检索结果-内蒙古大学图书馆

High Performance Big Data and Intelligent Systems (HPBD&IS)

作者： Huihui Cai Yakun Zhang Liang Xie Huijiong Yan Wei Qin Ye Yan Erwei Yin Minpeng Xu Dong Ming School of Academic of Medical Engineering and Translational Medicine Tianjin University Tianjin China Tianjin Artifical Intelligence Innoation Center(TAIIC) Tianjin China Defense Innovation Institute Academy of Military Sciences(AMS) Beijing China

ISBN: (纸本)9781665413282

Silent speech recognition (SSR) is a new application of human-computer interaction based on electromyography (EMG), which solves the limitation of acoustic signal dependence. In a low signal-to-noise ratio (SNR) environment, traditional methods cannot accurately segment the EMG active signal. This paper proposes an energy detection method based on spectral subtraction backtracking for detecting EMG active signals to assist silent speech recognition. The experiments are mainly based on energy detection. In addition to the energy detection, the spectral subtraction method is also used to improve the SNR and the accuracy of the endpoint information. Then, the experiments propose a backtracking method to make up for the deficiency of spectral subtraction. Finally, this paper adopts an end-to-end network model, which takes the pre-trained model of CNN as the front end, and the bidirectional gate recurrent unit (Bi-GRU) as the back end for classification. Experimental results show that the proposed activity detection method in this paper is more accurate than others in the low SNR environment.

关键词： Human computer interaction Backtracking Speech recognition Logic gates Electromyography Acoustics detection algorithms

来源：评论

学校读者我要写书评

暂无评论

Practical Methods for Joint Time and Carrier Synchronization in LPI/LPD Communications

Practical Methods for Joint Time and Carrier Synchronization...

引用

MILCOM, Military Communications Conference

作者： Haotian Zhai Bernd-Peter Paris Department of Electrical and Computer Engineering George Mason University Fairfax VA

To operate at low SNR, LPI/LPD communication systems rely on coherent processing which requires that the receiver is precisely synchronized in time, frequency and phase. To that end, this paper proposes a family of data-aided joint frame and carrier synchronization algorithms. A sequential detection algorithm based on the generalized likelihood ratio test (GLRT) is used to detect an embedded preamble signal in the received data stream. Novel algorithms for low-complexity, coarse carrier synchronization at low SNR provide carrier estimates during sequential detection. After detection, the coarse carrier estimate is refined for use in coherent demodulation. The proposed family of algorithms can be scaled to support operation over a wide range of SNR, including SNR below 0 dB. algorithms are validated through simulation. The practicality of our approach is demonstrated by real-time operation on a standard SDR platform with sample rates approaching 10MHz.

关键词： Military communication Time-frequency analysis Receivers Real-time systems Complexity theory Synchronization detection algorithms

来源：评论

学校读者我要写书评

暂无评论

Region Growing Algorithm Combined With Fast Peak detection for Segmenting Colloidal Gold Immunochromatographic Strip Images

引用

IEEE ACCESS 2019年 7卷 169715-169723页

作者： Guo, Wensheng Zhang, Yue Hu, Xiaoyan Zhang, Ting Liang, Ming Yang, Xiangliang Yang, Hai Huazhong Univ Sci & Technol Coll Life Sci & Technol Natl Engn Res Ctr Nanomed Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Engn Sci Wuhan Natl Lab Optoelect Wuhan 430074 Peoples R China Infinitus China Co Ltd Res & Dev Ctr Guangzhou 510663 Peoples R China

Accurate segmentation of test line and control line for colloidal gold immunochromatographic strip (GICS) images with image processing algorithms is essential to quantitative analysis of GICS. As common methods for GICS image segmentation, fuzzy c-means (FCM) algorithm and cellular neural network (CNN) algorithm both require presetting initial conditions (specifying initial parameters or training models) and take long running time, due to high calculation cost. Therefore neither is ideal for a point-of-care testing (POCT) device, which has low hardware cost and limited computing power. This paper designs a region growing algorithm combined with fast peak detection (RGFPD) to quickly and self-adaptively segment GICS images. Compared with FCM algorithm and CNN algorithm, the RGFPD algorithm has two obvious advantages. First, as a local algorithm rather than a local one, the region growing algorithm requires low calculation cost, which better suits the POCT device. Second, the fast peak detection algorithm calculates the seed points and growing criterions as initial conditions, realizing self-adaptive segmentation of images. In this paper, RGFPD algorithm is applied to segment GICS images of actual samples, taking FCM algorithm and CNN algorithm as contrast. The results show that RGFPD accurately segment images without presetting initial conditions, with shorter algorithm running time, and performs better in anti-interference.

关键词： Image segmentation detection algorithms Clustering algorithms Gray-scale Gold Strips Immune system Colloidal gold immunochromatographic strip image segmentation peak detection region growing point-of-care testing

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Deep Slow Feature Analysis for Change detection in Multi-Temporal Remote Sensing Images

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2019年第12期57卷 9976-9992页

作者： Du, Bo Ru, Lixiang Wu, Chen Zhang, Liangpei Wuhan Univ Sch Comp Sci Wuhan Peoples R China Wuhan Univ Collaborat Innovat Ctr Geospatial Technol Wuhan 430072 Peoples R China Wuhan Univ State Key Lab Informat Engn Surveying Mapping & R Wuhan 430072 Peoples R China

Change detection has been a hotspot in the remote sensing technology for a long time. With the increasing availability of multi-temporal remote sensing images, numerous change detection algorithms have been proposed. Among these methods, image transformation methods with feature extraction and mapping could effectively highlight the changed information and thus has a better change detection performance. However, the changes of multi-temporal images are usually complex, and the existing methods are not effective enough. In recent years, the deep network has shown its brilliant performance in many fields, including feature extraction and projection. Therefore, in this paper, based on the deep network and slow feature analysis (SFA) theory, we proposed a new change detection algorithm for multi-temporal remotes sensing images called deep SFA (DSFA). In the DSFA model, two symmetric deep networks are utilized for projecting the input data of bi-temporal imagery. Then, the SFA module is deployed to suppress the unchanged components and highlight the changed components of the transformed features. The change vector analysis pre-detection is employed to find unchanged pixels with high confidence as training samples. Finally, the change intensity is calculated with chi-square distance and the changes are determined by threshold algorithms. The experiments are performed on two real-world data sets and a public hyperspectral data set. The visual comparison and the quantitative evaluation have shown that DSFA could outperform the other state-of-the-art algorithms, including other SFA-based and deep learning methods.

关键词： Feature extraction Remote sensing Change detection algorithms detection algorithms Eigenvalues and eigenfunctions Artificial neural networks Training Change detection deep network remote sensing images slow feature analysis (SFA)

来源：评论

学校读者我要写书评

暂无评论

Classification of Indian Classical Music Ragas using Graph Theory

Classification of Indian Classical Music Ragas using Graph T...

引用

International Conference on Next Generation Intelligent Systems (ICNGIS)

作者： Aatmaj Mhatre Computer Science and Engineering K.J. Somaiya College of Engineering. K.J. Somaiya College of Engineering Mumbai India

A Raga is a melodic framework of a certain structure of notes. The Ragas of Indian classical music are roughly divided into 10 divisions by V.N. Bhatkhande based on certain rules and specific melodic styles. This is known as the traditional ’ Thaat’ system. This study aims at grouping Ragas using graph clustering methods and compare the results with the traditional methods of classification. This classification is based on the similarity between ragas on basis of 35 parameters based on the presence and absence of notes and characteristic notes. In this study, graphs are classified with two methods. One of them is a community detection algorithm using mixture models. The other method is based on classification using the nearest neighbour where traditional group members are selected as the classification criterion. In this study, the aim is not classifying music into ragas, on which a lot of research has been previously done, but classifying ragas themselves into groups. Our results match well with the traditional methods of classification.

关键词： Clustering methods Mixture models Graph theory Classification algorithms Intelligent systems detection algorithms Next generation networking

来源：评论

学校读者我要写书评

暂无评论

Community detection Based on Local Information and Dynamic Expansion

引用

IEEE ACCESS 2019年 7卷 142773-142786页

作者： Luo, Yongping Wang, Li Sun, Shiwen Xia, Chengyi Tianjin Univ Technol Key Lab Intelligence Comp & Novel Software Techno Tianjin 300384 Peoples R China Minist Educ Key Lab Comp Vis & Syst Tianjin 300384 Peoples R China

Mining the community structure in the real-world networks has been a hot topic in the field of complex networks, and has emerged as a prominent research area and continues to grow with the introduction of requirements for personalized recommendation. However, most of the existing community detection algorithms are based on global information, fewer works are devoted to detecting the communities hidden in the network by using local information. To this end, in this paper, we propose two improved signed modularity functions to evaluate the community properties in complex network, and then we apply these indicators to identify the community structure by using the local information since it is difficult to obtain the global information in practice. During the dynamic expansion, each local community will absorb the neighboring node with the highest positive energy;in addition, a new local community is generated when all local communities cannot contain the neighboring node. Finally, the algorithm has been applied to unsigned networks and signed networks, respectively. The experimental results show that the division results given by our proposed algorithm are in line with the actual ones in artificial and real-life networks.

关键词： Heuristic algorithms detection algorithms Clustering algorithms Image edge detection Complex networks Social network services Symmetric matrices Community detection global community local modularity local information local community

来源：评论

学校读者我要写书评

暂无评论

Novelty detection in Social Media by Fusing Text and Image Into a Single Structure

引用

IEEE ACCESS 2019年 7卷 132786-132802页

作者： Amorim, Marta Bortoloti, Frederico D. Ciarelli, Patrick M. Salles, Evandro O. T. Cavalieri, Daniel C. Univ Fed Espirito Santo Elect Engn Dept BR-29075910 Vitoria ES Brazil Fed Inst Espirito Santo Automat & Control Engn Dept BR-29173087 Serra Brazil

This work aims to propose an approach for detecting novelties, taking into account the temporal flow of data streams in social media. To this end, we present a completely new architecture for novelty detection. This new architecture entails three new contributions. First, we propose a new concept for novelty definition based on temporal windows. Second, we formulate an expression to determine the quality of a novelty. Third, we introduce a new approach to the fusion of heterogeneous data (image + text), using the COCO dataset and the MASK-RCNN convolutional neural network, which transforms image and text from social media into a single data format ready to be identified by machine learning algorithms. Since novelty detection is a task in which labeled samples are scarce or inexistent, unsupervised algorithms are used, and thus, the following baseline and state-of-the-art algorithms have been chosen: kNN, HBOS, FBagging, IForesting, and autoencoders. The new fusion approach is also compared to a state-of-the-art approach to outlier detection named AOM. Because of temporal particularities and the data types being fused, a new dataset was created, containing 27,494 tweets collected from Twitter. Our experiments show that data classification of social media using data fusion is superior to using only text or only images as input data.

关键词： detection algorithms knowledge based systems learning systems machine learning neural networks classification algorithms streaming media

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Rotation-Invariant Haar-Like Feature Integrated CNN-Based Ship detection Algorithm of Multiple-Target Environment in SAR Imagery

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2019年第12期57卷 10070-10087页

作者： Ai, Jiaqiu Tian, Ruitian Luo, Qiwu Jin, Jing Tang, Bo Hefei Univ Technol Key Lab Knowledge Engn Big Data Minist Educ Hefei 230009 Anhui Peoples R China Hefei Univ Technol Sch Comp Sci & Informat Engn Hefei 230009 Anhui Peoples R China Anhui Prov Key Lab Ind Safety & Emergency Technol Hefei 230009 Anhui Peoples R China Cent South Univ Sch Automat Changsha 410083 Peoples R China Mississippi State Univ Dept Elect & Comp Engn Starkville MS 39762 USA

This paper proposes a multi-scale rotation-invariant haar-like (MSRI-HL) feature integrated convolutional neural network (MSRIHL-CNN)-based ship detection algorithm of the multiple-target environment in synthetic aperture radar (SAR) imagery. Usually, ship detection includes preprocessing, prescreening, discrimination, and classification. Among them, prescreening and discrimination are the most two important stages so that they catch great intention. Based on our previous work, we propose a truncated-clutter-statistics-based joint, constant false alarm rate (CFAR) detector (TCS-JCFAR) for ship target prescreening in the multiple-target environment. TCS-JCFAR greatly enhances the prescreening rate in the multiple-target environment while achieving a low observed FAR. In the discrimination stage, conventional CNN extracts the deep features (high-level features);however, it will lose the local texture and edge information (low-level features) which are of great significance for target discrimination. Hence, the MSRI-HL features are used to represent the multi-scale, rotation-invariant texture, and edge information that conventional CNN fails to capture. The extracted low-level MSRI-HL features and the high-level deep features are optimally fused to a multi-layered feature vector. Finally, the multi-layered feature vector is fed into a typical support vector machine (SVM) classifier for ship target discrimination. The proposed MSRIHL-CNN combines the low-level texture and edge features and the high-level deep features;moreover, they are optimally fused to fully represent the ship targets. Undoubtedly, MSRIHL-CNN has better discrimination performance. The superiority of the proposed TCS-JCFAR-based prescreener and MSRIHL-CNN-based discriminator is validated on the Chinese Gaofen-3 SAR imagery.

关键词： Marine vehicles Feature extraction Clutter Synthetic aperture radar Radar polarimetry Correlation detection algorithms Multi-layered feature fusion multi-scale rotation-invariant haar-like (MSRI-HL) feature integrated convolutional neural network (MSRIHL-CNN)-based discrimination multiple-target environment synthetic aperture radar (SAR) ship detection truncated-clutter-statistics-based joint constant false alarm rate (CFAR) detector (TCS-JCFAR)-based prescreening

来源：评论

学校读者我要写书评

暂无评论

Inter-Patient CNN-LSTM for QRS Complex detection in Noisy ECG Signals

引用

IEEE ACCESS 2019年 7卷 169359-169370页

作者： Yuen, Brosnan Dong, Xiaodai Lu, Tao Univ Victoria Dept Elect & Comp Engn Victoria BC V8P 5C2 Canada

In this paper, a convolutional neural network (CNN) with long short-term memory (LSTM) is designed to detect QRS complexes in noisy electrocardiogram (ECG) signals. The CNN performs feature extraction while the LSTM determines the QRS complex timings. A multi-layer perception (MLP) after the LSTM is added to format the QRS complex detection predictions. With a unique data preparation procedure that includes proper design of training dataset, the proposed CNN-LSTM can achieve superior inter-patient testing performance, which means the testing and training datasets do not share any same patient ECG records. This generalization ability characteristic is critical to automated ECG analysis in an age of big data collected from noisy wearable ECG devices. The MIT-BIH and the European ST-T noise stress test databases are used to validate the effectiveness of the proposed algorithm in terms of sensitivity (recall), positive predictive value (precision), F-1 score and timing root mean square error of R peak positions.

关键词： Electrocardiography Testing Training detection algorithms Databases Noise measurement Neural networks Artificial neural networks electrocardiogram (ECG) QRS complex feedforward neural networks multi-layer neural network convolutional neural networks recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

SSAAM: Sentiment Signal-based Asset Allocation Method with Causality Information

SSAAM: Sentiment Signal-based Asset Allocation Method with C...

引用

IEEE International Conference on Big Data

作者： Rei Taguchi Hiroki Sakaji Kiyoshi Izumi School of Engineering The University of Tokyo Tokyo Japan

ISBN: (纸本)9781665480468

This study demonstrates whether financial text is useful for tactical asset allocation using stocks by using natural language processing to create polarity indexes in financial news. In this study, we performed clustering of the created polarity indexes using the change-point detection algorithm. In addition, we constructed a stock portfolio and rebalanced it at each change point utilizing an optimization algorithm. Consequently, the asset allocation method proposed in this study outperforms the comparative approach. This result suggests that the polarity index helps construct the equity asset allocation method.

关键词： Clustering algorithms Big Data Natural language processing Inference algorithms Resource management Indexes detection algorithms

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：