马兰,陈筱勇,邓国臣

• 1:信息工程大学
• 2:国际关系学院
• 3:中国测绘科学研究院

摘要(Abstract)：

针对低对比度、条纹噪声、低空间分辨率等特点而导致的热红外图像识别效果不佳问题,提出了一种港口目标热红外遥感图像特征提取与选择方法,实现了一定情况下港口目标的高精度分类。采用纹理、几何等29个特征,通过评估器选择最佳特征组合,并根据识别精度选择最佳分类器,能生成热红外图像港口目标22个最佳分类特征,且具有一定的鲁棒性。经过参数优化后的libSVM(一种支持向量机)分类器分类精度较高;白天图像比夜间图像分类精度更高;像素值、灰度直方图相关的一维和二维统计特征、局部二进制模式特征、边缘方向直方图特征等与灰度和纹理相关的特征对港口目标热红外图像识别影响较大。

关键词(KeyWords)： 特征提取;特征选择;热红外图像;港口目标;评估器;分类器;热红外遥感

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 马兰,陈筱勇,邓国臣

DOI: 10.16251/j.cnki.1009-2307.2017.05.015

参考文献(References)：

• [1]WILLIAM L,FEHLMAN II,HINDERS M K.Mobile robot navigation with intelligent infrared image interpretation[M].London,New York:Springer,Dordrecht,Heidelberg,2009:70-72.
• [2]高炳像.红外图像目标识别中的关键技术研究[D].杭州:杭州电子科技大学,2012:18-32.(GAO Bingxiang.Research on key technologies in infrared image target recognition[D].Hangzhou:Hangzhou Dianzi University,2012:18-32.)
• [3]GOUAILLIER V,GAGNON L.Ship silhouette recognition using principal components analysis[J].Proceedings of SPIE,the International Society for Optical Engineering,1997,3164:59-69.
• [4]WITHAGEN P,SCHUTTE K,VOSSEPOEL A,et al.Automatic classification of ships from infrared images[J].Proceedings of SPIE,the International Society for Optical Engineering,1999,3720:180-187.
• [5]ALVES J,HERMAN J,ROWE N C.Robust recognition of ship types from an infrared silhouette[C]//Command and Control Research and Technology Symposium,SanDiego,California,2004.
• [6]张铮,王艳平,薛桂香.数字图像处理与机器视觉:Visual C++与Matlab实现[M].北京:人民邮电出版社,2013:407-411.(ZHANG Zheng,WANG Yanping,XUE Guixiang.Digital image processing and machine vision:Programming implementation with visual C++and matlab[M].Beijing:Posts and Telecommunications Press,2013:407-411.)
• [7]胡小锋,赵辉.Visual C++/MATLAB图像处理与识别实用案例精选[M].北京:人民邮电出版社,2004:123-124.(HU Xiaofeng,ZHAO Hui.Practical and selected cases for image processing and identification with visual C++/matlab[M].Beijing:Posts and Telecommunications Press,2004:123-124.)
• [8]杨帆,王志陶,张华.精通图像处理经典算法(matlab版)[M].北京:北京航空航天大学出版社,2014:257-258.(YANG Fan,WANG Zhitao,ZHANG Hua.Proficient in image processing algorithms(matlab Edition)[M].Beijing:Beihang University Press,2014:257-258.)
• [9]BAY H,ESS A,TUYTELAARS T,et al.SURF:Speeded up robust features[J].Computer vision and image understanding(CVIU),2008,110(3):346-359.
• [10]王鑫.复杂背景下红外目标检测与跟踪算法研究[D].南京:南京理工大学,2010:29-33.(WANG Xin.Research on infrared target detection and tracking algorithms under complex background[D].Nanjing:Nanjing University of Science and Technology,2010:29-33.)
• [11]DE-LA-ROSA J J G,MUOZ A M.Higher-order cumulants and spectral kurtosis for early detection of subterranean termites[J].Mechanical Systems and Signal Processing,2008,22(2):279-294.
• [12]WITTEN I H,FRANK E,HALL M A.Data mining:Practical machine learning tools and techniques,third edition[M].[S.l.]:[s.n.],Elsevier Inc,2011:219-222.
• [13]袁梅宇.数据挖掘与机器学习:WEKA应用技术与实践[M].北京:清华大学出版社,2014:113-123.(YUAN Meiyu.Data mining and machine learning:WEKA application technology and practice[M].Beijing:Tsinghua University Press,2014:113-123.)
• [14]杨树文,董玉森,罗小波,等.遥感数字图像处理与分析:ENVI 5.x实验教程[M].北京:电子工业出版社,2015:24-28.(YANG Shuwen,DONG Yusen,LUO Xiaobo,et al.Remote sensing digital image processing and analysis-5.x ENVI experimental course[M].Beijing:Publishing House of Electronics Industry,2015:24-28.)
• [15]徐海祥.基于支持向量机方法的图像分割与目标分类[D].武汉:华中科技大学,2005:67-68.(XU Haixiang.Image segmentation and object classification based on support vector machines[D].Wuhan:Huazhong University of Science and Technology,2005:67-68.)