检索结果-南通市图书馆

GSM-MRF based classification approach for real-time moving object detection

在线全文

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2008年第2期9卷 250-255页

作者： xiang pan Yi-jun WU Institute of Information and Communication Engineering Zhejiang University Hangzhou 310027 China

Statistical and contextual information are typically used to detect moving regions in image sequences for a fixed *** this paper,we propose a fast and stable linear discriminant approach based on Gaussian Single Model(GSM)and Markov Random Field(MRF).The performance of GSM is analyzed first,and then two main improvements corresponding to the drawbacks of GSM are proposed:the latest filtered data based update scheme of the background model and the linear classification judgment rule based on spatial-temporal feature specified by *** results show that the proposed method runs more rapidly and accurately when compared with other methods.

关键词： Moving object detection Markov Random Field (MRF) Gaussian Single Model (GSM) Fisher Linear Discriminant Analysis (FLDA)

Image-based 3D model retrieval using manifold learning

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2018年第11期19卷 1397-1408页

作者： pan-pan MU San-yuan ZHANG Yin ZHANG Xiu-zi YE xiang pan College of Computer Science and Technology Zhejiang University College of Mathematics and Information Science Wenzhou University College of Computer Science and Technology Zhejiang University of Technology

We propose a new framework for image-based three-dimensional(3D) model retrieval. We first model the query image as a Euclidean point. Then we model all projected views of a 3D model as a symmetric positive definite(SPD) matrix, which is a point on a Riemannian manifold. Thus, the image-based 3D model retrieval is reduced to a problem of Euclid-to-Riemann metric learning. To solve this heterogeneous matching problem, we map the Euclidean space and SPD Riemannian manifold to the same high-dimensional Hilbert space, thus shrinking the great gap between them. Finally, we design an optimization algorithm to learn a metric in this Hilbert space using a kernel trick. Any new image descriptors, such as the features from deep learning, can be easily embedded in our framework. Experimental results show the advantages of our approach over the state-of-the-art methods for image-based 3D model retrieval.

关键词： Model retrieval Euclidean space Riemannian manifold Hilbert space Metric learning

A Unified Feature Representation and Learning Framework for 3D Shape

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2019年第5期28卷 993-999页

作者： MU panpan ZHANG Sanyuan pan xiang HONG Zhenjie Art Design College Zhejiang Gongshang University College of Computer Science and Technology Zhejiang University College of Computer Science and Technology Zhejiang University of Technology School of Mathematics and Information Science Wenzhou University

In conventional 3D shape retrieval and classification, they differentiate each other in their final stages. We propose a unified feature representation and learning framework for the instance-based shape retrieval and classification. Firstly, we render every 3D model in several directions and use the produced view-sets to represent the 3D models. In this way, both tasks can be tackled by measuring the distances between rendered views of 3D models. Secondly, we construct the viewsets as Symmetric positive definite matrices(SPDMs),which are points on a Riemannian manifold. Thus, the shape retrieval and classification tasks are reduced to a problem of measuring the distances between projected views and SPDMs. To solve this heterogeneous problem,we map them to a Hilbert space using a method of point-to-set matching. In this Hilbert space, the distances are surprisingly easy to calculate. Finally, we use a robust nearest-neighbor approach to unify the instancebased shape retrieval and classification. Our framework combines the state-of-the-art deep learning approaches with traditional mathematical optimization method, makes full use of both advantages, which is much more flexible than pure deep learning methods. Experimental results show the efficiency of our approach.

关键词： Shape retrieval Shape classification Deep learning Euclidean space Riemannian manifold Hilbert space Metric learning

Pseudo-dynamic tests on masonry residential buildings seismically retrofitted by precast steel reinforced concrete walls

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Earthquake Engineering and Engineering Vibration 2017年第3期16卷 587-597页

作者： Li Wenfeng Wang Tao Chen Xi Zhong xiang pan Peng Beijing Institute of Architectural Design Beijing 100045 China Key Laboratory of Earthquake Engineering and Engineering Vibration Institute of Engineering MechanicsChina Earthquake Administration Sanhe 065201 China Department of Civil Engineering Tsinghua University Beijing 100084 China

A retrofitting technology using precast steel reinforced concrete（PSRC） panels is developed to improve the seismic performance of old masonry buildings. The PSRC panels are built up as an external PSRC wall system surrounding the existing masonry building. The PSRC walls are well connected to the existing masonry building, which provides enough confinement to effectively improve the ductility, strength, and stiffenss of old masonry structures. The PSRC panels are prefabricated in a factory, significantly reducing the situ work and associated construction time. To demonstrate the feasibility and mechanical effectivenss of the proposed retrofitting system, a full-scale five-story specimen was constructed. The retrofitting process was completed within five weeks with very limited indoor operation. The specimen was then tested in the lateral direction, which could potentially suffer sigifnicant damage in a large earthquake. The technical feasibility, construction workability, and seismic performance were thoroughly demonstrated by a full-scale specimen construction and pseudo-dynamic tests.

关键词： masonry buildings precast SRC panel seismic retrofit pseudo-dynamic test seismic performance

Efficient photocatalysis triggered by thin carbon layers coating on photocatalysts:recent progress and future perspectives

在线全文

学校读者我要写书评

暂无评论

Science China Chemistry 2020年第10期63卷 1416-1427页

作者： Yu-Long Men Peng Liu Xingcui Peng Yun-xiang pan Institute of Nano Biomedicine and Engineering Shanghai Engineering Research Centre for Intelligent Diagnosis and Treatment InstrumentDepartment of Instrument Science and EngineeringSchool of Electronic Information and Electrical EngineeringShanghai Jiao Tong UniversityShanghai 200240China

Solar-energy-driven photocatalysis,such as photocatalytic reduction of CO2,is promising simultaneously for the energy and environmental *** thin carbon layers with the thickness less than 10 nm on photocatalysts has been developed as an efficient strategy for enhancing the photocatalytic efficiency in recent *** the present review,we summarize the crucial progress on carbon-coated *** for the improved light absorption,charge separation,reactant adsorption and photocatalytic stability on carbon-coated photocatalysts as well as the applications of carbon-coated photocatalysts are *** opportunities and challenges associated with carbon-coated photocatalysts are shown at the end of the *** hope that the present review can trigger more deep insights on carbon-coated photocatalysts and provide new opportunities for developing low-cost but efficient photocatalysts.

关键词： carbon-coated photocatalysts light absorption charge separation reactant adsorption photocatalytic stability

Influencing factors on elastic-plastic deformation of multi-layered surfaces under sliding contact

在线全文

学校读者我要写书评

暂无评论

中国有色金属学会会刊：英文版 2004年第Z1期14卷 472-477页

作者： YAN Li pan Xin-xiang XU Jiu-jun CHENG Dong Institute of Materials and Technology Dalian Maritime University China Institute of Materials and Technology Dalian Maritime University China Institute of Materials and Technology Dalian Maritime University China Institute of Materials and Technology Dalian Maritime University China

Stress distribution in the gradient multi-layered surface under a sliding contact was investigated using finite element method(FEM). The main structure parameters of layered surface discussed are total layer thickness,layer number and elastic modulus ratio of layer to the substrate. A model of multi-layered surface contact with rough slider was studied. The effect of the surface structure parameters on the elastic-plastic deformation was analyzed.

关键词： gradient multi-layered surface elastic-plastic deformation surface structure parameter FEM

Robust time reversal processing for active detection of a small bottom target in a shallow water waveguide

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

Journal of Zhejiang University-Science C(Computers and Electronics) 2010年第5期11卷 401-406页

作者： xiang pan Jian-long LI Wen XU Xian-yi GONG Institute of lnformation and Communication Engineering Zhejiang University Hangzhou 310027 China Zhejiang Provincial Key Laboratory of Information Network Technology Hangzhou 310027 China

With the spatial-temporal focusing of acoustic energy,time reversal processing(TRP) shows the potential application for active target detection in shallow *** turn the ‘potential’ into a reality,the TRP based on a model source(MS) instead of a physical probe source(PS) is *** uncertain ocean environments,the robustness of TRP is discussed for the narrowband and broadband signal *** channel transfer function matrix is first constructed in the acoustic perturbation *** a steering vector for time reversal transmission is obtained by singular value decomposition(SVD) of the *** verification of the robust TRP,the tank experiments of time reversal transmission focusing and its application for active target detection are *** experimental results have shown that the robust TRP can effectively detect and locate a small bottom target.

关键词： Robustness Time reversal processing Bottom target detection Uncertain environment Sonar

Robust time reversal focusing based on Maximin criterion in a waveguide with uncertain water depth

在线全文

学校读者我要写书评

暂无评论

Science China(Physics,Mechanics & Astronomy) 2013年第10期56卷 1822-1832页

作者： pan xiang WANG Nan ZHANG JiangFan XU Wen GONG XianYi Department of Information Science and Electronic Engineering Zhejiang University Hangzhou Applied Acoustic Research Institute

Time reversal processing （TRP） might be regarded as matched field processing with known environmental knowledge. However, the performance of TRP is degraded in an uncertain environment. A technique based on the Maximin criterion is pro- posed for enhancing the robustness of TRP in a waveguide with uncertain water depth. The relationship between the water depth and the focal spot translation is examined based on the waveguide-invariant theory. Then the time reversal transmissiou scheme with the Maximin criterion is performed to maximize the minimum transmission power on a target of interest. At the receiving end, coherent summation operation is carried out over the received data by a reception focusing bank. If it is neces- sary to enhance the target echo further, the iterative time reversal can be considered where the target echo corresponding to the first time reversal transmission is regarded as a secondary source. Numerical simulations and experimental results of the target localization in a waveguide tank have verified the effectiveness of robust TRP.

关键词： robustness time reversal processing uncertain water depth waveguide sonar

Mie Potential and Equation of State of Zr_(48)Nb_(8)Cu_(14)Ni_(12)Be_(18) Bulk Metallic Glass

在线全文

学校读者我要写书评

暂无评论

Chinese Physics Letters 2001年第6期18卷 805-807页

作者： ZHANG Yong pan Ming-xiang WANG Wei-Hua Institute of Physics and Center for Condensed Matter Physics Chinese Academy of SciencesBeijing 100080 National Microgravity Laboratory of Chinese Academy of Sciences Beijing 100080

The linear expansion of Zr_(48)Nb_(8)Cu_(14)Ni_(12)Be_(18) bulk metallic glass(BMG)with excellent glass forming ability and high thermal stability is investigated by a dilatometry *** average expansion coefficient is α TG=1.04 × 10^(-5) K^(-1)(300-656 K)for the BMG and α TG=1.11 × 10^(-5) K^(-1)(356~890 K)for the crystallized *** Mie potential as well as the equation of state of the BMG and its corresponding crystallized state are determined from the thermal expansion and ultrasonic data,and the differences among them are phenomenologically explained.

关键词： alloy. expansion thermal