检索结果-内蒙古大学图书馆

Nonlinear Prediction with Deep Recurrent Neural Networks for Non-Blind Audio Bandwidth Extension

China Communications 2018年第4期15卷 72-85页

作者： Lin Jiang Ruimin Hu Xiaochen Wang Weiping Tu Maosheng Zhang National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan 430072 China Institute of Big Data and internet Innovation Hunan University of Commerce Changsha 410205 China Software College East China University of Technology Nanchang 330013 China Collaborative Innovation Center for Economics crime investigation and prevention technology Jiangxi Province Nanchang 330103China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan 430072 China Collaborative Innovation Center of Geospatial Technology Wuhan 430079 China

Non-blind audio bandwidth extension is a standard technique within contemporary audio codecs to efficiently code audio signals at low bitrates. In existing methods, in most cases high frequencies signal is usually generated by a duplication of the corresponding low frequencies and some parameters of high frequencies. However, the perception quality of coding will significantly degrade if the correlation between high frequencies and low frequencies becomes weak. In this paper, we quantitatively analyse the correlation via computing mutual information value. The analysis results show the correlation also exists in low frequency signal of the context dependent frames besides the current frame. In order to improve the perception quality of coding, we propose a novel method of high frequency coarse spectrum generation to improve the conventional replication method. In the proposed method, the coarse high frequency spectrums are generated by a nonlinear mapping model using deep recurrent neural network. The experiments confirm that the proposed method shows better performance than the reference methods.

关键词： audio coding non-blind audiobandwidth extension context correlation deeprecurrent neural network

来源：评论

学校读者我要写书评

暂无评论

Robust visual tracking via combinative deep learning

Robust visual tracking via combinative deep learning

引用

International Conference on Mechanical, Electric and Industrial engineering, MEIE 2018

作者： Wen, Jia Zhao, Jiwei Liu, Pengfei Wang, Hongjun Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University Qinhuangdao China Information Science and Engineering College Yanshan University Qinhuangdao China State Key Laboratory of Software Engineering of Hebei Province Yanshan University Qinhuangdao China

Object tracking is a hot topic in computer vision. In recent years, a large number of trackers has been proposed, in which the deep learning tracker has achieved excellent performance. The real-time capability of the deep learning tracker is not good enough due to the high-complexity of the network structures. This paper proposed an innovative tracking method to solve this problem. There are three important differences between this tracker and the other deep learning trackers. Firstly, the overcomplete basis in the deep learning tracker results in heavy computational cost. In order to reduce the complexity of the network, fewer units are used in the first hidden layer to replace the overcomplete basis. Secondly, a training method combining two observation models is used in the tracking process. The denoising automatic encoder is used in the first layer and the backpropagation is used in the other layers. This can avoid the diffusion of gradients which is caused by BP and adapt to the change of the targets easier. Thirdly, this tracker using adaptive particle filter to track targets. The number of particles is dynamic changes in tracking process. In this paper, we use different kinds of unlabelled datum to train network and initialize observation model. The observation model uses the samples collected in the tracking to adjust dynamically so as to adapt to the target appearance and complex environment. Compared with the existing methods, the results of experiments in different video sequences show that this tracker has a higher speed and the similar accuracy compared. © Published under licence by IOP Publishing Ltd.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A voxel-based filtering algorithm for mobile LiDAR data

A voxel-based filtering algorithm for mobile LiDAR data

引用

2018 ISPRS TC III Mid-Term Symposium on Developments, Technologies and Applications in Remote Sensing

作者： Qin, Hejuan Guan, Haiyan Yu, Yongtao Zhong, Liang College of Remote Sensing and Geometics Engineering Nanjing University of Information Science and Technology Nanjing210044 China Faculty of Computer and Software Engineering Huaiyin Institute of Technology Huaian223003 China Changjiang Spatial Information Technology Engineering Co. Ltd. Wuhan410010 China

This paper presents a stepwise voxel-based filtering algorithm for mobile LiDAR data. In the first step, to improve computational efficiency, mobile LiDAR points, in xy-plane, are first partitioned into a set of two-dimensional (2-D) blocks with a given block size, in each of which all laser points are further organized into an octree partition structure with a set of three-dimensional (3-D) voxels. Then, a voxel-based upward growing processing is performed to roughly separate terrain from non-terrain points with global and local terrain thresholds. In the second step, the extracted terrain points are refined by computing voxel curvatures. This voxel-based filtering algorithm is comprehensively discussed in the analyses of parameter sensitivity and overall performance. An experimental study performed on multiple point cloud samples, collected by different commercial mobile LiDAR systems, showed that the proposed algorithm provides a promising solution to terrain point extraction from mobile point clouds. © Authors 2018. CC BY 4.0 License.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

DS-GCNs: Connectome Classification Using Dynamic Spectral Graph Convolution Networks with Assistant Task Training

arXiv

引用

arXiv 2019年

作者： Xing, Xiaodan Li, Qingfeng Wei, Hao Zhang, Minqing Zhan, Yiqiang Zhou, Xiang Sean Xue, Zhong Shi, Feng United Imaging Intelligence Shanghai China Shanghai Advanced Research Institute Shanghai China School of Biomedical Engineering Southern Medical University Guangdong China School of Computer Science and Engineering Central South University Hunan China College of Software Engineering Southeast University Jiangsu China Shanghai Advanced Research Institute Shanghai China

Functional Connectivity (FC) matrices measure the regional interactions in the brain and have been widely used in neurological brain disease classification. However, a FC matrix is neither a natural image which contains shape and texture information, nor a vector of independent features, which renders the extracting of efficient features from matrices as a challenging problem. A brain network, also named as connectome, could form a graph structure naturally, the nodes of which are brain regions and the edges are interregional connectivity. Thus, in this study, we proposed novel graph convolutional networks (GCNs) to extract efficient disease-related features from FC matrices. Considering the time-dependent nature of brain activity, we computed dynamic FC matrices with sliding-windows and implemented a graph convolution based LSTM (long short time memory) layer to process dynamic graphs. Moreover, the demographics of patients were also used to guide the classification. However, unlike in conventional methods where personal information, i.e., gender and age, were added as extra inputs, we argue that this kind of approach may not actually improve the classification performance, for such personal information given in dataset was usually balanced distributed. In this paper, we proposed to utilize the demographic information as extra outputs and to share parameters among three networks predicting subject status, gender and age, which serve as assistant tasks. We tested the performance of the proposed architecture in ADNI II dataset to classify Alzheimer’s disease patients from normal controls. The classification accuracy, sensitivity and specificity reach 90.0%, 91.7% and 88.6% on ADNI II dataset. Copyright © 2019, The Authors. All rights reserved.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Best bone of acetabulum for cup component placement in Crowe types Ⅰto Ⅲ dysplastic hips:a computer simulation study

引用

Chinese Medical Journal 2019年第23期132卷 2820-2826页

作者： Lin-Li Zheng Yang-Yang Lin Xiao-Yan Zhang Qian-Hui Ling Wei-Ming Liao Pei-Hui Wu Department of Joint Surgery the First Affiliated HospitalSun Yat-sen UniversityGuangzhouGuangdong 510080China Department of Rehabilitation Medicine the Sixth Affiliated HospitalSun Yat-sen UniversityGuangzhouGuangdong 510655China College of Computer Science&Software Engineering Shenzhen UniversityShenzhenGuangdong 518060China Zhongshan Medical School Sun Yat-sen UniversityGuangzhouGuangdong 510080China

Background:During cup implantation,vertical height of the cup center(V-HCC)should be precisely controlled to achieve sufficient bone-cup coverage(BCC).Our study aimed to investigate the acetabular bone stock and the quantitative relationship between V-HCC and BCC in Crowe types Ⅰto Ⅲ ***:From November 2013 to March 2016,pelvic models of 51 patients(61 hips)with hip dysplasia were retrospectively reconstructed using a computer *** height and doom thickness were measured on the mid-acetabular coronal cross section.V-HCC was defined as the vertical distance of cup rotational center to the interteardrop line(ITL).In the cup implantation simulation,the cup was placed at the initial preset position,with a V-HCC of 15 mm,and moved proximally by 3-mm *** each level,the BCC was automatically calculated by *** of variance and Kruskal-Wallis test were used to compare the differences between ***:There were no significant between-group differences in maximum thickness of the acetabular doom;however,peak bone stock values were obtained at heights of 41^3±5.14 mm(Crowe typeⅠ),47.58±4.10 mm(Crowe type Ⅱ),and 55.78±3.64 mm(Crowe type Ⅲ)above the *** 15 mm of V-HCC,median BCC was 78%(75-83%)(Crowe type Ⅰ),74%(66-71%)(Crowe type Ⅱ),and 61%(57-68%)(Crowe type Ⅲ).To achieve 80%of the BCC,the median V-HCC was 16.27(15.00-16.93)mm,18.19(15.01-21.53)mm,and 24.13(21.02-28.70)mm for Crowe typesⅠ,Ⅱ,and Ⅲ hips,***:During acetabular reconstruction,slightly superior placement with V-HCC<25 mm retained sufficient bone coverage in CroweⅠto Ⅲ hips.

关键词： computer simulation Congenital dysplasia Hip Three-dimensional image Total-hip replacement

来源：评论

学校读者我要写书评

暂无评论

An Emergency Scenario Reconstruction System Based on ESMM Event Situation Model

引用

Wuhan University Journal of Natural sciences 2017年第6期22卷 482-488页

作者： LI Shuoming CHEN Lei LIU Yu CHEN Shihong National Engineering Research Center for Multimedia Software/School of Computer Wuhan University College of Computer Science and Technology Wuhan University of Science and Technology

In order to keep decision-makers better informed with emergencies, it is useful to retrieve the user-oriented disaster relevant event information in an aggregated results list through meta-search engine. However, emergent event is dynamic which makes it difficult to use fixed search word or word combinations. This paper proposes an event situation monitoring model（ESMM） event detection model, which realizes heuristic query word vector dynamic expanding by adopting emergency fuzzy scenario reasoning ontology cluster. Disaster event facet information automatic searching is discussed as an example in this paper. The experimental results show that the proposed method can increase accuracy and extra clues not supplied by commercial search engines, which can be used as a supplement information source for government and individuals.

关键词： meta search event situation monitoring model（ESMM） model ontology event scenario

来源：评论

学校读者我要写书评

暂无评论

MPMPA: A Mitigation and Prevention Model for Social engineering Based Phishing attacks on Facebook

MPMPA: A Mitigation and Prevention Model for Social Engineer...

引用

2018 IEEE International Conference on Big Data, Big Data 2018

作者： Jamil, Abid Asif, Kashif Ghulam, Zikra Nazir, Muhammad Kashif Mudassar Alam, Syed Ashraf, Rehan Department of Computer Science FAST - NUCES Chinnot Faisalabad Campus Pakistan Department of Software Engineering Foundation University Islamabad Pakistan Institute of Avionics and Aeronautics Air University Islamabad Pakistan Riphah College of Computing Riphah International University Faisalabad Pakistan Department of Computer Science National Textile University Faisalabad Pakistan

ISBN: (纸本)9781538650356

The exponential increase in usage of the Internet can be seen in the present era. The use of social media and cyber-physical systems is rapidly increasing. Social Media platforms like Facebook have become an indispensable part of our lives. A large number of reported Facebook attacks have been observed in past few years. This study presents a novel model to detect and prevent the SEBPA (Social engineering Based Phishing Attacks) on Facebook. Current research is first attempt to address SEBPA. For the Validation of proposed model four realistic scenarios have been deviced. The results suggest that the proposed model provides comprehensive coverage of nearly all occurrences of SEBPA. Moreover, the proposed model also prefigure the threatening situation to users with different colors during the validation process. © 2018 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

SD-Unet: A Structured Dropout U-Net for Retinal Vessel Segmentation

SD-Unet: A Structured Dropout U-Net for Retinal Vessel Segme...

引用

IEEE Symposium on Bioinformatics and Bioengineering (BIBE)

作者： Changlu Guo Márton Szemenyei Yang Pei Yugen Yi Wei Zhou Department of Control Engineering and Information Technology Budapest University of Technology and Economics Budapest Hungary Budapest University of Technology and Economics Budapest Hungary School of Software Jiangxi Normal University Nanchang China College of Computer Science Shenyang Aerospace University Shenyang China

ISBN: (纸本)9781728146188

At present, artificial visual diagnosis of fundus diseases has low manual reading efficiency and strong subjectivity, which easily causes false and missed detections. Automatic segmentation of retinal blood vessels in fundus images is very effective for early diagnosis of diseases such as the hypertension and diabetes. In this paper, we utilize the U-shaped structure to exploit the local features of the retinal vessels and perform retinal vessel segmentation in an end-to-end manner. Inspired by the recently DropBlock, we propose a new method called Structured Dropout U-Net (SD-Unet), which abandons the traditional dropout for convolutional layers, and applies the structured dropout to regularize U-Net. Compared to the state-of-the-art methods, we demonstrate the superior performance of the proposed approach.

关键词： Image segmentation Biomedical imaging Blood vessels Training Retinal vessels Diseases

来源：评论

学校读者我要写书评

暂无评论

False Trajectory Privacy Protection Scheme Based on Location Service

引用

IOP Conference Series: Materials science and engineering 2020年第7期768卷

作者： Chen Yang Wei Yan School of Software Dalian University of Foreign Languages Dalian China School of Computer Science and Technology Harbin Engineering University Harbin China Department of Information Liaoning Police College Dalian China School of Software Dalian University of Technology Dalian China

When the user applies for a location continuous query service, the user location privacy may be revealed due to different types of trajectory identification. A false trajectory privacy protection scheme is proposed, which uses the real-time generation of similar trajectories to protect the user's personal location privacy. The false location set is established by initial anonymous location selection, similar trajectory location calculation and generated location selection. The false location is used to form a false trajectory similar to the real location trajectory, which reduces the probability of attacker identification and achieves real-time trajectory anonymity based on location service. Finally, the effectiveness of the scheme is verified by performance analysis and experimental results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An efficient quantum blind digital signature scheme

引用

science China(Information sciences) 2017年第8期60卷 226-239页

作者： Hong LAI Mingxing LUO Josef PIEPRZYK Zhiguo QU Shudong LI Mehmet A.ORGUN College of Computer and Information Science Southwest University Information Security and National Computing Grid Laboratory School of Information Science and TechnologySouthwest Jiaotong University School of Electrical Engineering and Computer Science Queensland University of Technology Institute of Computer Science Polish Academy of Sciences School of Computer and Software Nanjing University of Information Science and Technology College of Mathematics and Information Science Shandong Technology and Business University School of Computer Science National University of Defense Technology Department of Computing Macquarie University Faculty of Information Technology Macau University of Science and Technology

Recently, many quantum digital signature(QDS) schemes have been proposed to authenticate the integration of a message. However, these quantum signature schemes just consider the situation for bit messages,and the signing-verifying of one-bit modality. So, their signature efficiency is very low. In this paper, we propose a scheme based on an application of Fibonacci-, Lucas-and Fibonacci-Lucas matrix coding to quantum digital signatures based on a recently proposed quantum key distribution(QKD) system. Our scheme can sign a large number of digital messages every time. Moreover, these special matrices provide a method to verify the integration of information received by the participants, to authenticate the identity of the participants, and to improve the efficiency for signing-verifying. Therefore, our signature scheme is more practical than the existing schemes.

关键词： blind quantum digital signature Fibonacci- Lucas-and Fibonacci-Lucas matrix coding digital messages signing-verifying modality

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：