Please wait a minute...
J4  2012, Vol. 46 Issue (11): 2081-2088    DOI: 10.3785/j.issn.1008-973X.2012.11.021
Feature fusion algorithm based on two-dimensional
feature matrix
BAO Bi-sai1, WU Jian-rong 1, LOU Xiao-jun1, LIU Hai-tao1, 2
1. Key Lab of Wireless Sensor Network and Communication, Shanghai Institute of Microsystem and Information
Technology, Chinese Academy of Science, Shanghai 200050, China;
2. Wuxi SensingNet Industrialization Research Institute, Wuxi 214135, China
Download:   PDF(0KB) HTML
Export: BibTeX | EndNote (RIS)      


A strategy of two-dimensional feature fusion (2DFF) was proposed to improve efficiency of information fusion in wireless sensor networks (WSNs). The feature sets after normalized of multi-sensors were combined into a two-dimensional feature matrix. The techniques of image compression, twodimensional principal component analysis (2DPCA) and MatPCA were generalized for feature extraction from two-dimensional feature matrix to achieve multi-sensors feature fusion. The inherent essence of this method used in feature fusion was analyzed further in theory. Compared to traditional feature fusion method, this method can obtain more accurate fused feature and improve efficiency of information fusion. The experiment results on the real signals of ground targets show that this method can increase classification accuracy and reduce computational complexity.

Published: 11 December 2012
CLC:  TP 391  
Cite this article:

BAO Bi-sai, WU Jian-rong , LOU Xiao-jun, LIU Hai-tao. Feature fusion algorithm based on two-dimensional
feature matrix. J4, 2012, 46(11): 2081-2088.

URL:     OR


为了提高无线传感器网络信息融合的效率, 提出一种多传感器二维特征融合(2DFF)策略. 将多个传感器标准化后的特征集组合成二维特征矩阵, 引入图像压缩技术, 包括二维主成分分析(2DPCA)及MatPCA对特征矩阵进行特征提取, 实现特征融合. 从理论上剖析该方法之所以能够有效地适用于特征融合, 且区别于传统方法的内在本质. 相比传统的特征融合方法, 该方法能够获得更加精确的融合特征, 提高信息融合的效率. 基于实地采集的地面目标信号的实验结果表明,该方法既提高目标识别率, 又降低了计算复杂度.

[1] HALL D L, LLINAS J. Handbook of multisensor data fsuion [M]. New York: CRC Press, 2001: 33-66.
[2] MALHOTRA B, NIKOLAIDIS I, HARMS J. Distributed classification of acoustic taargets in wireless audiosensor networks  \
[J\]. Computer Networks, 2008, 52(13): 2582-2593.
[3] BARDWAJ A A, ANANDARAJ M, KAPIL K, et al. Multi sensor data fusion methods using sensor data compression and estimated weights[C]∥ International Conference on Signal Processing, Communications and Networking, 2008. ICSCN’08. Chennai India:Madras Institute of Technology, Anna University,   2008,(4/6): 250-254.
[4] KUNCHEVA L I, BEZDEK J C, DUIN R P W. Decision templates for multiple classifier fusion: an experimental comparison \
[J\]. Pattern Recognition, 2001, 34(2): 299-314.
[5] PAN Qiang, WEI Jianming, CAO Hongbin, et al. Improved DS acousticseismic modality fusion for groundmoving target classification in wireless sensor networks  \
[J\]. Pattern Recognition Letters, 2007, 28(16): 2419-2426.
[6] 曹红兵, 魏建明, 刘海涛. 无线传感网中多传感器特征融合算法研究 \
[J\]. 电子与信息学报, 2010, 32(1): 166-171.
CAO Hongbin, WEI Jianming, LIU Haitao. Research on multiSensor feature fusion algorithms in wireless sensor networks  \
[J\]. Journal of Electronics & Information Technology, 2010, 32(1):166-171.
[7] ARNAZ M, ROBERT X G. PCAbased feature selection scheme for machine defect classification  \
[J\]. IEEE Transactions on Instrumentation and Measurement, 2004, 53(6): 1517-1525.
[8] TURBANSAYAN G. Real time electromagnetic target classification using a novel feature extraction technique with PCAbased fusion  \
[J\]. IEEE Transactions on Antennas and Propagation,  2005, 53(2): 766-776.
[9] MICHAEL B R, INDRA A. Classification and regionalization through kernel principal component analysis  \
[J\]. Physics and Chemistry of the Earth, 2010(35): 316-328.
[10] SUN Bingyu, ZHANG Xiaoming, LI Jiuyong, et al. Feature fusion using locally linear embedding for classification  \
[J\]. IEEE Transactions on Neural Networks, 2010, 21(1): 163-168.
[11] YANG Jian, YANG Jingyu. Generalized K–L transform based combined feature extraction  \
[J\]. Pattern Recognition, 2002, 35(1): 295-297.
[12] YANG Jian, YANG Jingyu, ZHANG David, et al. Feature fusion: parallel strategy vs. serial strategy  \
[J\]. Pattern Recognition, 2003, 36(6): 1369-1381.
[13] SUN Quansen, ZENG Shenggen, LIU Yan, et al. A new method of feature fusion and its application in image recognition  \
[J\]. Pattern Recognition, 2005, 38(12): 2437-2448.
[14] YANG Jian, ZHANG David, Alejandro F F, et al. Twodimensional PCA: a new approach to appearancebased face representation and recognition  \
[J\]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(1): 131-137.
[15] CHEN Songcan, ZHU Yulian, ZHANG Daoqiang, et al. Feature extraction approaches based on matrix pattern: MatPCA and MatFLDA  \
[J\]. Pattern Recognition Letters, 2005,26(8): 1157-1167.
[16] YANG Jian, YANG Jingyu. From image vector to matrix: a straightforward image projection techniqueIMPCA vs. PCA  \
[J\]. Pattern Recognition, 2002, 35(9): 1997-1999.
[17] 李舜酩, 李香莲. 振动信号的现代分析技术与应用[M]. 1版.北京: 国防工业出版社, 2008: 57-117.
[18] 聂为荣, 朱继南. 运动目标地震动信号的时频特征分析 \
[J\]. 南京理工大学学报, 2002, 26(5): 478-481.
NIE Weirong, ZHU Jinan. Eigenvector Analysis in Time and frequency of seismic signals of moving target  \
[J\]. Journal of Nanjing University of Science and Technology, 2002, 26(5): 478-481.

[1] ZHAO Jian-jun, WANG Yi, YANG Li-bin. Threat assessment method based on time series forecast[J]. J4, 2014, 48(3): 398-403.
[2] CUI Guang-mang, ZHAO Ju-feng,FENG Hua-jun, XU Zhi-hai,LI Qi, CHEN Yue-ting. Construction of fast simulation model for degraded image by inhomogeneous medium[J]. J4, 2014, 48(2): 303-311.
[3] ZHANG Tian-yu, FENG Hua-jun, XU Zhi-hai, LI Qi, CHEN Yue-ting. Sharpness metric based on histogram of strong edge width[J]. J4, 2014, 48(2): 312-320.
[4] LIU Zhong, CHEN Wei-hai, WU Xing-ming, ZOU Yu-hua, WANG Jian-hua. Salient region detection based on stereo vision[J]. J4, 2014, 48(2): 354-359.
[5] WANG Xiang-bing,TONG Shui-guang,ZHONG Wei,ZHANG Jian. Study on  scheme design technique for hydraulic excavator's structure performance based on extension reuse[J]. J4, 2013, 47(11): 1992-2002.
[6] WANG Jin, LU Guo-dong, ZHANG Yun-long. Quantification-I theory based IGA and its application[J]. J4, 2013, 47(10): 1697-1704.
[7] LIU Yu, WANG Guo-jin. Designing  developable surface pencil through  given curve as its common asymptotic curve[J]. J4, 2013, 47(7): 1246-1252.
[8] HU Gen-sheng, BAO Wen-xia, LIANG Dong, ZHANG Wei. Fusion of panchromatic image and multi-spectral image based on
SVR and Bayesian method 
[J]. J4, 2013, 47(7): 1258-1266.
[9] WU Jin-liang, HUANG Hai-bin, LIU Li-gang. Texture details preserving seamless image composition[J]. J4, 2013, 47(6): 951-956.
[10] CHEN Xiao-hong,WANG Wei-dong. A HDTV video de-noising algorithm based on spatial-temporal filtering[J]. J4, 2013, 47(5): 853-859.
[11] ZHU Fan , LI Yue, JIANG Kai, YE Shu-ming, ZHENG Xiao-xiang. Decoding of rat’s primary motor cortex by partial least square[J]. J4, 2013, 47(5): 901-905.
[12] WU Ning, CHEN Qiu-xiao, ZHOU Ling, WAN Li. Multi-level method of optimizing vector graphs converted from remote sensing images[J]. J4, 2013, 47(4): 581-587.
[13] JI Yu, SHEN Ji-zhong, SHI Jin-he. Automatic ocular artifact removal based on blind source separation[J]. J4, 2013, 47(3): 415-421.
[14] WANG Xiang, DING Yong. Full reference image quality assessment based on Gabor filter[J]. J4, 2013, 47(3): 422-430.
[15] TONG Shui-guang, WANG Xiang-bing, ZHONG Wei, ZHANG Jian. Dynamic optimization design for rigid landing leg of crane
based on BP-HGA
[J]. J4, 2013, 47(1): 122-130.