Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Journal of Naval University of Engineering》 2018-01
Add to Favorite Get Latest Update

Target recognition of anti-ship missile based on meta-learning

HU Sheng-liang;FAN Xue-man;HE Jing-bo;College of Electronics Engineering,Naval Univ.of Engineering;  
In order to improve the correct recognition rate and the generalization ability of anti-ship missile under interference environment,a target recognition method based on meta-learning with stacked generalization strategy is proposed,in which a meta-level learner is constructed to rectify the error classification and consolidate the correct classification of base learners by learning the results of base learners.And then the decision tree is selected as base learning algorithm to construct homogeneous multiple classifiers system,and based on the self-built full polarization HRRP dataset,the influence of meta attribute vector,meta learning algorithm and the number of base classifiers on classification accuracy is studied in depth.Finally,the feasibility and effectiveness of the proposed method are verified by comparison with single classifier and commonly used ensemble algorithm.
【Fund】: 国家自然科学基金资助项目(61401493);; 国家部委基金资助项目(9140A01010415JB11002)
【CateGory Index】: TJ761.14;TP181
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved