost and get a good classification performance. Indicated in many different data set experimental result that, the proposed method gives large reduction in the number of human labeled samples to achieve similar classification accuracy, and has little computational overhead and good robustness.
A Semi-Supervised Clustering Method For P2P Traffic Classification  [cached]
Bin Liu
Journal of Networks , 2011, DOI: 10.4304/jnw.6.3.424-431
Abstract: In the last years, the use of P2P applications has increased significantly and currently they represent a significant portion of the Internet traffic. In consequence of this growth, P2P traffic identification and classification are becoming increasingly important for network administrators and designers. However, this classification was not simple. Nowadays, P2P applications explicitly tried to camouflage the original traffic in an attempt to go undetected. This paper present a methodology and selection of three P2P traffic metrics
Classification of 365 Chinese medicines in Shennong’s Materia Medica Classic based on a semi-supervised incremental clustering method
Rui Jin,Bing Zhang
Zhong Xi Yi Jie He Xue Bao , 2011,
Abstract: : Evidence of the pharmacological activity of traditional Chinese medicine (TCM) provides the basis for clinical prescription. Study of the classification of Chinese medicines according to these activities is key to understanding the general active tendencies of medicinal prescriptions, exploring their material basis, investigating their properties and searching for their alternatives. Taking the herbal classic Shennong’s Materia Medica Classic (Shennong Bencao Jing) for the data source, this paper studied the classification of Chinese medicines based on semi-supervised incremental clustering algorithm using “micro-cluster” concept in order to investigate the complex similarity among Chinese medicines. The results showed that 253 Chinese medicines were reasonably classified into 14 types, such as invigoration, clearing heat, diuresis, dredging blockages in the channels, treating gynecological conditions and treating strange diseases caused by ghosts. The results also showed that the other 112 Chinese medicines were classified into 112 individual types and the same high similarity to different known types was the main reason for this. The semi-supervised incremental clustering algorithm employed in the study had a high quality and a good development for clustering which is suitable for classification of Chinese medicines. This study illustrated the diversity of Chinese medicines and their complex similarities, thus aiming to provide innovative ideas and methods for related research.
International Journal of Engineering Science and Technology , 2011,
Abstract: Differences of physiological properties of the glottis and the vocal track are partly due to age and/or gender differences. Since these differences are reflected in the speech signal, acoustic measures related to those properties can be helpful for automatic gender classification. Acoustics measures of voice sources were extracted from 10 utterances spoken by 20 male and 20 female talkers (aged 19 to 25 year old). Speech long term features, including amplitude, zero crossing rate, short time energy, spectrum flux, and spectrogram is proposed for sex identification. An experimental framework is set-up for these classification task and the result of about 97% for gender classification clearly validate this hypothesis.
Performance Analysis of Gender Clustering and Classification Algorithms  [PDF]
International Journal on Computer Science and Engineering , 2012,
Abstract: In speech processing, gender clustering and classification plays a major role. In both gender clustering and classification, selecting the feature is an important process and the often utilized featurefor gender clustering and classification in speech processing is pitch. The pitch value of a male speech differs much from that of a female speech. Normally, there is a considerable frequency value difference between the male and female speech. But, in some cases the frequency of male is almost equal to female or frequency of female is equal to male. In such situation, it is difficult to identify the exact gender. By considering this drawback, here three features namely; energy entropy, zero crossing rate and short time energy are used for identifying the gender. Gender clustering and classification of speech signal are estimated using the aforementioned three features. Here, the gender clustering is computed using Euclidean distance, Mahalanobis distance, Manhattan distance & Bhattacharyya distance method and the gender classification method is computed using combined fuzzy logic and neural network, neuro fuzzy and support vector machine and its performance are analyzed.
Page 1 /100
Display every page Item

Copyright © 2008-2017 Open Access Library. All rights reserved.