自适应模糊C均值聚类的数据融合算法
|
摘要
针对基于改进模糊聚类的数据融合算法存在融合不精确、融合可信度较低等不足,为了解决多个同质传感器在无先验知识的情况下对同一个目标的某一特征进行测量的数据融合问题,提出了一种自适应模糊C均值聚类的数据融合算法,主要是把自适应模糊C均值聚类应用到数据融合中。该算法首先在改进的模糊聚类中通过引入自适应系数以发现不同形状和大小的聚类子集,使得融合结果更精确;其次将卡尔曼滤波原理和基于多层感知机的神经网络预测法应用到误差协方差估计中,提高了融合可信度。实验结果表明,与7种经典数据融合算法进行对比,该算法在4个模拟数据集与真实数据集上融合结果较好,特别在判别函数与融合误差方面优势更为明显。
For data fusion algorithm based on improved fuzzy clustering, there are some disadvantages such as inaccurate fusion and low reliability of fusion. In order to solve the data fusion problem of multiple homogenous sensors measuring a certain feature of the same target without prior knowledge, this paper presents a data fusion algorithm based on adaptive fuzzy C-means clustering, which mainly applies adaptive fuzzy C-means clustering to data fusion. The algorithm firstly introduces adaptive coefficients to find cluster subsets of different shapes and sizes in improved fuzzy clustering, making the fusion result more accurate. Secondly, Kalman filtering principle and neural network prediction method based on multilayer perceptron are applied to the error covariance estimation, which improves the credibility of the fusion. The experimental results show that compared with the four classical data fusion algorithms, the algorithm has better results in the fusion of the four simulated data sets with the real data sets, and the advantages are particularly obvious in criterion functions and fusion errors.
For data fusion algorithm based on improved fuzzy clustering, there are some disadvantages such as inaccurate fusion and low reliability of fusion. In order to solve the data fusion problem of multiple homogenous sensors measuring a certain feature of the same target without prior knowledge, this paper presents a data fusion algorithm based on adaptive fuzzy C-means clustering, which mainly applies adaptive fuzzy C-means clustering to data fusion. The algorithm firstly introduces adaptive coefficients to find cluster subsets of different shapes and sizes in improved fuzzy clustering, making the fusion result more accurate. Secondly, Kalman filtering principle and neural network prediction method based on multilayer perceptron are applied to the error covariance estimation, which improves the credibility of the fusion. The experimental results show that compared with the four classical data fusion algorithms, the algorithm has better results in the fusion of the four simulated data sets with the real data sets, and the advantages are particularly obvious in criterion functions and fusion errors.
引文
[1]魏秀蓉.无线传感器网络数据融合研究综述[J].无线互联科技,2015(14):22-26.
[2] Feng B,Chen B,Liu H.Radar HRRP target recognition withdeep networks[J].Pattern Recognition,2017,61:379-393.
[3] Xia Y,Xing Z.Comparison of centralised scaled unscentedKalman filter and extended Kalman filter for multisensordata fusion architectures[J].IET Signal Processing,2016,10(4):359-365.
[4]吴志奇.VTS多雷达目标融合技术[D].辽宁大连:大连海事大学,2017.
[5] Jing Hang,Yu Xiaoning.Dust concentration data fusionalgorithm based on DS evidence theory[J].Computer andDigital Engineering,2017.
[6] Singh A K,Purohit N,Goutele S,et al.An energy efficientapproach for clustering in WSN using fuzzy logic[J].International Journal of Computer Applications,2012,44(18):8-12.
[7] Jing Luyang,Wang Taiyong,Zhao Ming,et al.An adaptivemulti-sensor data fusion method based on deep convolu-tional neural networks for fault diagnosis of planetarygearbox[J].Sensors,2017,17(2):414-418.
[8] Yang Z,Chen M R,Wu W.Algorithm for wireless sensornetwork data fusion based on radial basis function neuralnetworks[J].Applied Mechanics and Materials,2014,577:873-878.
[9] Wang J,Acharya S,Kam M.Adaptive decision fusion usinggenetic algorithm[C]//Information Science and Systems,2016:401-406.
[10] Peng H,Cao X.Research conflict problems of DS evi-dence and its application in multi-sensor informationfusion technology[C]//International Conference on Infor-mation Theory and Information Security,2011.
[11] Chen B,Qi W,Yuan J,et al.Recognition of high-voltagecable partial discharge signal based on adaptive fuzzyC-means clustering[J].International Journal of PatternRecognition and Artificial Intelligence,2017,31(6):9-15.
[12] Rathore P,Ghafoori Z,Bezdek J C,et al.ApproximatingDunn’s cluster validity indices for partitions of bigdata[J].IEEE Transactions on Cybernetics,2018,99:1-13.
[13] Shen H Y,Feng Y M.Clustering adaptive weightedfusion algorithm of wireless sensor network for railwayembankment monitor[J].Journal of Transportation Sys-tems Engineering and Information Technology,2010,10(6):190-194.
[14] Shi Y,Nan J.A cluster validity index based on fuzzyhybrid hierarchical clustering[J].Journal of Computa-tional Theoretical Nanoscience,2016,13(3):2157-2161.
[15] Sara?li S,Do?an N,Do?an?.Comparison of hierarchi-cal cluster analysis methods by copheneticcorrelation[J].Journal of Inequalities and Applications,2013,15(1):203-209.
[16]高新波,裴继红,谢维信.模糊c-均值聚类算法中加权指数m的研究[J].电子学报,2000,28(4):80-83.
[17]刘宜平,沈毅,刘志言.一种FCM聚类算法的改进与优化[J].系统工程与电子技术,2000,22(4):1-4.
[18]苏卫星,朱云龙,刘芳,等.基于改进模糊聚类的同构多传感器在线数据融合方法[J].信息与控制,2015,44(5):557-563.
[19] Alyannezhadi M M,Pouyan A A,Abolghasemi V.Anefficient algorithm for multisensory data fusion underuncertainty condition[J].Journal of Electrical Systemsand Information Technology,2016.
[20] Zhao T,Peng X,Yu P,et al.Lithium-ion battery SOCestimation method with fusion improved Kalman filteralgorithm[J].Chinese Journal of Scientific Instrument,2016.
[21] Cappello F,Ramasamy S,Sabatini R,et al.Low-costsensors based multi-sensor data fusion techniques forRPAS navigation and guidance[C]//International Con-ference on Unmanned Aircraft Systems,2015.
[22]光俊叶,刘明霞,张道强.有效距离在聚类算法中的应用[J].计算机科学与探索,2017,11(3):406-413.
[23]刘沧生,许青林.基于密度峰值优化的模糊C均值聚类算法[J].计算机工程与应用,2018,54(14):153-157.
[24] Verma H,Agrawal R K,Sharan A.An improved intu-itionistic fuzzy c-means clustering algorithm incorporatinglocal information for brain image segmentation[J].AppliedSoft Computing,2016,46(6):543-557.
[25] Abdulhafiz W A,Alaa K.Handling data uncertaintyand inconsistency using multisensor data fusion[J].Advances in Artificial Intelligence,2013,11(3):1-11.
[26] Carvalho F D A T D,Barbosa G B N,Pimentel J T.Partitioning fuzzy C-means clustering algorithms forinterval-valued data based on city-block distances[C]//Intelligent Systems,2014.
[27] Zhang Y,Zeng C,Liang H,et al.A visual target trackingalgorithm based on improved kernelized correlationfilters[C]//International Conference on Mechatronics andAutomation,2016:199-204.
[28] Qin Xiaofei,Dai Shunfeng,Li Feng.Target trackingalgorithm based on improved kernel correlation filter[J].Measurement and Control Technology,2017.
[2] Feng B,Chen B,Liu H.Radar HRRP target recognition withdeep networks[J].Pattern Recognition,2017,61:379-393.
[3] Xia Y,Xing Z.Comparison of centralised scaled unscentedKalman filter and extended Kalman filter for multisensordata fusion architectures[J].IET Signal Processing,2016,10(4):359-365.
[4]吴志奇.VTS多雷达目标融合技术[D].辽宁大连:大连海事大学,2017.
[5] Jing Hang,Yu Xiaoning.Dust concentration data fusionalgorithm based on DS evidence theory[J].Computer andDigital Engineering,2017.
[6] Singh A K,Purohit N,Goutele S,et al.An energy efficientapproach for clustering in WSN using fuzzy logic[J].International Journal of Computer Applications,2012,44(18):8-12.
[7] Jing Luyang,Wang Taiyong,Zhao Ming,et al.An adaptivemulti-sensor data fusion method based on deep convolu-tional neural networks for fault diagnosis of planetarygearbox[J].Sensors,2017,17(2):414-418.
[8] Yang Z,Chen M R,Wu W.Algorithm for wireless sensornetwork data fusion based on radial basis function neuralnetworks[J].Applied Mechanics and Materials,2014,577:873-878.
[9] Wang J,Acharya S,Kam M.Adaptive decision fusion usinggenetic algorithm[C]//Information Science and Systems,2016:401-406.
[10] Peng H,Cao X.Research conflict problems of DS evi-dence and its application in multi-sensor informationfusion technology[C]//International Conference on Infor-mation Theory and Information Security,2011.
[11] Chen B,Qi W,Yuan J,et al.Recognition of high-voltagecable partial discharge signal based on adaptive fuzzyC-means clustering[J].International Journal of PatternRecognition and Artificial Intelligence,2017,31(6):9-15.
[12] Rathore P,Ghafoori Z,Bezdek J C,et al.ApproximatingDunn’s cluster validity indices for partitions of bigdata[J].IEEE Transactions on Cybernetics,2018,99:1-13.
[13] Shen H Y,Feng Y M.Clustering adaptive weightedfusion algorithm of wireless sensor network for railwayembankment monitor[J].Journal of Transportation Sys-tems Engineering and Information Technology,2010,10(6):190-194.
[14] Shi Y,Nan J.A cluster validity index based on fuzzyhybrid hierarchical clustering[J].Journal of Computa-tional Theoretical Nanoscience,2016,13(3):2157-2161.
[15] Sara?li S,Do?an N,Do?an?.Comparison of hierarchi-cal cluster analysis methods by copheneticcorrelation[J].Journal of Inequalities and Applications,2013,15(1):203-209.
[16]高新波,裴继红,谢维信.模糊c-均值聚类算法中加权指数m的研究[J].电子学报,2000,28(4):80-83.
[17]刘宜平,沈毅,刘志言.一种FCM聚类算法的改进与优化[J].系统工程与电子技术,2000,22(4):1-4.
[18]苏卫星,朱云龙,刘芳,等.基于改进模糊聚类的同构多传感器在线数据融合方法[J].信息与控制,2015,44(5):557-563.
[19] Alyannezhadi M M,Pouyan A A,Abolghasemi V.Anefficient algorithm for multisensory data fusion underuncertainty condition[J].Journal of Electrical Systemsand Information Technology,2016.
[20] Zhao T,Peng X,Yu P,et al.Lithium-ion battery SOCestimation method with fusion improved Kalman filteralgorithm[J].Chinese Journal of Scientific Instrument,2016.
[21] Cappello F,Ramasamy S,Sabatini R,et al.Low-costsensors based multi-sensor data fusion techniques forRPAS navigation and guidance[C]//International Con-ference on Unmanned Aircraft Systems,2015.
[22]光俊叶,刘明霞,张道强.有效距离在聚类算法中的应用[J].计算机科学与探索,2017,11(3):406-413.
[23]刘沧生,许青林.基于密度峰值优化的模糊C均值聚类算法[J].计算机工程与应用,2018,54(14):153-157.
[24] Verma H,Agrawal R K,Sharan A.An improved intu-itionistic fuzzy c-means clustering algorithm incorporatinglocal information for brain image segmentation[J].AppliedSoft Computing,2016,46(6):543-557.
[25] Abdulhafiz W A,Alaa K.Handling data uncertaintyand inconsistency using multisensor data fusion[J].Advances in Artificial Intelligence,2013,11(3):1-11.
[26] Carvalho F D A T D,Barbosa G B N,Pimentel J T.Partitioning fuzzy C-means clustering algorithms forinterval-valued data based on city-block distances[C]//Intelligent Systems,2014.
[27] Zhang Y,Zeng C,Liang H,et al.A visual target trackingalgorithm based on improved kernelized correlationfilters[C]//International Conference on Mechatronics andAutomation,2016:199-204.
[28] Qin Xiaofei,Dai Shunfeng,Li Feng.Target trackingalgorithm based on improved kernel correlation filter[J].Measurement and Control Technology,2017.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |