检索结果-南通市图书馆

Focusing on your subject:Deep subject-aware image composition recommendation networks

Computational Visual Media 2023年第1期9卷 87-107页

作者： Guo-Ye Yang Wen-Yang Zhou Yun cai Song-Hai Zhang Fang-Lue Zhang BNRist Department of Computer Science and TechnologyTsinghua UniversityBeiing 100084China School of Engineering and computer science Victoria University of WellingtonWellington 6012New Zealand

Photo composition is one of the most important factors in the aesthetics of *** a popular application,composition recommendation for a photo focusing on a specific subject has been ignored by recent deep-learning-based composition recommendation *** this paper,we propose a subject-aware image composition recommendation method,SAC-Net,which takes an RGB image and a binary subject window mask as input,and returns good compositions as crops containing the *** model first determines candidate scores for all possible coarse cropping *** crops with high candidate scores are selected and further refined by regressing their corner points to generate the output recommended cropping *** final scores of the refined crops are predicted by a final score regression *** existing methods that need to preset several cropping windows,our network is able to automatically regress cropping windows with arbitrary aspect ratios and *** propose novel stability losses for maximizing smoothness when changing cropping windows along with view *** results show that our method outperforms state-of-the-art methods not only on the subject-aware image composition recommendation task,but also for general purpose composition *** also have designed a multistage labeling scheme so that a large amount of ranked pairs can be produced *** use this scheme to propose the first subject-aware composition dataset SACD,which contains 2777 images,and more than 5 million composition ranked *** SACD dataset is publicly available at https://***/SACD/.

关键词： subject-aware image composition image cropping deep learning recommendation

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

SmartPipe: Towards Interoperability of Industrial Applications via Computational Reflection

引用

Journal of computer science & technology 2020年第1期35卷 161-178页

作者： Su Zhang Hua-Qian cai Yun Ma Tian-Yue Fan Ying Zhang Gang Huang Key Laboratory of High-Confidence Software technology(Peking university) Ministry of Education Beijing 100871China School of Software Tsinghua UniversityBeijing 100084China Hengyi Petrochemicals CO. LTD.Hangzhou 311215China National Engineering Research Center for Software Engineering Peking UniversityBeijing 100871China

With the advancement of new information technologies,a revolution is being taken place to bring the industry into a new era of intelligent *** of the key requirements of intelligent manufacturing is the interoperability of industrial ***,it is challenging to realize the interoperability for legacy industrial applications due to 1)the deficient semantic information of data transmitted over heterogeneous communication protocols,2)the difficulty to understand the complex process of business logic with no source code,and 3)the high cost and potential risk of reengineering the *** address the issues,in this paper,we propose an approach named SmartPipe to exposing existing functionalities of an industrial application as APIs without source code while simultaneously allowing the application to remain *** design a behavioral runtime model(BRM)as the self-representation of the industrial applications,based on which a computational reflection framework is designed to flexibly construct the model and generate APIs that encapsulate specific *** validate SmartPipe on a real industrial application that controls the spin-draw winding *** show that our approach is effective and more suitable for industrial scenes compared with traditional approaches.

关键词： computational reflection runtime model interoperability industrial application API generation

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

An Improved String-Searching Algorithm and Its Application in Component Security Testing

引用

tsinghua science and technology 2016年第3期21卷 281-294页

作者： Jinfu Chen Saihua cai Lili Zhu Yuchi Guo Rubing Huang Xiaolei Zhao Yunqi Sheng School of computer science and Communication Engineering Jiangsu University

Mass monitor logs are produced during the process of component security testing. In order to mine the explicit and implicit security exception information of the tested component, the log should be searched for keyword strings. However, existing string-searching algorithms are not very efficient or appropriate for the operation of searching monitor logs during component security testing. For mining abnormal information effectively in monitor logs, an improved string-searching algorithm is proposed. The main idea of this algorithm is to search for the first occurrence of a character in the main string. The character should be different and farther from the last character in the pattern string. With this algorithm, the backward moving distance of the pattern string will be increased and the matching time will be optimized. In the end, we conduct an experimental study based on our approach, the results of which show that the proposed algorithm finds strings in monitor logs 11.5% more efficiently than existing approaches.

关键词： component testing security vulnerabilities detection monitor log abnormal information string-searching

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

An efficient VLSI architecture of speeded-up robust feature extraction for high resolution and high frame rate video

引用

science China(Information sciences) 2013年第7期56卷 136-149页

作者： ZHANG WeiLong LIU LeiBo YIN ShouYi ZHOU RenYan cai ShanShan WEI ShaoJun Institute of Micro-Electronics and the National Laboratory for Information science and technology Tsinghua University

This paper proposes a VLSI architecture of the optimized Speeded-Up Robust Feature (SURF) algorithm. The SURF algorithm which is widely used in computer vision applications, locates interest points (IPoints) and extracts feature descriptors based on the surrounding gradients. In the proposed approach, the SURF algorithm is modified to make it more suitable for hardware implementation with little accuracy lost compared to OpenSURF, an open source software implementation based on OpenCV. The resource cost and throughput gain are balanced. Word Length Reduction (WLR) is adopted to compress the integral image and reduce the occupied memory resources. The orientations and the feature descriptors of the IPoints are calculated in a more efficient way, which significantly reduces memory accesses. Moreover, the operations are pipelined both in and among the proposed hardware modules. This VLSI architecture has been validated on FPGA (Xilinx Virtex-4 XC4VLX80), which is able to detect IPoints and extract SURF feature descriptors in a VGA (640 × 480) 64 fps video input with a 96 MHz working frequency while dissipating 1.278 W. This throughput is more than double of the ones reported in the latest literatures.

关键词： SURF paralleled VLSI architecture interest point feature extraction WLR

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

A single layer zero skew clock routing in X architecture

引用

science in China(Series F) 2009年第8期52卷 1466-1475页

作者： SHEN WeiXiang cai YiCi HONG XianLong HU Jiang LU Bing Electronic Design Automation Laboratory Departmeat of Computer Science and Technology Tsinghua University Beijing 100084 China Electrical and computer Engineering Texas A&M University College Station TX 77843 USA Cadence Design Systems Inc 35 Spring Street New Providence NJ 07974 USA

With its advantages in wirelength reduction and routing flexibility compared with conventional Manhattan routing, X architecture has been proposed and applied to modern IC design. As a critical part in high-performance integrated circuits, clock network design meets great challenges due to feature size decrease and clock frequency increase. In order to eliminate the delay and attenuation of clock signal introduced by the vias, and to make it more tolerant to process variations, in this paper, we propose an algorithm of a single layer zero skew clock routing in X architecture （called Pianar-CRX）. Our Planar- CRX method integrates the extended deferred-merge embedding algorithm （DME-X, which extends the DME algorithm to X architecture） with modified Ohtsuki＇s line-search algorithm to minimize the total wirelength and the bends. Compared with planar clock routing in the Manhattan plane, our method achieves a reduction of 6.81% in total wirelength on average and gets the resultant clock tree with fewer bends. Experimental results also indicate that our solution can be comparable with previous non-planar zero skew clock routing algorithm.

关键词： clock routing single layer X architecture zero skew

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

光学感前计算技术及其卫星遥感应用综述(特邀)

引用

激光与光电子学进展 2024年第2期61卷 425-439页

作者：李天宇王国庆李维陈宏伟刘勋王志斌刘少冲蔡滟馨电子科技大学计算机科学与工程学院四川成都611731 北京空间机电研究所北京100094 清华大学电子工程系北京100084

光学感前计算是一种在光电传感器前端的光学域进行信息计算处理的技术,包括编码压缩、全光智能推理等计算范式,具有传输即计算、结构即功能等显著特点,在卫星光学遥感领域有着广泛的应用前景。首先对用于感前计算的光场调制器件进行介绍... 详细信息

光学感前计算是一种在光电传感器前端的光学域进行信息计算处理的技术,包括编码压缩、全光智能推理等计算范式,具有传输即计算、结构即功能等显著特点,在卫星光学遥感领域有着广泛的应用前景。首先对用于感前计算的光场调制器件进行介绍,包括数字微镜器件(DMD)、液晶空间光调制器(LC-SLM)、衍射光学元件(DOE)及超表面等。然后分别梳理了感前编码压缩及全光智能推理的相关技术发展,在此基础上,着重讨论光学感前计算在卫星遥感领域的应用途径和未来发展趋势。

关键词：光学感前计算编码压缩全光智能推理卫星遥感

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Duple-EDA and sample density balancing

引用

science in China(Series F) 2009年第9期52卷 1640-1650页

作者： cai YunPeng XU Hua SUN XiaoMin JIA PeiFa LIU ZeHua State Key Laboratory on Intelligent technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing 100084 China School of computer Engineering Nanyang Technological University Singapore 639798 Singapore

In this paper, a new method is proposed to overcome the problem of local optima traps in a class of evolutionary algorithms, called estimation of distribution algorithms （EDAs）, in real-valued function optimization. The Duple-EDA framework is proposed in which not only the current best solutions but also the search history are modeled, so that long-term feedback can be taken into account. Sample Density Balancing （SDB） is proposed under the framework to alleviate the drift phenomenon in EDA. A selection scheme based on Pareto ranking considering both the fitness and the historical sample density is adopted, which prevents the algorithm from repeatedly sampling in a small region and directs it to explore potentially optimal regions, thus helps it avoid being stuck into local optima. An MBOA （mixed Bayesian optimization algorithm） version of the framework is implemented and tested on several benchmark problems. Experimental results show that the proposed method outperforms a standard niching method in these benchmark problems.

关键词： evolutionary computation estimation of distribution algorithm Boltzmann selection

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Patterns of Chromatin-Modifications Discriminate Different Genomic Features in Arabidopsis

引用

tsinghua science and technology 2013年第5期18卷 431-440页

作者： Anuj Srivastava Xiaoyu Zhang Sal LaMarca Liming cai Russell L. Malmberg Institute of Bioinformatics University of Georgia The Jackson Laboratory department of Plant Biology University of Georgia department of computer science University of Georgia

Dynamic regulation and packaging of genetic information is achieved by the organization of DNA into chromatin. Nucleosomal core histones, which form the basic repeating unit of chromatin, are subject to various post-translational modifications such as acetylation, methylation, phosphorylation, and ubiquitinylation. These modifications have effects on chromatin structure and, along with DNA methylation, regulate gene *** goal of this study was to determine if patterns in modifications were related to different categories of genomic features, and, if so, if the patterns had predictive value. In this study, we used publically available data（ChIP-chip）for different types of histone modifications（methylation and acetylation） and for DNA methylation for Arabidopsis thaliana and then applied a machine learning based approach（a support vector machine） to demonstrate that patterns of these modifications are very different among different kinds of genomic feature categories（protein, RNA,pseudogene, and transposon elements）. These patterns can be used to distinguish the types of genomic *** methylation and H3K4me3 methylation emerged as features with most discriminative power. From our analysis on Arabidopsis, we were able to predict 33 novel genomic features, whose existence was also supported by analysis of RNA-seq experiments. In summary, we present a novel approach which can be used to discriminate/detect different categories of genomic features based upon their patterns of chromatin modification and DNA methylation.

关键词： chromatin modification DNA methylation support vector machine machine learning Arabidopsis

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Application of optical proximity correction technology

引用

science in China(Series F) 2008年第2期51卷 213-224页

作者： cai YiCi ZHOU Qiang HONG XianLong SHI Rui WANG Yang department of computer science and technology Tsinghua University Beijing 100084 China

As process technology scales down to very deep sub-micron （VDSM） in semiconductor manufacturing technology, intrinsic size becomes close to or even shorter than the wavelength used for optical lithography. Thus, some distortions and deformations are introduced by optical proximity effects （OPE） mainly caused by the diffraction and interference of exposure light when layout patterns on a mask are transcribed to a wafer, which influence on the yield and performance of IC circuit. In order to compensate for the deformations, optical proximity correction （OPC） is the most commonly used methodology. Presently, the OPC method is to use a unitary toleration on the whole chip layer, which makes the run time of OPC algorithm longer, causes the size of GDSII files to follow exponential growth, and results in the cost of making mask grow immensely. Firstly, this paper proposes a selfadaptation OPC method with preprocessing function of patterns classification. According to the need of the correction precision, the OPC system divides patterns corrected into two groups with different toleration： critical patterns and general patterns, which enhance the efficiency of the OPC approach. Secondly, a model-based OPC method is presented based on pattern subsection and classifi- cation, which keeps the precision of the correction as well as enhances the efficiency. We also propose a rule-based OPC method with general, concise and complete correction rules, and achieve automatic-built rules-based and its looking-up. Thirdly, we also implement an OPC system, called MR-OPC; the MR-OPC system integrates both rule-based OPC and model-based OPC into a whole, so it can solve the confliction between the efficiency and precision. Experimental results show that the MR-OPC system we suggested has advantages of the efficiency and expansibility.

关键词： OPC rules-based model-based layout IC

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Reaction-diffusion modeling of malware propagation in mobile wireless sensor networks

引用

science China(Information sciences) 2013年第9期56卷 257-274页

作者： WANG XiaoMing HE ZaoBo ZHAO XueQing LIN Chuang PAN Yi cai ZhiPeng School of computer science Shaanxi Normal University department of computer science and technology Tsinghua University department of computer science Georgia State University

Mobile Wireless Sensor Networks (MWSNs) are employed in many fields, such as intelligent transportation, community health monitoring, and animal behavior monitoring. However, MWSNs may be vulnerable to malicious interference because of the large-scale characteristics. One of the threats is to inject malware into some nodes, especially mobile nodes. When a contaminated node communicates with its neighbors, multiple copies of the malware are transmitted to its neighbors, which may destroy nodes, block regular communications, or even damage the integrity of regular data packets. This work develops a modeling framework which mathematically characterizes the process of malware propagation in MWSNs based on the theory of reaction-diffusion equation. Our proposed model can efficiently predict the temporal dynamic behavior and spatial distribution of malware propagation over time, so that targeted immunization measures can be taken on infected nodes, whereas most of the existing models for malware propagation can only predict the temporal dynamic behavior rather than the spatial distribution of malware propagation over time. We conduct extensive simulations on large-scale MWSNs to evaluate the proposed model. The simulation results indicate that the proposed model and method are efficient, and that the mobile speed, communication range, and packet transmission rate of nodes are the main factors affecting malware propagation in MWSNs.

关键词： mobile wireless sensor network malware propagation reaction-diffusion equation equilibrium point stability spatial distribution targeted immunization strategy

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：