检索结果-内蒙古大学图书馆

IEEE International Conference on Image processing

作者： Jianchao Li Dan Wang Canxiang Yan Shiguang Shan Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS University of Chinese Academy of Sciences Institute of Deep Learning Baidu Inc.

ISBN: (纸本)9781479983407

Object segmentation has constantly received much attention due to its fundamental role in scene understanding. Traditional methods formulate it as a structured prediction problem, represented by graphical models (GMs). However, most GMs have difficulties in balancing the effectiveness of context modeling and efficiency of model inference. In this paper, we model the contexts implicitly using the deep convolutional neural network (DCNN). Specifically, we reformulate object segmentation as a regression problem and train a deep network end-to-end to learn the nonlinear mapping from the image to the object mask. The large receptive field of the network incorporates wide contexts to update the network parameters, giving an implicit context model. Moreover, the deep architecture is favorable for modeling nonlinearity. The inference of our method is quite efficient, involving only a simple feed-forward pass. Extensive experiments on public datasets demonstrate the advantages of our method.

关键词： Object segmentation Context modeling Deep convolutional neural networks Regression Object segmentation Context modeling Regression (Psychology) network parameters Regression Regression

来源：评论

学校读者我要写书评

暂无评论

Fusing PLSA model and Markov random fields for automatic image annotation

引用

High Technology Letters 2014年第4期20卷 409-414页

作者：田东平 Zhao Xiaofei Shi Zhongzhi Institute of Computer Software Baoji University of Arts and Sciences Key Laboratory of Intelligent Information Processing Institute of Computing TechnologyChinese Academy of Sciences

A novel image auto-annotation method is presented based on probabilistic latent semantic analysis(PLSA) model and multiple Markov random fields(MRF).A PLSA model with asymmetric modalities is first constructed to estimate the joint probability between images and semantic concepts,then a subgraph is extracted served as the corresponding structure of Markov random fields and inference over it is performed by the iterative conditional modes so as to capture the final annotation for the *** novelty of our method mainly lies in two aspects:exploiting PLSA to estimate the joint probability between images and semantic concepts as well as multiple MRF to further explore the semantic context among keywords for accurate image *** demonstrate the effectiveness of this approach,an experiment on the Corel5 k dataset is conducted and its results are compared favorably with the current state-of-the-art approaches.

关键词： automatic image annotation probabilistic latent semantic analysis （PLSA） expectation maximization Markov random fields （MRF） image retrieval

来源：评论

学校读者我要写书评

暂无评论

Portrait image segmentation based on improved grabcut algorithm 6th

Portrait image segmentation based on improved grabcut algori...

引用

6th International Conference on Swarm Intelligence, ICSI 2015 held in conjunction with the 2nd BRICS Congress on Computational Intelligence, CCI 2015

作者： Li, Shuai Zheng, Xiaohui Chen, Xianjun Zhan, Yongsong Key Laboratory of Cloud Computing and Complex System Guilin University of Electronic Technology Guilin Guangxi541004 China Guangxi Experiment Center of Information Science Guilin University of Electronic Technology Guilin Guangxi541004 China Key Laboratory of Intelligent Processing of Computer Image and Graphics Guilin University of Electronic Technology Guilin Guangxi541004 China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin Guangxi541004 China Information Engineering School Haikou College of Economics Haikou Hainan571127

ISBN: (纸本)9783319204680

Traditional computer portrait caricature system mainly take the method that exaggerate and deform real images directly, that lead the facial image background also been deformed when exaggerate facial image. If in pretreatment stage, we segmented the characters and background of the input image, and then do the subsequent processing, the problem may be solved. But for better portrait caricature effects, we need an excellent segmentation algorithm. So, we propose an improved Grabcut image segmentation algorithm and use it to extract the prospect character image for exaggeration and deformation. In practical application, we separate deform and exaggerate the foreground characters image with TPS method, then fuse it with the original or new background picture, get the final image. Application proves, the method solves the background deformation problem well, and improves the quality and rate of image segmentation, caricature synthesis effect reality and natural. © Springer International Publishing Switzerland 2015.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Unscented Kalman consensus filters based on Rauch-Tung-Striebel Smoother

引用

Journal of Computational information Systems 2014年第10期10卷 4159-4167页

作者： Zou, Juan Pei, Tingrui Zhou, Yan Long, Jiachuan College of Information Engineering Xiangtan University Xiangtan 411105 China Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Xiangtan 411105 China

One of the fundamental problems in sensor networks is the target's state estimation. The Unscented Kalman consensus filter (C-UKF) makes use of the consensus protocol and applies it to an optimal filter of nonlinear state-space models. It does this in order to acquire a consensus of n sensor Kalman filters' measurements, but the estimates of C-UKF are dispersed for a relatively long period. In this study, a new Unscented Kalman consensus filter based on Rauch-Tung-Striebel Smoother (C-URTS) is proposed. This new algorithm and Unscented Kalman consensus filter (C-UKF) differ in their reliance on a traditional Unscented Kalman filter. Unscented Kalman consensus filters are used to apply state estimates for the initial smooth value of URTS. The C-URTS uses a neighbor's information to reach a consensus, and then runs a backward smoother with the obtained Kalman consensus filters' value for better accuracy. Simulation results are provided to demonstrate the effectiveness of C-URTS in sensor networks. 1553-9105/Copyright © 2014 Binary information Press.

关键词： Sensor networks

来源：评论

学校读者我要写书评

暂无评论

On the encoding complexity of systematic polar codes

On the encoding complexity of systematic polar codes

引用

IEEE International SOC Conference

作者： Liping Li Wenyi Zhang Key Laboratory of Intelligent Computing and Signal Processing of the Ministry of Education of China Anhui University China Chinese Academy of Sciences and Department of Electronic Engineering and Information Sciences University of Science and Technology of China

Systematic polar codes are proposed by Arikan and are shown to have better BER performance than non-systematic polar codes. From a recursive decomposition of the generator matrix of polar codes, Arikan showed that the encoding complexity of systematic polar codes is also O(N log N) where N is the code block length. But the recursive process involves some additional calculations in transforming the problem instances back and forth. In this paper, by using the sparsity property of the generator matrix, we propose an encoding process which has the same complexity as non-systematic polar codes in the presence of an additional memory array. Without the additional memory elements, the number of additions of the proposed encoding process increases compared with non-systematic polar codes. We also provide an analysis to quantify this additional increase of the complexity.

关键词： Encoding Systematics Complexity theory Decoding Generators Bit error rate Electronic mail

来源：评论

学校读者我要写书评

暂无评论

A saliency detection method based on global contrast

International Journal of Signal Processing, Image Processing...

引用

International Journal of Signal processing, Image processing and Pattern Recognition 2015年第7期8卷 111-122页

作者： Yu, Chun-Yan Zhang, Wei-Shi Wang, Chun-Li Institute of Information and Science Technology Dalian Maritime University China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University No.3 Feixi Road Hefei China

To highlight the saliency object clearly from the foreground, we propose a saliency detection method based on global contrast with cluster. Due to the fact that background pixels usually have similar patches, we use cluster analysis to merge the background regions. By using mean shift filter, the background pixels with similar color level are clustered and the saliency calculation can be decreased a lot. In the method, we use the contrast of color feature with all the other pixels to compute the saliency map. A weight coefficient is utilized to improve the detection accuracy in global contrast differences evaluation. The results of extensive experiments on public dataset show that our method perform well and can highlight the salient object clearly against the other five state-of-the-art methods. Besides, we demonstrate that the applications in image segmentation and fusion with our saliency map can get satisfactory results. © 2015 SERSC.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

An Interrupt Scheduling Mechanism of Virtual Network Interface Card for Xen-Based System

An Interrupt Scheduling Mechanism of Virtual Network Interfa...

引用

2015 5th International Conference on Computer Sciences and Automation Engineering(ICCSAE 2015)

作者： Liuzheng Lu Runze Wan Jinrong He School of Computer Hubei University of Education Co-Innovation Centre of Elementary Education Information Technology Services of Hubei Science Computing and Intelligent Information Processing of Guangxi Higher Education Key Laboratory Guangxi Teachers Education University

As one of typical paravirtualization hypervisors, Xen has received widespread attentions especially its scaling capability under some kinds of workload. In this paper, we focus on the problem that the most CPU resources are occupied by frequent interrupt from the NIC（Network Interface Card） and it will cause bottleneck of the system for Xen. To alleviate this problem, this paper proposes an adaptive interrupt latency scheduling mechanism based on XEN, which use the polling or interrupt method in accordance with the queue length of virtual buffer without supplementing any additional processing unit. Also, the method can guarantee different quality of service to some extent by means of the definition of the two types of priority virtual buffers. Simulation results show that the mechanism can reduce CPU overhead significantly and improve system performance effectively.

关键词： virtualization interrupt latency virtual buffer

来源：评论

学校读者我要写书评

暂无评论

An efficient topology-based algorithm for transient analysis of power grid 11th

An efficient topology-based algorithm for transient analysis...

引用

11th International Conference on intelligent computing, ICIC 2015

作者： Yang, Lan Wang, Jingbin Azevedo, Lorenzo Wang, Jim Jing-Yan School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing400067 China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University Tianjin300072 China National Time Service Center Chinese Academy of Sciences Xi’an710600 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou215006 China Computer Science Department University of Säo Paulo Säo Paulo SP05508-070 Brazil Thuwal23955-6900 Saudi Arabia

ISBN: (纸本)9783319221793

In the design flow of integrated circuits, chip-level verification is an important step that sanity checks the performance is as expected. Power grid verification is one of the most expensive and time-consuming steps of chip-level verification, due to its extremely large size. Efficient power grid analysis technology is highly demanded as it saves computing resources and enables faster iteration. In this paper, a topology-base power grid transient analysis algorithm is proposed. Nodal analysis is adopted to analyze the topology which is mathematically equivalent to iteratively solving a positive semi-definite linear equation. The convergence of the method is proved. © Springer International Publishing Switzerland 2015.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Fast assignment reduction in inconsistent incomplete decision systems

引用

Journal of Systems Engineering and Electronics 2014年第1期25卷 83-94页

作者： Min Li Shaobo Deng Shengzhong Feng Jianping Fan Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Nanchang Institute of Technology Graduate School of Chinese Academy of Sciences Key Laboratory of Intelligent Information Processing Institute of Computing TechnologyChinese Academy of Sciences

This paper focuses on fast algorithm for computing the assignment reduct in inconsistent incomplete decision systems. It is quite inconvenient to judge the assignment reduct directly ac-cording to its definition. We propose the judgment theorem for the assignment reduct in the inconsistent incomplete decision system, which greatly simplifies judging this type reduct. On such basis, we derive a novel attribute significance measure and construct the fast assignment reduction algorithm （F-ARA）, intended for com-puting the assignment reduct in inconsistent incomplete decision systems. Final y, we make a comparison between F-ARA and the discernibility matrix-based method by experiments on 13 Univer-sity of California at Irvine （UCI） datasets, and the experimental results prove that F-ARA is efficient and feasible.

关键词： assignment reduction upper approximation reduction,inconsistent incomplete decision system rough set.

来源：评论

学校读者我要写书评

暂无评论

Dynamic clustering algorithm in mobile wireless sensor network

引用

Journal of information and Computational Science 2015年第13期12卷 4897-4907页

作者： Wang, Bingting Liang, Hui Lin, Qibin Zhang, Yangyi Wang, Yujie Hong, Peilin Hu, Yanjun Department of Electronic and Electrical Engineering Chuzhou University Chuzhou China Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei China

In this paper, we analyzed the blindness of traditional clustering algorithms, which select cluster head based on residual energy. Then we proposed the Dynamic Clustering Algorithm (DCA) in Mobile Wireless Sensor Network, which takes into consideration residual buffer level when selecting cluster heads and allots time slots dynamically base on the occupied buffer level. We analyzed the energy model of the proposed DCA algorithm and calculated the optimal distance between cluster heads. Simulation results show that our DCA clustering algorithm has good performances in terms of residual energy, data transmission and network lifetime by comparing it with LEACH and LEACH-CI. ©, 2015, Journal of information and Computational Science. All right reserved.

关键词： Clustering algorithms

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：