Dataset classification is an essential fundament of computational intelligence in cyber-physical systems(CPS).Due to the complexity of CPS dataset classification and the uncertainty of clustering number,this paper foc...Dataset classification is an essential fundament of computational intelligence in cyber-physical systems(CPS).Due to the complexity of CPS dataset classification and the uncertainty of clustering number,this paper focuses on clarifying the dynamic behavior of acceleration dataset which is achieved from micro electro mechanical systems(MEMS)and complex image segmentation.To reduce the impact of parameters uncertainties with dataset classification,a novel robust dataset classification approach is proposed based on neighbor searching and kernel fuzzy c-means(NSKFCM)methods.Some optimized strategies,including neighbor searching,controlling clustering shape and adaptive distance kernel function,are employed to solve the issues of number of clusters,the stability and consistency of classification,respectively.Numerical experiments finally demonstrate the feasibility and robustness of the proposed method.展开更多
An innovative damage identification method using the nearest neighbor search method to assess 3D structures is presented.The frequency response function was employed as the input parameters to detect the severity and ...An innovative damage identification method using the nearest neighbor search method to assess 3D structures is presented.The frequency response function was employed as the input parameters to detect the severity and place of damage in 3D spaces since it includes the most dynamic characteristics of the structures.Two-dimensional principal component analysis was utilized to reduce the size of the frequency response function data.The nearest neighbor search method was employed to detect the severity and location of damage in different damage scenarios.The accuracy of the approach was verified using measured data from an experimental test;moreover,two asymmetric 3D numerical examples were considered as the numerical study.The superiority of the method was demonstrated through comparison with the results of damage identification by using artificial neural network.Different levels of white Gaussian noise were used for polluting the frequency response function data to investigate the robustness of the methods against noise-polluted data.The results indicate that both methods can efficiently detect the damage properties including its severity and location with high accuracy in the absence of noise,but the nearest neighbor search method is more robust against noisy data than the artificial neural network.展开更多
The core of smoothed particle hydrodynamics (SPH) is the nearest neighbor search subroutine. In this paper, a nearest neighbor search algorithm which is based on multiple background grids and support variable smooth...The core of smoothed particle hydrodynamics (SPH) is the nearest neighbor search subroutine. In this paper, a nearest neighbor search algorithm which is based on multiple background grids and support variable smooth length is introduced. Through tested on lid driven cavity flow, it is clear that this method can provide high accuracy. Analysis and experiments have been made on its parallelism, and the results show that this method has better parallelism and with adding processors its accuracy become higher, thus it achieves that efficiency grows in pace with accuracy.展开更多
In this study, we extend our previous adaptive steganographic algorithm to support point geometry. For the purpose of the vertex decimation process presented in the previous work, the neighboring information between p...In this study, we extend our previous adaptive steganographic algorithm to support point geometry. For the purpose of the vertex decimation process presented in the previous work, the neighboring information between points is necessary. Therefore, a nearest neighbors search scheme, considering the local complexity of the processing point, is used to determinate the neighbors for each point in a point geometry. With the constructed virtual connectivity, the secret message can be embedded successfully after the vertex decimation and data embedding processes. The experimental results show that the proposed algorithm can preserve the advantages of previous work, including higher estimation accuracy, high embedding capacity, acceptable model distortion, and robustness against similarity transformation attacks. Most importantly, this work is the first 3D steganographic algorithm for point geometry with adaptation.展开更多
Although the distance between binary codes can be computed fast in Hamming space, linear search is not practical for large scale datasets. Therefore attention has been paid to the efficiency of performing approximate ...Although the distance between binary codes can be computed fast in Hamming space, linear search is not practical for large scale datasets. Therefore attention has been paid to the efficiency of performing approximate nearest neighbor search, in which hierarchical clustering trees (HCT) are widely used. However, HCT select cluster centers randomly and build indexes with the entire binary code, this degrades search performance. In this paper, we first propose a new clustering algorithm, which chooses cluster centers on the basis of relative distances and uses a more homogeneous partition of the dataset than HCT has to build the hierarchical clustering trees. Then, we present an algorithm to compress binary codes by extracting distinctive bits according to the standard deviation of each bit. Consequently, a new index is proposed using compressed binary codes based on hierarchical decomposition of binary spaces. Experiments conducted on reference datasets and a dataset of one billion binary codes demonstrate the effectiveness and efficiency of our method.展开更多
The continuous top-t most influential place (CTtMIP) query is defined formally and solved efficiently in this paper. A CTtMIP query continuously monitors the t places with the maximum influence from the set of place...The continuous top-t most influential place (CTtMIP) query is defined formally and solved efficiently in this paper. A CTtMIP query continuously monitors the t places with the maximum influence from the set of places, where the influence of a place is defined as the number of its bichromatic reverse k nearest neighbors (BRkNNs). Two new metrics and their corresponding rules are introduced to shrink the search region and reduce the candidates of BRkNNs checked. Extensive experiments confirm that our proposed approach outperforms the state-of-the-art competitor significantly.展开更多
The Markov chain random field(MCRF)model is a spatial statistical approach for modeling categorical spatial variables in multiple dimensions.However,this approach tends to be computationally costly when dealing with l...The Markov chain random field(MCRF)model is a spatial statistical approach for modeling categorical spatial variables in multiple dimensions.However,this approach tends to be computationally costly when dealing with large data sets because of its sequential simulation processes.Therefore,improving its computational efficiency is necessary in order to run this model on larger sizes of spatial data.In this study,we suggested four parallel computing solutions by using both central processing unit(CPU)and graphics processing unit(GPU)for executing the sequential simulation algorithm of the MCRF model,and compared them with the nonparallel computing solution on computation time spent for a land cover post-classification.The four parallel computing solutions are:(1)multicore processor parallel computing(MP),(2)parallel computing by GPU-accelerated nearest neighbor searching(GNNS),(3)MP with GPU-accelerated nearest neighbor searching(MPGNNS),and(4)parallel computing by GPU-accelerated approximation and GPU-accelerated nearest neighbor searching(GA-GNNS).Experimental results indicated that all of the four parallel computing solutions are at least 1.8×faster than the nonparallel solution.Particularly,the GA-GNNS solution with 512 threads per block is around 83×faster than the nonparallel solution when conducting a land cover post-classification with a remotely sensed image of 1000×1000 pixels.展开更多
The paper proposes a novel symmetrical encoding-based index structure, which is called EDD-tree (for encoding-based dual distance tree), to support fast k-nearest neighbor (k-NN) search in high-dimensional spaces....The paper proposes a novel symmetrical encoding-based index structure, which is called EDD-tree (for encoding-based dual distance tree), to support fast k-nearest neighbor (k-NN) search in high-dimensional spaces. In the EDD-tree, all data points are first grouped into clusters by a k-means clustering algorithm. Then the uniform ID number of each data point is obtained by a dual-distance-driven encoding scheme, in which each cluster sphere is partitioned twice according to the dual distances of start- and centroid-distance. Finally, the uniform ID number and the centroid-distance of each data point are combined to get a uniform index key, the latter is then indexed through a partition-based B^+-tree. Thus, given a query point, its k-NN search in high-dimensional spaces can be transformed into search in a single dimensional space with the aid of the EDD-tree index. Extensive performance studies are conducted to evaluate the effectiveness and efficiency of our proposed scheme, and the results demonstrate that this method outperforms the state-of-the-art high-dimensional search techniques such as the X-tree, VA-file, iDistance and NB-tree, especially when the query radius is not very large.展开更多
基金supported by National Natural Science Foundation of China(61403244,61304031)Key Project of Science and Technology Commission of Shanghai Municipality(14JC1402200)+3 种基金the Shanghai Municipal Commission of Economy and Informatization under Shanghai Industry-University-Research Collaboration(CXY-2013-71)the Science and Technology Commission of Shanghai Municipality under’Yangfan Program’(14YF1408600)National Key Scientific Instrument and Equipment Development Project(2012YQ15008703)Innovation Program of Shanghai Municipal Education Commission(14YZ007)
文摘Dataset classification is an essential fundament of computational intelligence in cyber-physical systems(CPS).Due to the complexity of CPS dataset classification and the uncertainty of clustering number,this paper focuses on clarifying the dynamic behavior of acceleration dataset which is achieved from micro electro mechanical systems(MEMS)and complex image segmentation.To reduce the impact of parameters uncertainties with dataset classification,a novel robust dataset classification approach is proposed based on neighbor searching and kernel fuzzy c-means(NSKFCM)methods.Some optimized strategies,including neighbor searching,controlling clustering shape and adaptive distance kernel function,are employed to solve the issues of number of clusters,the stability and consistency of classification,respectively.Numerical experiments finally demonstrate the feasibility and robustness of the proposed method.
文摘An innovative damage identification method using the nearest neighbor search method to assess 3D structures is presented.The frequency response function was employed as the input parameters to detect the severity and place of damage in 3D spaces since it includes the most dynamic characteristics of the structures.Two-dimensional principal component analysis was utilized to reduce the size of the frequency response function data.The nearest neighbor search method was employed to detect the severity and location of damage in different damage scenarios.The accuracy of the approach was verified using measured data from an experimental test;moreover,two asymmetric 3D numerical examples were considered as the numerical study.The superiority of the method was demonstrated through comparison with the results of damage identification by using artificial neural network.Different levels of white Gaussian noise were used for polluting the frequency response function data to investigate the robustness of the methods against noise-polluted data.The results indicate that both methods can efficiently detect the damage properties including its severity and location with high accuracy in the absence of noise,but the nearest neighbor search method is more robust against noisy data than the artificial neural network.
基金Project supported by the National Natural Science Foundation of China(Grant No.11002086)the Shanghai Leading Academic Discipline Project(Grant No.J50103)
文摘The core of smoothed particle hydrodynamics (SPH) is the nearest neighbor search subroutine. In this paper, a nearest neighbor search algorithm which is based on multiple background grids and support variable smooth length is introduced. Through tested on lid driven cavity flow, it is clear that this method can provide high accuracy. Analysis and experiments have been made on its parallelism, and the results show that this method has better parallelism and with adding processors its accuracy become higher, thus it achieves that efficiency grows in pace with accuracy.
基金supported by the National Science Council under Grant No. NSC98-2221-E-468-017 and NSC 100-2221-E-468-023the Research Project of Asia University under Grant No. 100-A-04
文摘In this study, we extend our previous adaptive steganographic algorithm to support point geometry. For the purpose of the vertex decimation process presented in the previous work, the neighboring information between points is necessary. Therefore, a nearest neighbors search scheme, considering the local complexity of the processing point, is used to determinate the neighbors for each point in a point geometry. With the constructed virtual connectivity, the secret message can be embedded successfully after the vertex decimation and data embedding processes. The experimental results show that the proposed algorithm can preserve the advantages of previous work, including higher estimation accuracy, high embedding capacity, acceptable model distortion, and robustness against similarity transformation attacks. Most importantly, this work is the first 3D steganographic algorithm for point geometry with adaptation.
文摘Although the distance between binary codes can be computed fast in Hamming space, linear search is not practical for large scale datasets. Therefore attention has been paid to the efficiency of performing approximate nearest neighbor search, in which hierarchical clustering trees (HCT) are widely used. However, HCT select cluster centers randomly and build indexes with the entire binary code, this degrades search performance. In this paper, we first propose a new clustering algorithm, which chooses cluster centers on the basis of relative distances and uses a more homogeneous partition of the dataset than HCT has to build the hierarchical clustering trees. Then, we present an algorithm to compress binary codes by extracting distinctive bits according to the standard deviation of each bit. Consequently, a new index is proposed using compressed binary codes based on hierarchical decomposition of binary spaces. Experiments conducted on reference datasets and a dataset of one billion binary codes demonstrate the effectiveness and efficiency of our method.
基金Supported by the National Natural Science Foundation of China (61003049)the Natural Science Foundation of Zhejiang Province (Y110278, 2010QNA5051)Zheda Zijin Plan
文摘The continuous top-t most influential place (CTtMIP) query is defined formally and solved efficiently in this paper. A CTtMIP query continuously monitors the t places with the maximum influence from the set of places, where the influence of a place is defined as the number of its bichromatic reverse k nearest neighbors (BRkNNs). Two new metrics and their corresponding rules are introduced to shrink the search region and reduce the candidates of BRkNNs checked. Extensive experiments confirm that our proposed approach outperforms the state-of-the-art competitor significantly.
基金supported in part by the U.S.National Science Foundation[grant number 1414108]Division of Behavioral and Cognitive Sciences.
文摘The Markov chain random field(MCRF)model is a spatial statistical approach for modeling categorical spatial variables in multiple dimensions.However,this approach tends to be computationally costly when dealing with large data sets because of its sequential simulation processes.Therefore,improving its computational efficiency is necessary in order to run this model on larger sizes of spatial data.In this study,we suggested four parallel computing solutions by using both central processing unit(CPU)and graphics processing unit(GPU)for executing the sequential simulation algorithm of the MCRF model,and compared them with the nonparallel computing solution on computation time spent for a land cover post-classification.The four parallel computing solutions are:(1)multicore processor parallel computing(MP),(2)parallel computing by GPU-accelerated nearest neighbor searching(GNNS),(3)MP with GPU-accelerated nearest neighbor searching(MPGNNS),and(4)parallel computing by GPU-accelerated approximation and GPU-accelerated nearest neighbor searching(GA-GNNS).Experimental results indicated that all of the four parallel computing solutions are at least 1.8×faster than the nonparallel solution.Particularly,the GA-GNNS solution with 512 threads per block is around 83×faster than the nonparallel solution when conducting a land cover post-classification with a remotely sensed image of 1000×1000 pixels.
基金the key program of the National Natural Science Foundation of China (Grant No.60533090)the National Natural Science Fund for Distinguished Young Scholars (Grant No.60525108)China-America Academic Digital Library Project
文摘The paper proposes a novel symmetrical encoding-based index structure, which is called EDD-tree (for encoding-based dual distance tree), to support fast k-nearest neighbor (k-NN) search in high-dimensional spaces. In the EDD-tree, all data points are first grouped into clusters by a k-means clustering algorithm. Then the uniform ID number of each data point is obtained by a dual-distance-driven encoding scheme, in which each cluster sphere is partitioned twice according to the dual distances of start- and centroid-distance. Finally, the uniform ID number and the centroid-distance of each data point are combined to get a uniform index key, the latter is then indexed through a partition-based B^+-tree. Thus, given a query point, its k-NN search in high-dimensional spaces can be transformed into search in a single dimensional space with the aid of the EDD-tree index. Extensive performance studies are conducted to evaluate the effectiveness and efficiency of our proposed scheme, and the results demonstrate that this method outperforms the state-of-the-art high-dimensional search techniques such as the X-tree, VA-file, iDistance and NB-tree, especially when the query radius is not very large.