检索结果-内蒙古大学图书馆

International Symposium on Knowledge Acquisition and Modeling, KAM

作者： Qing He Qun Wang Chang-ying Du Xu-dong Ma Zhong-zhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China Chinese Academy of Sciences Beijing China

The enlarging volumes of data resources produced in real world makes classification of very large scale data a challenging task. Therefore, parallel process of very large high dimensional data is very important. Hyper-Surface Classification (HSC) is approved to be an effective and efficient classification algorithm to handle two and three dimensional data. Though HSC can be extended to deal with high dimensional data with dimension reduction or ensemble techniques, it is not trivial to tackle high dimensional data directly. Inspired by the decision tree idea, an improvement of HSC is proposed to deal with high dimensional data directly in this work. Furthermore, we parallelize the improved HSC algorithm (PHSC) to handle large scale high dimensional data based on MapReduce framework, which is a current and powerful parallel programming technique used in many fields. Experimental results show that the parallel improved HSC algorithm not only can directly deal with high dimensional data, but also can handle large scale data set. Furthermore, the evaluation criterions of scaleup, speedup and sizeup validate its efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE WIC ACM International Conference on Web Intelligence (WI)

作者： Xu Han Zhongzhi Shi Wenjia Niu Kunrong Chen Xinghua Yang Chinese Academy of Sciences Beijing China The Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

With the rapid development of XML language which has good flexibility and interoperability, more and more log files of software running information are represented in XML format, especially for Web services. Fault diagnosis by analyzing semi-structured and XML like log files is becoming an important issue in this area. For most related learning methods, there is a basic assumption that training data should be in identical structure, which does not hold in many situations in practice. In order to learn from training data in different structures, we propose a similarity-based Bayesian learning approach for fault diagnosis in this paper. Our method is to first estimate similarity degrees of structural elements from different log files. Then the basic structure of combined Bayesian network (CBN) is constructed, and the similarity-based learning algorithm is used to compute probabilities in CBN. Finally, test log data can be classified into possible fault categories based on the generated CBN. Experimental results show our approach outperforms other learning approaches on those training datasets which have different structures.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image denoising based on Non-Local means and multi-scale dyadic wavelet transform

Image denoising based on Non-Local means and multi-scale dya...

引用

International Conference on Computer Science and Information Technology (CSIT)

作者： Gang Yu Yong Yin Hongjun Wang Zhi Liu Dengwang Li School of Information science and engineering Shandong University Lin'an China Key Laboratory of Intelligent Computing & Information Processing Ministry of Education Xiangtan China

A variety of wavelet transform methods have been introduced to remove noise from images. However, many of these algorithms remove the fine details and smooth the structures of the image when removing noise. The wavelet coefficient magnitude sum (WCMS) algorithm can preserve edges, but it is at the expense of removing noise. The Non-Local means algorithm can removing noise effective. But it tend to cause distortion ( eg white). Meanwhile, when the noise is large, the method is not so effective. In this paper, we propose an efficient denoising algorithm. we denoised the image with non-local means algorithm in the spatial domain and WCMS algorithm in wavelet domain, weighted, combined them and got the image that we want. The experiment shows that our algorithm can improve PSNR form 0.6 dB to 1.0 dB and the image boundary is more clearly.

关键词： Discrete wavelet transforms Image edge detection PSNR Histograms Noise reduction

来源：评论

学校读者我要写书评

暂无评论

Voting conditional random fields for multi-label image classification

Voting conditional random fields for multi-label image class...

引用

Congress on Image and signal processing, CISP

作者： Xishun Wang Xi Liu Zhiping Shi Zhongzhi Shi Hongjian Sui Chinese Academy of Sciences Beijing China The Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

In our real world, there usually exist several different objects in one image, which brings intractable challenges to the traditional pattern recognition methods to classify the images. In this paper, we introduce a Conditional Random Fields (CRFs) model to deal with the Multi-label Image Classification problem. Considering the correlations of the objects, a second-order CRFs is constructed to capture the semantic associations between labels. Different initial feature weights are set to introduce the voting techniques for a better performance. We evaluate our methods on MSRC dataset and demonstrate high precision, recall and F 1 measure, showing that our method is competitive.

关键词： Visualization Training Image classification Correlation Buildings Image segmentation Semantics

来源：评论

学校读者我要写书评

暂无评论

CBCT image denoising based on multi-scale wavelet transform

CBCT image denoising based on multi-scale wavelet transform

引用

International Conference on BioMedical Engineering and Informatics (BMEI)

作者： Yong Yin Gang Yu Hongjun Wang Zhi Liu Dengwang Li School of Information science and engineering Shandong University Jinan China Key Laboratory of Intelligent Computing Information Processing of Ministry of Education Xiangtan China

Cone-Beam Computed Tomography (CBCT) has always been in the forefront of medical image processing. The denoising as a image pre-processing, has a great affected on the image analysis and recognition. In this paper, a new algorithm for image denoising was proposed. By thresholding the interscale wavelet coefficient magnitude sum(WCMS) within a cone of influence (COI), the wavelet coefficients are classified into 2 categories: irregular coefficients, and edge-related and regular coefficients. They are processed by different ways. Meanwhile according to the projection image sequences characteristics in CBCT system, an effective noise variance estimated methods was proposed. The experiment shows that our algorithm can improve PSNR form 1.3dB to 2.6dB, and the image border is more clearly.

关键词： Noise Wavelet coefficients Image reconstruction Classification algorithms Noise reduction Image edge detection Image denoising

来源：评论

学校读者我要写书评

暂无评论

A HSC-based sample selection method for Support Vector Machine

A HSC-based sample selection method for Support Vector Machi...

引用

International Conference on Machine Learning and Cybernetics

作者： He, Qing Li, Ning Shi, Zhong-Zhi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Hebei University Baoding 071002 Hebei China Graduate University of Chinese Academy of Sciences Hebei University Baoding 071002 Hebei China Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei China

ISBN: (纸本)9781424465262

Support Vector Machine (SVM) is a classification technique of machine learning based on statistical learning theory. A quadratic optimization problem needs to be solved in the algorithm, and with the increase of the samples, the time complexity will also increase. So it is necessary to shrink training sets to reduce the time complexity. A sample selection method for SVM is proposed in this paper. It is inspired from the Hyper surface classification (HSC), which is a universal classification method based on Jordan Curve Theorem, and there is no need for mapping from lower-dimensional space to higher-dimensional space. The experiments show that the algorithm shrinks training sets keeping the accuracy for unseen vectors high. © 2010 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

A joint block diagonalization approach to convolutive blind source separation

A joint block diagonalization approach to convolutive blind ...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Xianfeng Xu Da-Zheng Feng Wei Xing Zheng National Key Laboratory of Radar Signal Processing Xidian University Xi'an China School of Computing and Mathematics University of Western Ontario Penrith South NSW Australia

This paper is concerned with blind separation of convolutive sources. The main idea is to make an explicit exploitation of block Toeplitz structure and block-inner diagonal structure in autocorrelation matrices of source signals at different time delays as well as of inherent relations among these matrices. With implementation of joint block diagonalization, a tri-quadratic cost function is introduced so that the mixture matrix can be extracted from a set of the correlation matrices of the observed vector sequence without pre-whitening. In this novel one-stage algorithm, every iteration step involves finding the closed solution to the corresponding least squares problem. Once the estimate of the mixing matrix is obtained, the source signals are retrieved by the classical least squares methods. The performance of the proposed algorithm is illustrated by simulation results.

关键词： Blind source separation signal processing algorithms Source separation Iterative algorithms Autocorrelation Time domain analysis Australia Delay effects Least squares methods Higher order statistics

来源：评论

学校读者我要写书评

暂无评论

Study of high-resolution range profile for monostatic MIMO radar based on SIAR

引用

Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics 2010年第4期32卷 734-739+763页

作者： Zhao, Guang-Hui Chen, Bai-Xiao Gao, Zhao-Zhao Intelligent Perception and Image Understanding Key Lab of Ministry of Education of China Xidian Univ. Xi'an 710071 China National Key Lab for Radar Signal Processing Xidian Univ. Xi'an 710071 China National Information Control Laboratory Chengdu 610036 China

Conventional pulse compression use a periodical echo of single receive antenna, which is modulated by a certain carrier-frequency, in other words, single spectrum is exploited. But for MIMO radar, as the multi-carrier-frequency signals are transmitted simultaneously, if the spectrum of the target echo after channel separation can be combined to form the whole band spectrum echo, the corresponding range resolution can improve several times as compared with the conventional method, and it will be more convenient for follow-up detection and tracking. Considering the difference between the frequency modulation band and the interval between the adjacent frequencies, the spectrum joint after channel separation will be overlapped or spaced. The methods of spectrum moving of each echo and the spectrum extrapolation with Root-MUSIC algorithm are proposed, by which high-resolution range profile of the target is obtained. Simulation results verify the validity of these methods.

关键词： Multiple signal classification

来源：评论

学校读者我要写书评

暂无评论

On Dynamic Weighting of Data in Clustering with K-Alpha Means

On Dynamic Weighting of Data in Clustering with K-Alpha Mean...

引用

International Conference on Pattern Recognition

作者： Si-Bao Chen Hai-Xian Wang Bin Luo Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Hefei Anhui China Key Laboratory of Child Development and Learning Science of Ministry of Education Research Center of Learning Science South-East University Nanjing Jiangsu China

ISBN: (纸本)9781424475421

Although many methods of refining initialization have appeared, the sensitivity of K-Means to initial centers is still an obstacle in applications. In this paper, we investigate a new class of clustering algorithm, K-Alpha Means (KAM), which is insensitive to the initial centers. With K-Harmonic Means as a special case, KAM dynamically weights data points during iteratively updating centers, which deemphasizes data points that are close to centers while emphasizes data points that are not close to any centers. Through replacing minimum operator in K-Means by alpha-mean operator, KAM significantly improves the clustering performances.

关键词： Clustering algorithms Heuristic algorithms Indexes Partitioning algorithms signal processing algorithms Euclidean distance Iris

来源：评论

学校读者我要写书评

暂无评论

Fast object detection using boosted co-occurrence histograms of oriented gradients

Fast object detection using boosted co-occurrence histograms...

引用

IEEE International Conference on Image processing

作者： Haoyu Ren Cher-Keng Heng Wei Zheng Luhong Liang Xilin Chen Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China Panasonic Singapore Laboratories Private Limited Singapore

Co-occurrence histograms of oriented gradients (CoHOG) are powerful descriptors in object detection. In this paper, we propose to utilize a very large pool of CoHOG features with variable-location and variable-size blocks to capture salient characteristics of the object structure. We consider a CoHOG feature as a block with a special pattern described by the offset. A boosting algorithm is further introduced to select the appropriate locations and offsets to construct an efficient and accurate cascade classifier. Experimental results on public datasets show that our approach simultaneously achieves high accuracy and fast speed on both pedestrian detection and car detection tasks.

关键词： Classification algorithms Training Support vector machines Feature extraction Accuracy Boosting Object detection

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：