检索结果-内蒙古大学图书馆

Scaling Up Single Image Dehazing Algorithm by Cross-Data Vision Alignment for Richer Representation Learning and Beyond

引用

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Shi, Yukai Weng, Zhipeng Lin, Yupei Shi, Cidan Guo, Xingyuan Yang, Xiaojun Lin, Liang Guangdong University of Technology School of Information Engineering Guangzhou510006 China Anhui University Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Hefei230601 China Southern Power Grid Ltd. Guangzhou510000 China Ministry of Education of China Key Laboratory of Photonic Technology for Integrated Sensing and Communication Guangdong Guangzhou510006 China Sun Yat-sen University School of Computer Science Guangzhou510006 China

In recent years, deep neural network tasks have increasingly relied on high-quality image inputs. With the development of high-resolution representation learning, the task of image dehazing has received significant attention. Previously, many methods collected diverse image data for large-scale training to boost the performance on a target scene. Ignoring the domain gap between different data, former dehazing methods simply adopt multiple datasets for explicit large-scale training, which often makes the methods themselves violated. To address this problem, we propose a novel method of cross-data vision alignment for richer representation learning to improve the existing dehazing methodology. Specifically, we call for the internal and external knowledge should be further adapted in a self-supervised manner to fill up the domain gap. By using cross-data external alignment, the datasets inherit samples from different domains that are firmly aligned, making the model learn more robust and generalizable features. By using the internal augmentation method, the model can fully exploit local information within the images and then obtain more image details. To demonstrate the effectiveness of our proposed method, we conduct training on the natural image dataset (NID). Experimental results show that our method clearly resolves the domain gap in different dehazing datasets and presents a new pipeline for large-scale training in the dehazing task. Our approach significantly outperforms other advanced methods in dehazing and produces dehazed images that are closest to real haze-free images. © 1963-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Local reputation for P2P MMOG design

Local reputation for P2P MMOG design

引用

18th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2007

作者： Shanshan, Liu Jinlong, Li Xufa, Wang Anhui Key Laboratory of Software in Computation and Communication Computer Science Department University of Science and Technology of China

ISBN: (纸本)0769530494

The assignment of the computation and communication is one of the important problems in the design of peer-to-peer (P2P) Massively Multi-player Online Game (MMOG). A reputation mechanism for P2P MMOG is proposed in this paper to avoid the problems in P2P design. The communications among peers are described by whether the messages are received timely and correctly in the process of synchronization, and then the reliabilities of peers are quantified. The concept of local reputation is proposed, based on which the adaptive adjusting algorithm for the function of peers and the algorithm for getting update peer selecting reliable computing peers are brought forward to assign the computation and communication of the system. According to the experiments, the algorithms are effective for the distribution of the computation and communication. © 2007 IEEE.

关键词： Peer to peer networks

来源：评论

学校读者我要写书评

暂无评论

A Power Control Scheme for Target Localization in WSN

引用

电子学报(英文版) 2008年第3期17卷 443-448页

作者： GU Yu ZHAO Baohua Department of Computer Science University of Science and Technology of China Anhui Province Key Laboratory of Software in Computing and Communication Hefei 230027 China

In this paper, we focus on the target tracking problem in sensor networks and propose an Powersaving target localization scheme (PSTL) based on a conjecture model that reflects the moving pattern of a target, and also a corresponding two-step communication protocol between Base station (BS) and sensors. BS executes a query mechanism to determine which sensors should be used for detailed information according to a limited amount of data received from sensors. This scheme reduces both energy consumption and communication bandwidth requirement, prolongs the lifetime of the wireless sensor networks. Simulation results indicate that it can achieve a high accuracy while saving a large amount of energy.

关键词：无线传感器网络功率控制目标定位推测模型

来源：评论

学校读者我要写书评

暂无评论

Minimum-Energy Broadcast Trees in All-Wireless Networks

引用

电子学报(英文版) 2008年第2期17卷 227-230页

作者： HU Tian ZHAO Baohua Department of Computer Science University of Science and Technology of China Anhui Province Key Laboratory of Software in Computing and Communication Hefei 230027 China

This paper focuses on the problem of constructing the minimum-energy broadcast trees in all-wireless networks. We proposed a greedy algorithm called Greedy maximum-branch replacement algorithm (GMBR) to decrease the total power of the broadcast tree further. This algorithm can be developed to a distributed one easily. Compared with the previous algorithms, GMBR was proved to have better performance, especially when the propagation loss exponent and the network size become larger.

关键词：最小能量广播天线网络技术性能

来源：评论

学校读者我要写书评

暂无评论

Secure Two-Party Point-Circle Inclusion Problem

引用

Journal of Computer Science & Technology 2007年第1期22卷 88-91页

作者：罗永龙黄刘生仲红 Department of Computer Science University of Science and Technology of China Hefei 230027 China Department of Computer Science Anhui Normal University Wuhu 241000 China Anhui Province Key laboratory of Software in Computing and Communication Hefei 230027 China

Privacy-preserving computational geometry is a special secure multi-party computation and has many applications. Previous protocols for determining whether a point is inside a circle are not secure enough. We present a two-round protocol for computing the distance between two private points and develop a more efficient protocol for the point-circle inclusion problem based on the distance protocol. In comparison with previous solutions, our protocol not only is more secure but also reduces the number of communication rounds and the number of modular multiplications significantly.

关键词： secure multi-party computation computational geometry homomorphic encryption scheme private comparison

来源：评论

学校读者我要写书评

暂无评论

A preliminary study on why using the nonself detector set for anomaly detection in artificial immune Systems

A preliminary study on why using the nonself detector set fo...

引用

2009 International Conference on computational Intelligence and Security, CIS 2009

作者： Xu, Baoliang Luo, Wenjian Wang, Xufa Nature Inspired Computation and Applications Laboratory School of Computer Science and Technology University of Science and Technology of China Hefei 230027 Anhui China Anhui Key Laboratory of Software in Computing and Communication University of Science and Technology of China Hefei 230027 Anhui China

ISBN: (纸本)9780769539317

In artificial immune systems, the detectors in the nonself space are often adopted to detect anomaly changes, such as the negative selection algorithm and its improvements. Since the detectors in the self space can also be used to detect anomaly changes, a frequently asked question is which kind of detector sets is more efficient for a specific problem. In this paper, firstly, the advantages and disadvantages of the self detector set and the nonself detector set are briefly reviewed. Secondly, when the complete matching rule is adopted, the average time costs of employing the nonself detector set and the self detector set for anomaly detection are compared theoretically. Thirdly, simulated experiments are done, and experimental results demonstrate that the theoretical conclusion is essentially correct. © 2009 IEEE.

关键词： Immune system

来源：评论

学校读者我要写书评

暂无评论

A novel updating strategy for associative memory scheme in cyclic dynamic environments

A novel updating strategy for associative memory scheme in c...

引用

International Workshop on Advanced computational Intelligence (IWACI)

作者： Cao, Yong Luo, Wenjian Nature Inspired Computation and Applications Laboratory School of Computer Science and Technology University of Science and Technology of China Hefei 230027 Anhui China Anhui Key Laboratory of Software in Computing and Communication University of Science and Technology of China Hefei 230027 Anhui China

ISBN: (纸本)9781424463343

Associative memory schemes have been developed for Evolutionary Algorithms (EAs) to solve Dynamic Optimization Problems (DOPs), and demonstrated powerful performance. In these schemes, how to update the memory could be important for their performance. However, little work has been done about the associative memory updating strategies. In this paper, a novel updating strategy is proposed for associative memory schemes. In this strategy, the memory point whose associated environmental information is most similar to the current environmental information is first picked out from the memory. Then, the selected memory individual is updated according to the fitness value, and the associated environmental information is updated according to the matching degree between environmental information and individuals. This updating strategy is embedded into a stateof-the-art algorithm, i.e. the MPBIL, and tested by experiments. Experimental results demonstrate that the proposed updating strategy is helpful for associative memory schemes to enhance their search ability in cyclic dynamic environments.

关键词： Associative processing

来源：评论

学校读者我要写书评

暂无评论

Classifying and clustering in negative databases

引用

Frontiers of Computer Science 2013年第6期7卷 864-874页

作者： Ran LIU Wenjian LUO Lihua YUE School of Computer Science and Technology University of Science and Technology of China Hefei 230027 China Anhui Province Key Laboratory of Software Engineering in Computing and Communication University of Science and Technology of China Hefei 230027 China

Recently, negative databases （NDBs） are proposed for privacy protection. Similar to the traditional databases, some basic operations could be conducted over the NDBs, such as select, intersection, update, delete and so on. However, both classifying and clustering in negative databases have not yet been studied. Therefore, two algorithms, i.e., a k nearest neighbor （kNN） classification algorithm and a k-means clustering algorithm in NDBs, are proposed in this paper, respectively. The core of these two algorithms is a novel method for estimating the Hamming distance between a binary string and an NDB. Experimental results demonstrate that classifying and clustering in NDBs are promising.

关键词： negative databases classification clustering k nearest neighbor k-means hamming distance

来源：评论

学校读者我要写书评

暂无评论

Boosting performance of virtualized desktop infrastructure with physical GPU and SPICE

引用

Science China(Information Sciences) 2020年第7期63卷 236-238页

作者： Shupan LI Limin XIAO Chungang SHI Liequan CHE Changyou ZHANG Yuanzhang LI State Key Laboratory of Software Development Environment Beihang University School of Computer Science and Engineering Beihang University Beijing Jinghang Computation and Communication Research Institute Institute of Software Chinese Academy of Sciences School of Computer Science and Technology Beijing Institute of Technology

Dear editor,The virtual desktop infrastructure (VDI) is widely used by enterprises, as it can be easily deployed,has a reduced management cost, improves resource utilization, and can be accessed at any location by the network. Currently, there is a demand for the VDIs to support applications, such as playing 3D games, playing HD videos, and using 3D design-

关键词： visualized desktop infrastructure SPICE physical GPU The virtual machine (VM) virtual machine monitor (VMM)

来源：评论

学校读者我要写书评

暂无评论

An adaptive strategy for updating the memory in evolutionary algorithms for dynamic optimization

An adaptive strategy for updating the memory in evolutionary...

引用

IEEE Symposium on computational Intelligence in Dynamic and Uncertain Environments

作者： Zhu, Tao Luo, Wenjian Li, Zhifang Nature Inspired Computation and Applications Laboratory School of Computer Science and Technology University of Science and Technology of China Hefei 230027 Anhui China Anhui Key Laboratory of Software Engineering in Computing and Communication University of Science and Technology of China Hefei 230027 China

ISBN: (纸本)9781424499311

The memory scheme is one of the most widely employed techniques in Evolutionary Algorithms for solving dynamic optimization problems. The updating strategy is a key concern for the memory scheme. Unfortunately, the existent memory updating strategies neglect the characteristics of the memory updating behaviors, and sometimes this could lead results against the original intention. In this paper, a novel updating strategy is proposed, which can adaptively update the memory according to the characteristics of the memory updating behaviors. Experiments are carried out in different kinds of dynamic environments, and the experimental results show that the proposed strategy is better than the traditional strategies. © 2011 IEEE.

关键词： Evolutionary 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：