检索结果-内蒙古大学图书馆

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

作者： Li, Qianwen Zhou, Yue Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9783642420566

As one of the most traditional tracking methods, particle filter has been improved in many previous tracking methods due to its non-Gaussian and non-linear distribution. Meanwhile, pure tracking methods cannot achieve good performance in complex tracking scenarios where there enormous deformation and occlusion occur. We present a combination of patch-based tracking and detecting methodology for visual tracking. In our tracking stage, a hierarchical patch-based histogram is used to describe the observation model, computed by an improved L 1 bin-ratio dissimilarity( L 1 -BRD) distance. While in the detecting stage, a patch-based binary feature is obtained through centersymmetric local binary pattern (CS-LBP) and then used to train a randomize fern forest. We combine the two parts collaboratively and experiments demonstrate that the proposed tracking framework outperforms the state-of-theart methods in challenging scenarios. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

Bundle CDN: A highly parallelized approach for large-scale 1-regularized logistic regression

Bundle CDN: A highly parallelized approach for large-scale 1...

引用

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

Speed estimation for scene objects using stereo visual odometry methods

Speed estimation for scene objects using stereo visual odome...

引用

2013 IEEE 9th International Conference on Intelligent Computer Communication and processing, ICCP 2013

作者： Golban, Catalin Nedevschi, Sergiu Image Processing and Pattern Recognition Group Computer Science Department Technical University of Cluj-Napoca Romania

ISBN: (纸本)9781479914937

This paper proposes a novel method to determine the speed of the surrounding vehicles in traffic scenarios. Relying on the video information obtained from a stereo camera mounted on a moving vehicle, we first determine the vehicle ego motion based on static scene features then we determine the relative motion between objects based on features situated on the moving objects. For robustness to false feature matches everything is plugged into a multi-RANSAC framework. The novelty of the method consist in the fact that the relative motion between the objects can be determined with the same algorithm that was previously used for ego motion estimation, the only difference consisting in the geometric constraints that are imposed to the subset of point features considered for inliers set detection and evaluation. Also, the proposed method does not rely on the fact that objects are detected previously and it does not detect the objects. © 2013 IEEE.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

Improved Predictive Personalized Modelling with the use of Spiking Neural Network System and a Case Study on Stroke Occurrences Data

Improved Predictive Personalized Modelling with the use of S...

引用

International Joint Conference on Neural Networks

作者： Muhaini Othman Nikola Kasabov Enmei Tu Valery Feigin Rita Krishnamurthi Zhengguang Hou Yixiong Chen Jin Hu Knowledge Engineering and Discovery Research Institute (KEDRI) Auckland University of Technology Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University The National Institute for Stroke and Applied Neurosciences (NISAN) AUT University Institute of Automation Chinese Academy of Sciences (CASIA)

ISBN: (纸本)9781479914821

This paper is a continuation of previous published work by the same authors on Personalized Modelling and Evolving Spiking Neural Network Reservoir architecture (PMeSNNr). The focus is on improvement of predictive modeling methods for the stroke occurrences case study utilizing an enhanced NeuCube architecture. The adaptability of the new architecture leads towards understanding feature correlations that affect the outcome of the study and extracts new knowledge from hidden patterns that reside within the associations. Through this new method, estimation of the earliest time point for stroke prediction is possible. This study also highlighted the improvement from designing a new experimental dataset compared to previous experiments. Comparative experiments were also carried out using conventional machine learning algorithms such as kNN, wkNN, SVM and MLP to prove that our approach can result in much better accuracy level.

关键词： Neurons Data models Reservoirs Accuracy Brain modeling Computer architecture Predictive models

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

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

Kernel based weighted group sparse representation classifier

Kernel based weighted group sparse representation classifier

引用

15th International Conference on Human-Computer Interaction, HCI International 2013

作者： Xu, Bingxin Guo, Ping Chen, C. L. Philip Image Processing and Pattern Recognition Laboratory Beijing Normal University Beijing China Faculty of Science and Technology University of Macau China

ISBN: (纸本)9783642393419

Sparse representation classification (SRC) is a new framework for classification and has been successfully applied to face recognition. However, SRC can not well classify the data when they are in the overlap feature space. In addition, SRC treats different samples equally and ignores the cooperation among samples belong to the same class. In this paper, a kernel based weighted group sparse classifier (KWGSC) is proposed. Kernel trick is not only used for mapping the original feature space into a high dimensional feature space, but also as a measure to select members of each group. The weight reflects the importance degree of training samples in different group. Substantial experiments on benchmark databases have been conducted to investigate the performance of proposed method in image classification. The experimental results demonstrate that the proposed KWGSC approach has a higher classification accuracy than that of SRC and other modified sparse representation classification. © 2013 Springer-Verlag.

关键词： Classification (of information)

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

Optimal classification of epileptic EEG signals using neural networks and harmony search methods

Journal of Software

引用

Journal of Software 2014年第1期9卷 230-239页

作者： Gao, Xiao-Zhi Wang, Jing Tanskanen, Jarno M. A. Bie, Rongfang Wang, Xiaolei Guo, Ping Zenger, Kai College of Information Engineering Shanghai Maritime University China Department of Automation and Systems Technology Aalto University School of Electrical Engineering Finland Laboratory of Image Processing and Pattern Recognition Beijing Normal University China Department of Biomedical Engineering Tampere University of Technology Finland School of Foundational Education Peking University Health Science Center China

In this paper, the Harmony Search (HS)-aided BP neural networks are used for the classification of the epileptic electroencephalogram (EEG) signals. It is well known that the gradient descent-based learning method can result in local optima in the training of BP neural networks, which may significantly affect their approximation performances. Three HS methods, the original version and two new variations recently proposed by the authors of the present paper, are applied here to optimize the weights in the BP neural networks for the classification of the epileptic EEG signals. Simulations have demonstrated that the classification accuracy of the BP neural networks can be remarkably improved by the HS method-based training. © 2014 ACADEMY PUBLISHER.

关键词： Optimization

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

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：