Lattice structure based metric for feature data fusion

ABSTRACT For stably reducing data redundancy which has negative impact on data processing, this paper integrates the ideas from lattice structure and develops a lattice structure based metric for feature fusion. The metric distance between two lattice nodes, which is a key factor for the lattice structure based metric, is determined by the shortest path length from these two nodes to their nearest descendant node in lattice structure. Then, based on the metrics between lattice nodes, the source feature dataset is divided into different subsets following the principles of minimum metric precedence and gradual expansion. The duplicate feature data in a subset is fused into an object data unit by using the median operation. The experimental results show that the proposed feature fusion method can better improve the completeness and conciseness of the existing feature data stably and has wide range of applications in the related fields on data processing.

[1]  Santiago Mazuelas,et al.  Adaptive Data Fusion for Wireless Localization in Harsh Environments , 2012, IEEE Transactions on Signal Processing.

[2]  Xiaoming Zhang,et al.  Feature Fusion Using Locally Linear Embedding for Classification , 2010, IEEE Transactions on Neural Networks.

[3]  Eleanor G. Rieffel,et al.  J an 2 00 0 An Introduction to Quantum Computing for Non-Physicists , 2002 .

[4]  W. Wootters.  Statistical distance and Hilbert space , 1981 .

[5]  Stephen P. Boyd,et al.  A space-time diffusion scheme for peer-to-peer least-squares estimation , 2006, 2006 5th International Conference on Information Processing in Sensor Networks.

[6]  K. Mehrotra,et al.  A clustering-based discretization for supervised learning , 2010 .

[7]  Theodoros Damoulas,et al.  Pattern recognition with a Bayesian kernel combination machine , 2009, Pattern Recognit. Lett..

[8]  Guangfeng Lin,et al.  Heterogeneous feature structure fusion for classification , 2016, Pattern Recognit..

[9]  Peter Willett,et al.  Bayesian Data Fusion for Distributed Target Detection in Sensor Networks , 2010, IEEE Transactions on Signal Processing.

[10]  Huifang Deng,et al.  Quantum inspired method of feature fusion based on von Neumann entropy , 2014, Inf. Fusion.

[11]  G. Gibbons.  Typical states and density matrices , 1992 .

[12]  Tad Hogg.  Quantum Computing and Phase Transitions in Combinatorial Search , 1996, J. Artif. Intell. Res..

[13]  D. Deutsch,et al.  Rapid solution of problems by quantum computation , 1992, Proceedings of the Royal Society of London. Series A: Mathematical and Physical Sciences.

[14]  Yunhong Wang,et al.  Automatic object classification using motion blob based local feature fusion for traffic scene surveillance , 2012, Frontiers of Computer Science.

[15]  Huifang Deng,et al.  A Collision and Reaction Model of Feature Fusion: Mechanism and Realization , 2015, IEEE Intelligent Systems.

[16]  D. J. Newman,et al.  UCI Repository of Machine Learning Database , 1998 .

[17]  Aihong Chen,et al.  Feature fusion based on Wootters metric , 2016, Int. J. Syst. Sci..

[18]  A. Uhlmann.  The Metric of Bures and the Geometric Phase. , 1992 .

[19]  Weizhong Yan,et al.  Fusion in multi-criterion feature ranking , 2007, 2007 10th International Conference on Information Fusion.

[20]  Zheyao Wang,et al.  Motion Measurement Using Inertial Sensors, Ultrasonic Sensors, and Magnetometers With Extended Kalman Filter for Data Fusion , 2012, IEEE Sensors Journal.

[21]  Elena Console,et al.  Data Fusion , 2009, Encyclopedia of Database Systems.

[22]  Lambert Schomaker,et al.  Variants of the Borda count method for combining ranked classifier hypotheses , 2000 .