检索结果-内蒙古大学图书馆

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2013

作者： Bian, Yatao Li, Xiong Cao, Mingqi Liu, Yuncai Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai 200240 China

ISBN: (纸本)9783642409936

Parallel coordinate descent algorithms emerge with the growing demand of large-scale optimization. In general, previous algorithms are usually limited by their divergence under high degree of parallelism (DOP), or need data pre-process to avoid divergence. To better exploit parallelism, we propose a coordinate descent based parallel algorithm without needing of data pre-process, termed as Bundle Coordinate Descent Newton (BCDN), and apply it to large-scale 1-regularized logistic regression. BCDN first randomly partitions the feature set into Q non-overlapping subsets/bundles in a Gauss-Seidel manner, where each bundle contains P features. For each bundle, it finds the descent directions for the P features in parallel, and performs P-dimensional Armijo line search to obtain the stepsize. By theoretical analysis on global convergence, we show that BCDN is guaranteed to converge with a high DOP. Experimental evaluations over five public datasets show that BCDN can better exploit parallelism and outperforms state-of-the-art algorithms in speed, without losing testing accuracy. © 2013 Springer-Verlag.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Salient object segmentation based on automatic labeling

Salient object segmentation based on automatic labeling

引用

20th International Conference on Neural Information processing, ICONIP 2013

作者： Zhou, Lei Gong, Chen Li, YiJun Qiao, Yu Yang, Jie Kasabov, Nikola Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University China Auckland University of Technology New Zealand

ISBN: (纸本)9783642420504

This paper proposes an automatic salient object extraction framework. Firstly, the saliency model are developed by applying the low level color features and the boundary prior. The initial salient regions are extracted by adaptive thresholding. Multiple classifiers are trained with extracted initial region, which reflect color information of images or adopt label propagation. Then, the labels for segmentation are generated automatically via classifier composition. Finally, the conditional random field (CRF) model based on multi-feature fusion is applied for salient object segmentation. Empirical study reveals that the proposed algorithm achieves satisfying performance. © Springer-Verlag 2013.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Unsupervised segmentation using cluster ensembles

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2014年 8836卷 76-84页

作者： Zhang, Wei Yang, Jie Jia, Wenjing Kasabov, Nikola Jia, Zhenhong Zhou, Lei Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University China School of Computing and Communications University of Technology Sydney United States Knowledge Engineering and Discovery Research Institute Auckland University of Technology Auckland New Zealand School of Information Science and Engineering Xinjiang University Urumqi China

We propose a novel framework for automatic image segmentation. In this approach, a mixture of several over-segmentation methods are used to produce superpixels and then aggregation is achieved using a cluster ensemble method. Generated by different existing segmentation algorithms, superpixels can describe the manifold patterns of a natural image such as color space, smoothness and texture. We use them as the initial superpixels. Grouping cues which affect the performance of segmentation can also be captured. After the oversegmentation, the simultaneous collection of superpixels is expected to achieve synergistic effects and ensure the accuracy of the segmentation. For this purpose, cluster ensemble methods are used to process the initial segmentation results and produce the final result. Our method achieves significantly better performance on the Berkeley Segmentation Database compared to state-of-theart techniques. © Springer International Publishing Switzerland.

关键词： Superpixels

来源：评论

学校读者我要写书评

暂无评论

Salient object detection via color contrast and color distribution

Salient object detection via color contrast and color distri...

引用

11th Asian Conference on Computer Vision, ACCV 2012

作者： Fu, Keren Gong, Chen Yang, Jie Zhou, Yue Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Shanghai 200240 China

ISBN: (纸本)9783642373305

In this paper, we take the advantages of color contrast and color distribution to get high quality saliency maps. The overall procedure flow of our unified framework contains superpixel pre-segmentation, color contrast and color distribution computation, combination, final refinement and then object segmentation. During color contrast saliency computation, we combine two color systems and then introduce the using of distribution prior before saliency smoothing. It works to select correct color components. In addition, we propose a novel saliency smoothing procedure that is based on superpixel regions and is realized in color space. This processing step leads to total object being highlighted evenly, contributing to high quality color contrast saliency maps. Finally, a new refinement approach is utilized to eliminate artifacts and recover unconnected parts in the combined saliency maps. In visual comparison, our method produces higher quality saliency maps which stress out the total object meanwhile suppress background clutters. Both qualitative and quantitative experiments show our approach outperforms 8 state-of-the-art methods, achieving the highest precision rate 96% (3% improvement from the current highest), when evaluated via one of the most popular data sets [1]. Excellent content-aware image resizing also can be achieved with our saliency maps. © 2013 Springer-Verlag.

关键词： Superpixels

来源：评论

学校读者我要写书评

暂无评论

Algorithm Research and Real-time Simulation of Neural Network Sliding Mode Position Control

Algorithm Research and Real-time Simulation of Neural Networ...

引用

第25届中国控制与决策会议

作者： LI Wei ZHANG Yanyu GAO Yong CHAI Xiuli Institute of Image Processing and Pattern Recognition Henan University School of Computer and Information Engineering Henan University School of Automation Science and Engineering Beihang University

ISBN: (纸本)9781467355339

This paper presents a neural network sliding mode control algorithm for position control of modular robot. This method adopts BP neural network to approximate the functional relation between the sliding hyperplane and the exponential approximation rate. At the same time, the saturation function of sliding mode control algorithm is replaced by a hyperbolic tangent function to realize the boundary design method of the sliding mode control. The results of real-time simulation show that the algorithm proposed in this paper has the merits of fast response, strong robustness, and reducing the chattering of sliding mode control. This method solves the problems that conventional PID algorithm can’t solve under some circumstances, such as complicated environment, great load change, etc.

关键词： Robot control actuator sliding mode control neural network control

来源：评论

学校读者我要写书评

暂无评论

NeuCube(ST) for spatio-temporal data predictive modelling with a case study on ecological data

NeuCube(ST) for spatio-temporal data predictive modelling wi...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Enmei Tu Nikola Kasabov Muhaini Othman Yuxiao Li Susan Worner Jie Yang Zhenghong Jia The Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China The Knowledge Engineering and Discovery Research Institute Auckland University of Technology Auckland New Zealand Universiti Tun Hussein Onn Malaysia Lincoln University Lincoln NZ School of Information Science and Engineering Xinjiang University Urumqi China

Early event prediction challenges most of existing modeling methods especially when dealing with complex spatio-temporal data. In this paper we propose a new method for predictive data modelling based on a new development of the recently proposed NeuCube spiking neural network architecture, called here NeuCube (ST) . The NeuCube uses a Spiking Neural Network reservoir (SNNr) and dynamic evolving Spiking Neuron Network (deSNN) classifier. NeuCube (ST) is an integrated environment including data conversion into spike trains, input variable mapping, unsupervised learning in the SNNr, supervised classification learning, activity visualization and network structure analysis. A case study on a real world ecological data set is presented to demonstrate the validity of the proposed method.

关键词： Neurons Training Reservoirs Data models Predictive models Biological neural networks Biological system modeling

来源：评论

学校读者我要写书评

暂无评论

Optimization of level set segmentation with nonlinear coherent diffusion

引用

2013 2nd International Symposium on Materials Science and Engineering Technology, ISMSET 2013

作者： Gao, Yong Hui Wang, Sheng Zheng Yang, Jie Institute of Image Processing and Pattern Recognition University of Shanghai Jiaotong Shanghai 200240 China University of Shanghai Maritime Shanghai 201306 China

ISBN: (纸本)9783037858028

Level set method is convenient in image segmentation for the stabilization and veracity. Gaussian filter is usually taken as a preprocess to reduce the influence of weak edges due to noises, but the disadvantage is obvious: blur fine structures specially the important boundaries and lead to inaccurate segmentation result. This paper introduces a robust method which filters the images with a Nonlinear Coherent Diffusion (NCD) to accelerate the evolution of level set in a spatially varying manner. Experimental results show the performance of the proposed method in improving precision of segmentation. © (2013) Trans Tech Publications, Switzerland.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Region merging based segmentation with cellular automaton

Region merging based segmentation with cellular automaton

引用

2013 International Conference on Mechatronics, Applied Mechanics and Energy Engineering, MAMEE 2013

作者： Gao, Yong Hui Wang, Sheng Zheng Yang, Jie Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai 200240 China Shanghai Maritime University Shanghai 201306 China

ISBN: (纸本)9783037858325

Since fully automatic image segmentation on natural images is usually hard to provide guaranteed results, interactive scheme with a few simple user inputs becomes a good alternative. This paper presents a novel interactive method based on regional attacking and merging mechanism within a cellular automaton (CA) framework. With an attacking rule based on regions maximal similarity, the adjacent homogeneous regions that are initialized by pre-segmentation are automatically merged and labeled, the users only need to indicate the object and background regions with rough markers. The whole process needn't set any similarity threshold in advance and the desired contours are effectively extracted by labeling all the non-marker regions as either background or object. Extensive experiments are performed and the results show that the proposed scheme can reliably extract the object contours from the complex background. © (2013) Trans Tech Publications, Switzerland.

关键词： Cellular automata

来源：评论

学校读者我要写书评

暂无评论

An improved 3D edge surface tracking algorithm based on 3D fractional-order differentiation within confocal microscopy images

An improved 3D edge surface tracking algorithm based on 3D f...

引用

4th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2013

作者： Ma, Yu Zhang, Yanning Wang, Lisheng Institute of Computer Science Northwestern Polytechnical University Xi'an 710129 China Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai 200240 China

ISBN: (纸本)9783642420566

Fractional-order differentiation enhances the image nonlinearly, but only has been applied in the 2D image. The 2D fractional differentiation operator is extended to 3D and the 3D fractional differentiation discrete filtering masks are deduced. The 3D fractional differentiation is implemented to improve the traditional 3D image edge surface tracking algorithm in two aspects. Firstly, the 3D data fields of neuron slices are enhanced by the 3D fractional differentiation, ensures more detailed structures of edge surfaces with low contrast are extracted. Then integral-order gradient is modified by the fractional differentiation to get more 3D detailed structures. The proposed method has been applied to 3D confocal microscopy images, and more 3D detail structures of neuron are tracked compared to the traditional 3D edge surface tracking algorithm. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

3D model retrieval using density-based silhouette descriptor

3D model retrieval using density-based silhouette descriptor

引用

2nd International Conference on pattern recognition Applications and Methods, ICPRAM 2013

作者： Tang, Qi Yang, Xin Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai 200240 China

ISBN: (纸本)9789898565419

In this paper we present a new content-based retrieval descriptor, density-based silhouette descriptor (DBS). It characterizes a 3D object with multivariate probability functions of its 2D silhouette features. The new descriptor is computationally efficient and induces a permutation property that guarantees invariance at the matching stage. Also, it is insensitive to small shape perturbations and mesh resolution. The retrieval performance on several 3D databases shows that the DBS provides state-of-art discrimination over a broad and heterogeneous set of shape categories.

关键词： Probability density function

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：