检索结果-内蒙古大学图书馆

Unsupervised noise removal technique based on constrained NMF

IET SIGNAL PROCESSING 2017年第7期11卷 788-795页

作者： Vaghmare, Anil Kumar Rao, Chevula Venkata Rama Natl Inst Technol Dept Elect & Commun Engn Shillong Meghalaya India

This paper presents a novel approach to mitigate background interference noise in a noisy audio environment based on unsupervised speech enhancement without the prior knowledge of speech as well as noise signal. In this scheme, a noisy speech signal is represented by a non-negative matrix factorisation (NMF) with sparseness constraints. Using NMF with constraints, noise density spectrum is estimated to design the spectral subtraction method. In the earlier method of NMF, correlation between the noise and the audio signals was not considered. The proposed method explores the correlation between the noise and speech signals. A sparse constraint NMF-based spectral subtraction method is designed to eliminate the noise. The performance is evaluated by considering various noise environments in terms of quality and intelligibility measures. Simulation results highlight the improvement in the enhanced speech with the suggested approach.

关键词： speech enhancement speech intelligibility signal denoising matrix decomposition interference suppression correlation methods spectral analysis unsupervised noise removal technique constrained NMF background interference noise mitigation noisy audio environment unsupervised speech enhancement noisy speech signal representation nonnegative matrix factorisation noise density spectrum noise-speech signal correlation sparse constraint NMF-based spectral subtraction method intelligibility measures quality measures

来源：评论

学校读者我要写书评

暂无评论

Word cloud segmentation for simplified exploration of trending topics on Twitter

引用

IET SOFTWARE 2017年第5期11卷 214-220页

作者： Shahid, Nabila Ilyas, Muhammad U. Alowibdi, Jalal S. Aljohani, Naif R. Natl Univ Sci & Technol SEECS Applied Network & Data Sci Res AN DASH Lab Islamabad 44000 Pakistan Univ Jeddah Fac Comp & Informat Technol Dept Comp Sci Jeddah Mecca Province Saudi Arabia King Abdulaziz Univ Fac Comp & Informat Technol Jeddah Saudi Arabia

Twitter is a popular microblogging platform, with 310 million monthly active users as of the first quarter of 2016. It is a rapidly growing microblogging platform where people share opinions, news on any topic of their interest. More than 7000 tweets are posted every second. Due to the enormous volume of data being generated, it becomes difficult to extract useful/meaningful information. Tweets collected from Twitter on a certain topic may consist of numerous conversation threads about relevant sub-topics. However, it is difficult to discern these sub-topics if the data is visualised as a single word cloud. The authors transform a corpus of tweets to a spectral domain and evaluate the results from a number of clustering algorithms, including K-means, latent semantic indexing and non-negative matrix factorisation to construct clustered word clouds that helps identify sub-topics under a broader topic.

关键词： cloud computing natural language processing social networking (online) pattern clustering indexing matrix decomposition word cloud segmentation trending topic exploration Twitter microblogging platform spectral domain clustering algorithms k-means latent semantic indexing nonnegative matrix factorisation clustered word clouds

来源：评论

学校读者我要写书评

暂无评论

Novel feature extraction method for cough detection using NMF

引用

IET SIGNAL PROCESSING 2017年第5期11卷 515-520页

作者： You, Mingyu Wang, Huihui Liu, Zeqin Chen, Chong Liu, Jiaming Xu, Xiang-Huai Qiu, Zhong-Min Tongji Univ Dept Control Sci & Engn Shanghai Peoples R China Tongji Univ Dept Resp Med Tongji Hosp Sch Med Shanghai Peoples R China

Cough is a common symptom in respiratory diseases. To provide valuable clinical information for cough diagnosis and monitoring, objectively evaluating the quantity and intensity of cough based on cough detection by pattern recognition technologies is needed. Cough detection aims to extract the boundaries of cough events from an audio stream. From spectral visualisation, it is found that the energy spectrum of cough signal spreads widely in the whole frequency band, which is very different from a speech signal. However, almost all feature extraction methods for cough detection in the previous work are derived from speech recognition region. In this study, to find the difference of cough and other audios in a more compact representation, non-negative matrix factorisation (NMF) is exploited to extract the spectral structure from signals. Furthermore, the spectral structure from cough signal can be used as filter banks of feature extraction methods, which makes the filter banks more suitable for cough detection than manually designed ones. Besides, parameterisation for the spectral structure also provides an optimising strategy for the authors' NMF-based feature extraction method. Experiments are conducted on real data. The results demonstrate that NMF-based feature extraction method has considerable potential in improving performance for cough detection.

关键词： feature extraction diseases pattern recognition audio streaming matrix decomposition channel bank filters feature extraction method cough detection respiratory diseases cough diagnosis pattern recognition technologies audio stream energy spectrum speech signal speech recognition region nonnegative matrix factorisation filter banks feature extraction methods parameterisation spectral structure NMF-based feature extraction method

来源：评论

学校读者我要写书评

暂无评论

Effective supervised multiple-feature learning for fused radar and optical data classification

引用

IET RADAR SONAR AND NAVIGATION 2017年第5期11卷 768-777页

作者： Karimi, Danya Akbarizadeh, Gholamreza Rangzan, Kazem Kabolizadeh, Mostafa Shahid Chamran Univ Ahvaz Dept Remote Sensing & GIS Fac Earth Sci Ahvaz Iran Shahid Chamran Univ Ahvaz Dept Elect Engn Fac Engn Ahvaz Iran

In multi-sensor data fusion based on multiple features, the high dimensionality of feature space increases the runtime and computational complexity. The present study proposes a new algorithm based on the combination of random subspace (RS), linear discriminant analysis and sparse regularisation (LDASR), namely RS-LDASR for feature space dimensionality reduction, supervised feature selection, and learning. The use of RSs can effectively solve the problem of high dimensionality and high feature-to-instance ratio. The extraction of multiple features from the images raises the possibility of a correlation between features which reduce classification accuracy. In this study, after the construction of several RSs, supervised feature selection and learning based on LDASR were applied with very high accuracy. Classification and image fusion for remote sensing data analysis were tested by the implementation of feature-based fusion on two pairs of fused synthetic aperture radar and optical data. Four feature matrices were constructed using attribute profiles (APs), multi-APs (MAPs), non-negative matrix factorisation (NMF), and textural features. Support vector machine and rotation forest were applied as the base classifiers. The results show that use of RS-LDASR significantly improved the classification accuracy based on NMF plus texture features and even NMF alone.

关键词： geophysical image processing image texture matrix decomposition remote sensing by radar correlation methods feature extraction random processes computational complexity image fusion image classification learning (artificial intelligence) support vector machines radar imaging supervised multiple-feature learning fused radar data classification optical data classification multisensor data fusion computational complexity random subspace linear discriminant analysis and sparse regularisation RS-LDASR feature space dimensionality reduction supervised feature selection feature-to-instance ratio feature extraction feature correlation classification accuracy reduction image fusion remote sensing data analysis feature-based fusion fused synthetic aperture radar feature matrices attribute profiles multiAP MAP nonnegative matrix factorisation NMF textural features support vector machine rotation forest

来源：评论

学校读者我要写书评

暂无评论

MINI-BATCH STOCHASTIC APPROACHES FOR ACCELERATED MULTIPLICATIVE UPDATES IN nonnegative matrix factorisationWITH BETA-DIVERGENCE 26

MINI-BATCH STOCHASTIC APPROACHES FOR ACCELERATED MULTIPLICAT...

引用

26th IEEE International Workshop on Machine Learning for Signal Processing (MLSP)

作者： Serizel, Romain Essid, Slim Richard, Gael Univ Paris Saclay LTCI CNRS Telecom ParisTech F-75013 Paris France

ISBN: (纸本)9781509007462

nonnegative matrix factorisation (NMF) with beta-divergence is a popular method to decompose real world data. In this paper we propose mini-batch stochastic algorithms to perform NMF efficiently on large data matrices. Besides the stochastic aspect, the mini-batch approach allows exploiting intensive computing devices such as general purpose graphical processing units to decrease the processing time and in some cases outperform coordinate descent approach.

关键词： nonnegative matrix factorisation GPGPU multiplicative rules online learning

来源：评论

学校读者我要写书评

暂无评论

Polarimetric SAR image fusion using nonnegative matrix factorisation and improved-RGB model

引用

ELECTRONICS LETTERS 2010年第20期46卷 1399-1401页

作者： Guo, R. Zhang, L. Xing, M. Li, J. Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Shaanxi Peoples R China

Since different polarimetrie channels have different dynamical ranges, nonnegative matrix factorisation (NMF) is explored for fusing polarimetric SAR (PolSAR) data in a single colour image. The data of the three polarimetric channels (HH, HV and VV) with total power (TP) data is fully utilised to implement the NMF, and the choice of normalised factors for different polarimetric channels based on the same probability is applied, which produces an improved-RGB model. Combining the improved-RGB model with NMF, this approach brings in fusion images with higher spatial resolution and the power component preserved.

关键词： image fusion improved-RGB model matrix decomposition nonnegative matrix factorisation polarimetrie SAR image fusion polarimetrie channels probability radar imaging radar polarimetry synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Time-Dependent Passive Building Thermography for Detecting Delamination of Adhered Ceramic Cladding

引用

JOURNAL OF NONDESTRUCTIVE EVALUATION 2015年第3期34卷 1-16页

作者： Edis, Ecem Flores-Colen, Ines De Brito, Jorge Istanbul Tech Univ Fac Architecture TR-34437 Istanbul Turkey Univ Lisbon DECivil IST P-1049001 Lisbon Portugal

Time-dependent passive infrared thermography (td-PIRT) is an inspection technique that can be used to detect delamination. td-PIRT was used to inspect two buildings in Lisbon (Portugal) with adhered ceramic cladding to evaluate its potential to detect delamination problems in such claddings, which is a common problem and has safety risks. The td-PIRT data were analysed quantitatively by means of simple image subtraction (SIS), nonnegative matrix factorisation (NMF), and principal component analysis (PCA). Comparative evaluations showed that td-PIRT has a great potential to detect delamination problems in adhered ceramic cladding, depending on the analysis method selected. PCA and NMF were found to be efficient methods because of their good, consistent performance in revealing defective areas.

关键词： Time-dependent passive infrared thermography Delamination Adhered ceramic tile Simple image subtraction nonnegative matrix factorisation Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Automatic image annotation by a loosely joint non-negative matrix factorisation

引用

IET COMPUTER VISION 2015年第6期9卷 806-813页

作者： Rad, Roya Jamzad, Mansour Sharif Univ Technol Dept Comp Engn Tehran Iran

Nowadays, the number of digital images has increased so that the management of this volume of data needs an efficient system for browsing, categorising and searching. Automatic image annotation is designed for assigning tags to images for more accurate retrieval. Non-negative matrix factorisation (NMF) is a traditional machine learning technique for decomposing a matrix into a set of basis and coefficients under the non-negative constraints. In this study, the authors propose a two-step algorithm for designing an automatic image annotation system that employs the NMF framework for its first step and a variant of K-nearest neighbourhood as its second step. In the first step, a new multimodal NMF algorithm is proposed to extract the latent factors which reflect the content of images. This is done by jointly factorising the visual and textual data feature matrices so that they have close representation, although not necessarily the same. In the second step, after mapping images to the latent factors space a few tags are predicted for the new images based on a weighted average of similar data. They evaluated the performance of the proposed method and compared it to the state-of-the-art literature. Comparison results demonstrate the effectiveness and potential of the proposed method in image annotation applications.

关键词： image classification image retrieval matrix decomposition learning (artificial intelligence) visual databases nonnegative matrix factorisation multimodal NMF algorithm digital images machine learning technique automatic image annotation system K-nearest neighbourhood visual data feature matrices textual data feature matrices

来源：评论

学校读者我要写书评

暂无评论

Target recognition in synthetic aperture radar images via non-negative matrix factorisation

引用

IET RADAR SONAR AND NAVIGATION 2015年第9期9卷 1376-1385页

作者： Cui, Zongyong Cao, Zongjie Yang, Jianyu Feng, Jilan Ren, Hongliang Univ Elect Sci & Technol China Sch Elect Engn Chengdu 611731 Peoples R China Natl Univ Singapore Fac Engn Singapore 117575 Singapore Natl Univ Singapore Adv Robot Ctr Singapore 117575 Singapore

This study proposes a novel non-negative matrix factorisation (NMF) variant L-1/2-NMF after visualisation and analysis of the process of target recognition via NMF for synthetic aperture radar (SAR) images. NMF has been applied to obtain pattern feature in SAR images. This study considers the intrinsic character and the physical meaning of NMF feature when applied for SAR automatic target recognition. At the base of obtaining the linear relationship between the sample to be recognised and the train samples, the whole recognition process is detailed and vividly visualised. Meanwhile, lots of researches have been done to improve NMF methods by enforcing sparse constraint with L-1-norm, such as non-negative sparse coding (NNSC), local NMF and sparse NMF. Compared with L-1-norm, L-1/2-norm has been shown to have a more natural sparseness. In this study, a novel variant of NMF with L-1/2 constraint, called L-1/2-NMF is proposed, and is carried out a thorough study by applying it in SAR target recognition. Experimental results on MSTAR public database show that both the basis and coding matrices obtained by L-1/2-NMF have higher sparseness than those obtained by NMF, NNSC and NMF with sparseness constraints (NMFsc). The recognition results demonstrate that the L-1/2-NMF outperforms NNSC, NMFsc and non-smooth NMF.

关键词： radar target recognition radar imaging matrix decomposition feature extraction image recognition image coding synthetic aperture radar images nonnegative matrix factorisation SAR images pattern feature NMF feature SAR automatic target recognition NMF method L-1-norm nonnegative sparse coding NNSC local NMF sparse NMF L1-2-norm L1-2-NMF MSTAR public database coding matrices sparseness constraint nonsmooth NMF

来源：评论

学校读者我要写书评

暂无评论

Graph clustering by congruency approximation

引用

IET COMPUTER VISION 2015年第6期9卷 841-849页

作者： Ren, Weiya Li, Guohui Tu, Dan Natl Univ Def Technol Coll Informat Syst & Management Changsha 410073 Hunan Peoples R China

The authors consider the general problem of graph clustering. Graph clustering manipulates the graph-based data structure and the entries of the solution vectors are only allowed to take non-negative discrete values. Finding the optimal solution is NP-hard, so relaxations are usually considered. Spectral clustering retains the orthogonality rigorously but ignores the non-negativity and discreteness of the solution. Sym non-negative matrix factorisation can retain the non-negativity rigorously but it is hard to reach the orthogonality. In this study, they proposed a novel method named congruent approximate graph clustering (CAC), which can retain the non-negativity rigorously and can reach the orthogonality properly by congruency approximation. Furthermore, the solution obtained by CAC is sparse, which is approximate with the ideal discrete solution. Experimental results on several real image benchmark datasets indicate that CAC achieves encouraging results compared with state-of-the-art methods.

关键词： graph theory pattern clustering computational complexity matrix decomposition congruency approximation graph-based data structure solution vectors nonnegative discrete values NP-hard optimal solution spectral clustering nonnegative matrix factorisation congruent approximate graph clustering CAC ideal discrete solution real image benchmark datasets

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：