检索结果-内蒙古大学图书馆

Journal of software 2011年第7期6卷 1257-1264页

作者： Zheng, Gang Yu, Tian Laboratory of Biological signal and Intelligent Processing Tianjin University of Technology Tianjin China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin University of Technology Tianjin China School of Computer and Communication Engineering Tianjin University of Technology Tianjin China Laboratory of Biological signal and Intelligent Processing Tianjin University of Technology Tianjin China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin University of Technology Tianjin China

A hybrid strategy has been proposed to reduce the wrong clustering on Ambulatory ECG (electrocardiogram). Since Ambulatory ECG is usually composed by 24 hours data, the number of individual ECG waveform can reach to 100,000, the request for accurate clustering result is highly required. The proposed strategy adopted some intelligent algorithms to solve the above problem. It clusters ECG waveform sample (selected from Ambulatory ECG) synchronously by Max-Min distance clustering algorithm, K-means algorithm and Simulated annealing algorithm first. And then, it adopted all three outputs from the above three algorithms as input on Back Propagation Artificial Neural Network (BP ANN). In the end, we got more accurate clustering result from the output of ANN. For testing the results, data of MIT/BIH arrhythmia database were used for experiments. After the controlled trial on MIT/BIH data, it can be safely concluded that the clustering result achieved by improved strategy can got more accurate than that by the traditional clustering algorithm. An average accuracy ratio is about 94.6%, 1.6% higher than k-means algorithm averagely and 1.3% higher than Simulated Annealing algorithm averagely. © 2011 ACADEMY PUBLISHER.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Privacy-Protected Facial Expression Recognition Augmented by High-Resolution Facial Images

Privacy-Protected Facial Expression Recognition Augmented by...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Cong Liang Shangfei Wang Xiaoping Chen Key Lab of Computing and Communication Software of Anhui Province University of Science and Technology of China Hefei China Anhui Robot Technology Standard Innovation Base University of Science and Technology of China Hefei China

Cloud-based expression recognition from high-resolution facial images may put the subjects’ privacy at risk. We identify two kinds of privacy leakage, the appearance leakage in which the visual appearances of subjects are disclosed and the identity-pattern leakage in which the identity information of subjects is dug out. To address both leakages, we propose privacy-protected facial expression recognition from low-resolution facial images with the help of high-resolution facial images. Specifically, to prevent appearance leakage, we propose to extract identity-invariant representations from downsampled images, from which the visually distinguishable appearances cannot be recovered. To prevent identity-pattern leakage, we propose to eliminate the identity information from the extracted representations by leveraging the disentangled representations of high-resolution images as privileged information. After training, our method can fully capture identity-invariant representations from downsampled images for expression recognition without the requirement of high-resolution samples. These privacy-protected representations can be safely transmitted through the Internet. Experimental results in different scenarios demonstrate that the proposed method protects privacy without significantly inhibiting facial expression recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Relation-Based Multi-Teacher Knowledge Distillation

Relation-Based Multi-Teacher Knowledge Distillation

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Yutian Wang Yi Xu School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu China Laboratory of Intelligent Collaborative Computing University of Electronic Science and Technology of China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Traditional knowledge distillation adopts one teacher model to instruct the training of a lightweight student model. To improve the performance of knowledge distillation, multi-teacher knowledge distillation utilizing multi-party knowledge to guide model learning was proposed. However, existing methods mainly use the summation-result of teacher models of a single sample to integrate multi-party knowledge directly, ignoring the relation knowledge among samples. In this paper, we propose a novel multi-teacher knowledge distillation method, which utilizes the data relation knowledge to allocate weights for teachers adaptively. To obtain the data relation knowledge, we design output-based strategy and feature-based strategy, which would help to allocate more weight for the teacher who has better learned the data relation knowledge. Extensive experiments have demonstrated the performance and efficiency of our proposed multi-teacher knowledge distillation method.

关键词： Training Adaptation models Neural networks

来源：评论

学校读者我要写书评

暂无评论

ECG signal feature selection for emotion recognition

引用

Telkomnika - Indonesian Journal of Electrical Engineering 2013年第3期11卷 1363-1370页

作者： Xun, Lichen Zheng, Gang Key Laboratory of Tianjin Intelligent Computing and Software Technologies Tianjin University of Technology Tianjin China Department of Computer and Communication Engineering Tianjin University of Technology Tianjin 300384 China

This paper aims to study the selection of features based on ECG in emotion recognition. In the process of features selection, we start from existing feature selection algorithm, and pay special attention to some of the intuitive value on ECG waveform as well. Through the use of ANOVA and heuristic search, we picked out the different features to distinguish joy and pleasure these two emotions, then we combine this with pathological analysis of ECG signals by the view of the medical experts to discuss the logic corresponding relation between ECG waveform and emotion distinguish. Through experiment, using the method in this paper we only picked out five features and reached 92% of accuracy rate in the recognition of joy and pleasure. © 2013 Universitas Ahmad Dahlan.

关键词： Electrocardiography

来源：评论

学校读者我要写书评

暂无评论

Realistic and real-time simulation of water

Realistic and real-time simulation of water

引用

International Conference on Computer Science and Information Technology (CSIT)

作者： Xian Wu Lan-fang Dong De-tang Lu University of Science and Technology of China Hefei Anhui CN Anhui Province Key Lab of Software in Computing and Communication Hefei China Institute of Engineering and Science Software University of Science and Technology Hefei China Institute of Engineering and Science Software University of Science and Technology of China Hefei China

A modified physics-based method is adopted to control the water waves, and the discretization of the governing 2D N-S equation is taken as the interaction between the variables in spaces, based on the idea of spreading though the neighboring regions. Meshing model of the water surface is based on a terrain LOD scheme called Tiled Quad-tree. With that the region of water surface can be extended unboundedly. A new controlling strategy is used, that the driving parameters of the water movement is taken separately, so it's easy to change parameters in real time while generating various water waves. Reflection and refraction of rays in water is simulated to improve the visual impression. It is shown that these methods can efficiently produce realistic movement of various water waves in real-time.

关键词： Computational modeling Navier-Stokes equations Animation Layout Automatic control Computer graphics Fluid dynamics Shape control communications technology Control system synthesis

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Genetic Algorithm for Vehicle Routing Problems with Dynamic Requests

A Hybrid Genetic Algorithm for Vehicle Routing Problems with...

引用

IEEE Symposium Series on Computational Intelligence

作者： Ruikang Yi Wenjian Luo Chenyang Bu Xin Lin Anhui Province Key Laboratory of Software Engineering in Computing and Communication University of Science and Technology of China Hefei Anhui China

In this paper, a hybrid Genetic Algorithm (GA) is proposed to solve the Vehicle Routing Problems with Dynamic Requests (VRPDR). The proposed hybrid GA primarily involves the following parts. First, a greedy split is introduced, which is fit for the VRPDR. Second, the adjacent-exchange-based local search is proposed. Third, a novel mutation operator, named the insert mutation, is proposed. The experimental results show that the proposed hybrid GA is effective.

关键词： Dynamic Vehicle Routing Genetic Algorithm Split Algorithm Local Search

来源：评论

学校读者我要写书评

暂无评论

A FUZZY IDENTIFICATION METHOD OF CONTOURS BASED ON CHAIN-CODE FEATURES

A FUZZY IDENTIFICATION METHOD OF CONTOURS BASED ON CHAIN-COD...

引用

第三届全国智能视觉监控学术会议

作者： Qi Dong Guangping Xu Liu Jing Yanbing Xue Hua Zhang Key Laboratory of Intelligence Computing and New Software Technology TianjinChina300384 Key Laboratory of Computer Vision and SystemMinistry of EducationTianjinChina300384 School of Computer and CommunicationTianjin University of TechnologyTianjinChina300384

ISBN: (纸本)9781457718342

The contour analysis and identification are the important aspects in visual surveillance *** paper proposes a fuzzy identification method of ***,according to the description of a contour based on the chain-code method,the proposed method utilizes the statistical features of contours including the chain-code entropy and chain-code space distribution entropies,from which the feature vector of a contour is ***,the method generates the contour pattern from some contour samples and uses the approaching principle to identify a *** our method integrates effectively multiple statistical feathers of its chain-code and employs a fuzzy pattern recognition technique,the experiments show quantitatively that it can achieve good results from various metrics.

关键词： Image Contour Fuzzy Pattern Chain Code Object Recognition

来源：评论

学校读者我要写书评

暂无评论

LinNet: Linear Network for Efficient Point Cloud Representation Learning 38

LinNet: Linear Network for Efficient Point Cloud Representat...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Deng, Hao Jing, Kunlei Cheng, Shengmei Liu, Cheng Ru, Jiawei Bo, Jiang Wang, Lin State-Province Joint Engineering and Research Center of Advanced Networking and Intelligent Information Services School of Information Science and Technology Northwest University China Shaanxi Key Laboratory of Higher Education Institution of Generative Artificial Intelligence and Mixed Reality China School of Software Engineering Xi'an Jiaotong University China Department of Computing The Hong Kong Polytechnic University Hong Kong

Point-based methods have made significant progress, but improving their scalability in large-scale 3D scenes is still a challenging problem. In this paper, we delve into the point-based method and develop a simpler, faster, stronger variant model, dubbed as LinNet. In particular, we first propose the disassembled set abstraction (DSA) module, which is more effective than the previous version of set abstraction. It achieves more efficient local aggregation by leveraging spatial anisotropy and channel anisotropy separately. Additionally, by mapping 3D point clouds onto 1D space-filling curves, we enable parallelization of downsampling and neighborhood queries on GPUs with linear complexity. LinNet, as a purely point-based method, outperforms most previous methods in both indoor and outdoor scenes without any extra attention, and sparse convolution but merely relying on a simple MLP. It achieves the mIoU of 73.7%, 81.4%, and 69.1% on the S3DIS Area5, NuScenes, and SemanticKITTI validation benchmarks, respectively, while speeding up almost 10x times over PointNeXt. Our work further reveals both the efficacy and efficiency potential of the vanilla point-based models in large-scale representation learning. Our code will be available at https://***/DengH293/LinNet. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Metamorphic Relations Prioritization And Selection Based on Test Adequacy Criteria

Metamorphic Relations Prioritization And Selection Based on ...

引用

Science and Technology Innovation (IAECST), International Academic Exchange Conference on

作者： Dafei Huang Yang Luo Meng Li School of Computing University of South China Hengyang China Engineering and Technology Research Center of Software Evaluation and Testing for Intellectual Equipment of Hunan Province Hengyang China CNNC Key Laboratory on High Trusted Computing Hengyang China

Metamorphic testing is the mainstream technique to solve oracle problems, and the effectiveness of metamorphic relations directly affects the testing cost. Most existing studies are summaries of engineering experience and lack theoretical discussions. A distance-based definition of metamorphic relation validity is provided, and a quantitative relationship model based on distance and test coverage is established to guide metamorphic relation selection and ranking. Experiments on four procedures and nearly 50 metamorphic relations show that, compared with random, coverage-based, and distance-based methods, this method has higher defect detection capability and can improve test effectiveness under the condition of using the same number of metamorphic relations. The number of metamorphic relations used in this method is lower than the number of programs under test and the cost of testing is lower while meeting the same defect detection level.

关键词： Technological innovation Costs Testing

来源：评论

学校读者我要写书评

暂无评论

A fuzzy identification method of contours based on chain-code features

A fuzzy identification method of contours based on chain-cod...

引用

2011 3rd Chinese Conference on Intelligent Visual Surveillance, IVS 2011

作者： Dong, Qi Xu, Guangping Jing, Liu Xue, Yanbing Zhang, Hua Key Laboratory of Intelligence Computing and New Software Technology Tianjin 300384 China Key Laboratory of Computer Vision and System Ministry of Education Tianjin 300384 China School of Computer and Communication Tianjin University of Technology Tianjin 300384 China

ISBN: (纸本)9781457718311

The contour analysis and identification are the important aspects in visual surveillance research. The paper proposes a fuzzy identification method of contours. First, according to the description of a contour based on the chain-code method, the proposed method utilizes the statistical features of contours including the chain-code entropy and chain-code space distribution entropies, from which the feature vector of a contour is composed. Then, the method generates the contour pattern from some contour samples and uses the approaching principle to identify a contour. Since our method integrates effectively multiple statistical feathers of its chain-code and employs a fuzzy pattern recognition technique, the experiments show quantitatively that it can achieve good results from various metrics. © 2011 IEEE.

关键词： Object recognition

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：