检索结果-内蒙古大学图书馆

Web image clustering based-on Multi-Instance

Web image clustering based-on Multi-Instance

2008 IEEE/WIC/ACM International Conference on Web Intelligence and intelligent Agent Technology - Workshops, WI-IAT Workshops 2008

作者： Lu, Jing Ma, Shaoping Zhang, Min State Key Lab. of Intelligent Tech. and Systems Tsinghua National Lab. for Information Science and Tech. Department of Computer Science and Tech. Tsinghua Uni Beijing 100084 China

ISBN: (纸本)9780769534961

In image retrieval and annotation, Multi-Instance Learning has been studied actively. Most of the methods solve the MIL problem in a supervised way. In this paper, we proposed two unsupervised frameworks for clustering multi-instance objects based on Expectation Maximization (EM) and iterative heuristic optimization respectively. For each framework, we introduced three new algorithms of finding users' interests on specific web images without any manual lab.led data. And comparative studies have shown the effectiveness of the proposed algorithms. © 2008 IEEE.

关键词： Maximum principle

来源：评论

学校读者我要写书评

暂无评论

Efficient learning of robust quadruped bounding usingpretrained neural networks

引用

IET Cyber-systems and Robotics 2022年第4期4卷 331-338页

作者： Zhicheng Wang Anqiao Li Yixiao Zheng Anhuan Xie Zhibin Li Jun Wu Qiuguo Zhu Institute of Cyber-Systems and Control Zhejiang UniversityZhejiangChina Intelligent Robot Research Center Zhejiang LabZhejiangChina Department of Computer Science University College LondonLondonUK State Key Laboratory of Industrial Control Technology Zhejiang UniversityZhejiangChina

Bounding is one of the important gaits in quadrupedal locomotion for negotiating *** authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation in dynamic body *** authors first pretrained the neural network(NN)based on data from a robot operated by conventional model-based controllers,and then further optimised the pretrained NN via deep reinforcement learning(DRL).In particular,the authors designed a reward function considering contact points and phases to enforce the gait symmetry and periodicity,which improved the bounding *** NN-based feedback controller was learned in the simulation and directly deployed on the real quadruped robot Jueying Mini successfully.A variety of environments are presented both indoors and outdoors with the authors’*** authors’approach shows efficient computing and good locomotion results by the Jueying Mini quadrupedal robot bounding over uneven *** cover image is based on the Research Article Efficient learning of robust quadruped bounding using pretrained neural networks by Zhicheng Wang et al.,https://***/10.1049/csy2.12062.

关键词： legged locomotion reinforcement learning robot learning

来源：评论

学校读者我要写书评

暂无评论

Fast double-parallel image processing based on FPGA

Fast double-parallel image processing based on FPGA

引用

IEEE International Conference on Vehicular Electronics and Safety

作者： Li, Ye Yao, Qingming Tian, Bin Xu, Wencong State Key Lab. of Intelligent Control and Management of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China School of Mechanical Electrical and Information Engineering Shandong University Weihai 264209 China

ISBN: (纸本)9781457705762

Due to FPGA's flexibility and parallelism, it is popular for accelerating image processing. In this paper, a double-parallel architecture based on FPGA has been exploited to speed up median filter and edge detection tasks, which are essential steps during image processing. The double-parallel scheme includes an image-level parallel and an operation-level parallel. The image-level parallel is a high-level parallel which divides one image into different parts and processes them concurrently. The operation-level parallel, which is embedded in each image-level parallel thread, fully explores every parallel part inside the concrete algorithms. The corresponding design is based on a DE2 Development Board which contains a CYCLONE II FPGA device. Meanwhile, the same task has also been implemented on PC and DSP for performance comparison. Despite the fact that operating frequencies of used PC and DSP are much higher than FPGA's, FPGA costs less time per computed image than both of them. By taking advantage of the double-parallel technique, the speed/frequency ratio of FPGA is 202 times faster than PC and 147 times faster than DSP. Finally, a detailed discussion about different platforms is conducted, which analyzes advantages and disadvantages of used computing platforms. This paper reveals that the proposed double-parallel scheme can dramatically speed up image processing methods even on a low-cost FPGA platform with low frequency and limited resources, which is very meaningful for practical applications. © 2011 IEEE.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Photoelectric intelligent Detection Platform Based on Multifunctional Edge Computing Equipment 43

Photoelectric Intelligent Detection Platform Based on Multif...

引用

43rd Chinese control Conference, CCC 2024

作者： Xu, Qianjing Chen, Wenjie Xue, Zhijun State Key Lab of Intelligent Control and Decision of Complex Systems Beijing10081 China Beijing Institute of Technology School of Automation Beijing100081 China

ISBN: (纸本)9789887581581

In order to study to design a photoelectric intelligent platform based on multifunctional edge computing equipment by using photoelectric detection equipment TC505C, RK-Series Development Kits computing platform and Nvidia Jetson Series. The photoelectric intelligent platform can be used to run multiple target detection algorithms after design and implementation. Completed as follows. First of all, after the first stage of the photoelectric intelligent platform is completed, the host computer can establish communication with the photoelectric detection equipment, carry out automatic or manual servo control on the photoelectric detection equipment, switch its visible and infrared field of view and view it in real time through the monitor, and realize simple tracking function through the image processing box. Secondly, after the second stage of the photoelectric intelligent platform is built, under the premise of ensuring the normal availab.lity of servo function, a variety of target detection algorithms can be debugged on multifunctional edge computing equipment. Overall, this paper uses the modular design concept to design, debug and implement the system in stages. Pay attention to the synchronous debugging of hardware and software functions of each module to ensure the stability of each module. Finally completed the design goals of each stage. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Program debugging

来源：评论

学校读者我要写书评

暂无评论

Model set optimization method for complex plant multi-model control

引用

IMACS Multiconference on "Computational Engineering in systems Applications", CESA

作者： Jiansheng, Xu Xiong, Hou Yongji, Wang IEEE Department of Control Science and Engineering Huazhong University of Science and Technology Wuhan 430074 China State Key Lab. of National Defense for Aerospace Intelligent Control Beijing 100854 China

ISBN: (纸本)7302139229

In this paper we present a novel model set optimization method that is applicable to the complex systems with large parameters variations and multiple operating modes. The proposed method is based on the multi-model approach to improve the system performance of a complex control system of linear or nonlinear characteristics when it operates at various operating conditions. The multi-model control scheme depends on the multiple representations of a process using different models that generate the control signal needed to make the system follow a prescribed desired trajectory. It uses a set of models-combining the offline identified models and the dynamic model bank with the iterative learning, which guides the adaptation process. The offline models are obtained from the physics or chemistry principles. Dynamic model bank summarizes the parameters of the models that successfully approximate the plant. The model bank can be automatically created and updated and does not call for an initial set of models. By virtue of the iterative learning, we expect to make the models set consist the models which best represented the character of the plant. It uses a soft switching mechanism that provides a smooth transition from an interpolate to a pure hard switching scheme between the models in the model set We also demonstrate the advantage of using this approach on an examples considering control of systems with large parameter variations. Simulation results show that the method proposed in this paper has obvious advantage over the only dynamic model bank, and illustrate the potential of the algorithm of the developed method.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Optimal budget allocation across search advertising markets

Optimal budget allocation across search advertising markets

引用

21st Workshop on Information Technologies and systems, WITS 2011

作者： Yang, Yanwu Zhang, Jie Liu, Baiyu Zeng, Daniel State Key Lab. of Intelligent Control and Management of Complex Systems Chinese Academy of Sciences Beijing 100190 China School of Mathematics and Physics University of Science and Technology Beijing China Department of Management Information Systems University of Arizona United States

How to rationally allocate the limited advertising budget is a critical issue in search auctions. However, due to the heterogeneousness of major search markets in terms of auction mechanisms, ranking algorithms and advertising structures, it is becoming increasingly difficult for an advertiser to manipulate advertising budget simultaneously across several search markets. In this paper, we establish a novel optimal budget allocation model across search advertising markets, under a finite time horizon. By considering distinctive features of search auctions, we introduce the quality score q and the dynamic advertising effort u to extend the advertising response function to fit budget decision scenarios. We also provide a feasible solution to our model and study some desirable properties: (a) the marginal return is non-increasing with respect to the advertising budget;(b) the optimal budget solution satisfies the condition that the advertising effort u is positively proportional to the product of the change of accumulated revenue in a market ∂V, the change of market share ∂ θ and the advertising elasticity α. Computational experiments are made to evaluate our model and identified properties. Experimental results show that the advertiser with increasing advertising elasticity is suggested to invest more budget in the late stages, but the advertiser with decreasing advertising elasticity should invest more budget in the initial stage, in order to maximize net profits.

关键词： Marketing

来源：评论

学校读者我要写书评

暂无评论

Efficient Mobile Deployment of Lightweight CNN-Transformer Model for Crop Disease Diagnosis in Smart Agriculture 4

Efficient Mobile Deployment of Lightweight CNN-Transformer M...

引用

4th Asia Conference on Information Engineering, ACIE 2024

作者： Lv, Zirong China University of Geosciences Hubei Key Lab. of Adv. Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education School of Automation Wuhan China

ISBN: (纸本)9798350371925

Crop disease diagnosis is of great significance for crop yield and agricultural production. Deep learning methods have become a major research direction for solving crop disease diagnosis. However, recent studies have shown that deploying complex neural network models on mobile devices poses a significant challenge to mobile hardware performance. To address this issue, this paper proposes a lightweight network model based on the Transformer and deploys it on mobile devices. Compared to other lightweight network models, our approach significantly reduces computational complexity using Transformer modules and exhibits a particular advantage in recognition accuracy. It achieves a recognition rate of 97.76% for ten categories of tomato diseases, while the model size is only 3.86 MB. Furthermore, we deploy the model on embedded devices and apply it to tomato crop recognition, enabling farmers to monitor crop conditions in real-time and realizing smart agriculture. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A fast algorithm of address lines extraction on complex Chinese mail pieces

A fast algorithm of address lines extraction on complex Chin...

引用

3rd IASTED International Conference on Signal Processing, Pattern Recognition, and Applications

作者： Liu, Tong Ding, Xiaoqing Fu, Qiang Ren, Zheng State Key Lab. of Intelligent Technology and Systems Tsinghua University Beijing China Siemens AG Buecklestrasse 1-5 D-78467 Konstanz Germany

ISBN: (纸本)0889865477

A fast and efficient method is presented to extract address lines on both machine printed and handwritten Chinese mail envelopes. The algorithm is based on a bottom-up approach. First, we select out text blocks from connected components (CCs) and immediately group the text blocks into the initial lines. Then, the average text block features are computed to validate the initial text lines and guide an iterative split and merge process. Lines are split by merging the text CCs in detail according to criteria for similarity and consistency of neighborhood text blocks. Particularly, some non-text blocks within the lines are recovered if they are similar with other text blocks. A skew detection and, accordingly, deskew step is followed. We have tested the performance of our methods on a large mail sample test deck with different categories of envelopes, and an obvious improvement both on accuracy and on computation time could be achieved compared to our previous system.

关键词： Algorithms

来源：评论

学校读者我要写书评

暂无评论

Dynamic budget adjustment in search auctions

Dynamic budget adjustment in search auctions

引用

21st Workshop on Information Technologies and systems, WITS 2011

作者： Zhang, Jie Yang, Yanwu Qin, Rui Zeng, Daniel Li, Xin State Key Lab. of Intelligent Control and Management of Complex Systems Chinese Academy of Sciences Beijing 100190 China Department of Management Information Systems University of Arizona United States Department of Information Systems City University of Hong Kong Kowloon Tong Hong Kong

With serious advertising budget constraints, advertisers have to adjust their daily budget according to the performance of advertisements in real time. Thus we can leave precious budgets to better opportunities in the future, and avoid the surge of ineffective clicks for unnecessary costs. However, advertisers usually have no sufficient knowledge and time for real-time advertising operations in search auctions. We formulate the budget adjustment problem as a state-action decision process in the reinforcement learning (RL) framework. Considering dynamics of marketing environments and some distinctive features of search auctions, we extend continuous reinforcement learning to fit the budget decision scenarios. The market utility is defined as discounted total clicks to get during the remaining period of an advertising schedule. We conduct experiments to validate and evaluate our strategy of budget adjustment with real world data from search advertising campaigns. Experimental results showed that our strategy outperforms the two other baseline strategies.

关键词： Marketing

来源：评论

学校读者我要写书评

暂无评论

H∞ analysis method for the small signal stability of power system

H∞ analysis method for the small signal stability of power ...

引用

26th Chinese control Conference, CCC 2007

作者： Mei, Shengwei Che, Wenyan State Key Lab. of Control and Simulation of Power Systems and Generation Equipments Dept. of Electrical Engineering Tsinghua University Beijing 100084 China

ISBN: (纸本)7900719229

This work is concerned with the small signal stability problem in multi-machine systems. By means of H∞ theory, firstly the math model with disturbances is constructed;and secondly, the excitation control law for each generator is designed based on solving Riccati Equation. Finally, simulations performed on 4-machine system shows that the proposed H ∞ method has more advantages over LQR method in both damping ratio and H∞ norm of the corresponding closed-loop system and so can enhance the small signal stability of power system greatly.

关键词： Electric power systems

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：