基于直方图的遥感图像相似性检索方法比较
|
摘要
相似性度量是用于研究多源数据之间相似程度的,是对空间数据进行模式识别的基础。通过单波段遥感图像的检索对两组直方图相似性检索方法进行了实验研究,即基于特征向量的相似性度量和基于概率的相似性度量。实验中发现第一组相似度量中有两种以往较少用于遥感图像检索的方法表现出色,它们分别是2χ统计距离和相似夹角余弦度量。第二组实验中,针对其中包含较明显的目标物体且背景较为单一的遥感图像(其直方图可看作混合高斯分布),在类别可分离判据的基础上,根据K-近邻法则提出了一种计算该类图像之间相似值的方法。实验结果表明基于K-近邻法则的计算方法行之有效。所得出的结论将对多源数据分析中相似性度量的理解与选择有积极意义。
Similarity measure is usually used to study the similar degree between multisource data,which is the basis of pattern recognition on spatial data.In this paper two kinds of similarity measures are experimentally investigated through some remote sensing image retrievals,they are feature vector based measures and probabilistic measures,accordingly two groups experiments are designed to compare the measures for application to remote sensing image retrieval.From the experiment results we find that in the first group two measures seldom used in the literature perform well,they are χ~2 statistical distance measure and cosine of the angle measure.And in the second group experiments,for computing the similarity degree of two images with their histograms obeying mixture Gaussian distributions,we present a method on the basis of class separability measures according to the K-nearest neighbor rule.The experiment results show that the method has good performance.We believe that the results described in this paper will be of significance in applications to multisource data analysis.
Similarity measure is usually used to study the similar degree between multisource data,which is the basis of pattern recognition on spatial data.In this paper two kinds of similarity measures are experimentally investigated through some remote sensing image retrievals,they are feature vector based measures and probabilistic measures,accordingly two groups experiments are designed to compare the measures for application to remote sensing image retrieval.From the experiment results we find that in the first group two measures seldom used in the literature perform well,they are χ~2 statistical distance measure and cosine of the angle measure.And in the second group experiments,for computing the similarity degree of two images with their histograms obeying mixture Gaussian distributions,we present a method on the basis of class separability measures according to the K-nearest neighbor rule.The experiment results show that the method has good performance.We believe that the results described in this paper will be of significance in applications to multisource data analysis.
引文
[1]Maxwell B A,Buddem eier R W.Coastal Typology D evelopm entw ith Heterogeneous Data Sets[J].Regional Environm entalChange,2002,3(1—3):77—87.
[2]Tian Y Q,Guo P,Lyu M R.Comparative Stud ies on FeatureExtraction M ethods for Mu ltispectral Remote Sensing ImageC lassification[A].Proceed ings of IEEE International Conferenceon System s,Man and Cybernetics[C].2005.
[3]N ing S N.Remote Sensing Image Processing and Applications[M].Beijing:Earthquake Press,1995.[宁书年.遥感图像处理与应用[M].北京:地震出版社,1995.]
[4]Ram esh Jain,Jayaram Murthy S N,Peter L J Chen(LuongTran).S im ilarity M easures for Image Databases[A]Proc.4 thIEEE International Conference on Fuzzy System s,Yokohama[C].1995.
[5]Bao Q,Guo P.Comparative Stud ies on S im ilarity M easures forRemote Sensing Image Retrieval[A].Proceed ings of the IEEEConference on System s,Man and Cybernetics[C].2004.
[6]Zhang Y J.Content-based V ision Information Retrieval[M].Beijing:Science Press,2003.[章毓晋.基于内容的视觉信息检索[M].北京:科学出版社,2003.]
[7]Ku llback S.Information Theory and Statistics[M].New York,1968.
[8]Cover TM,Thomas J A.E lem ents of Information Theory,W ileySeries in Telecommun ications[M].John W iley&Sons,NewYork,USA,1991.
[9]Swain M J,Ballard D H.Color Indexing[J].InternationalJournal ofComputer Vision,1991,7(1):11—32.
[10]Rafael C Gonzalez,R ichard E W oods.D igital Image Processing[M].2nd ed.Add ison-W esley Longman Pub lish ing Co.,Inc.USA,1992.
[11]Bozkaya T,Oasoyoglu M.D istance-based Indexing for H igh-d im ensionalM etric Spaces[J].ACM Special Interest Group onManagem ent ofData,1997,26(2):357—368.
[12]B ian Z Q,Zhang X G.Pattern Recogn ition[M].2nd ed.,Beijing:Tsinghua Un iversity Press,2000.[边肇祺,张学工.模式识别[M].第二版.北京:清华大学出版社,2000.]
[13]X ie H.Som e Strategies to Improve the Ob ject Profile-based ImageRetrieval Model and Its Prototype[D].M.S.thesis,JinanUn iv.,Guangzhou,Ch ina,2001.[谢红.基于目标形状的图像检索模型的改进策略及其原型系统[D].暨南大学硕士学位论文,中国广州,2001.]
[14]Hu G S.D igital S ignal Process-Theory,A lgorithm andRealization[M].2nd ed,Beijing:Tsinghua Un iversity Press,2003.[胡广书.数字信号处理———理论、算法与实现[M].第二版,北京:清华大学出版社,2003.]
[15]Sergios Theodorid is,Konstantinos Koutroumbas.Patternrecogn ition[M].2nd ed.,USA:Academ ic Press,2003.
[16]ChengM D,Shen X C.Introduction to Image Recogn ition[M].Shanghai:Shanghai Science Technology Press,1983.[程民德,沈燮昌.图像识别导论[M].上海:上海科学技术出版社,1983.]
[17]Jeffrey Bach,Santanu Pau l,Ram esh Jain.A V isual InformationManagem ent System for the Interactive Retrieval of Faces[J].IEEE Transactions on Knowledge and Data Engineering,1993,5:619—628.
[18]W angW H.Content-based Image Retieval Techn ique Research[D].Ph.D.d issertation,Changsha,Ch ina,2001.[王文惠.基于内容的图像检索技术研究[D].国防科学技术大学博士学位论文,2001.]
[2]Tian Y Q,Guo P,Lyu M R.Comparative Stud ies on FeatureExtraction M ethods for Mu ltispectral Remote Sensing ImageC lassification[A].Proceed ings of IEEE International Conferenceon System s,Man and Cybernetics[C].2005.
[3]N ing S N.Remote Sensing Image Processing and Applications[M].Beijing:Earthquake Press,1995.[宁书年.遥感图像处理与应用[M].北京:地震出版社,1995.]
[4]Ram esh Jain,Jayaram Murthy S N,Peter L J Chen(LuongTran).S im ilarity M easures for Image Databases[A]Proc.4 thIEEE International Conference on Fuzzy System s,Yokohama[C].1995.
[5]Bao Q,Guo P.Comparative Stud ies on S im ilarity M easures forRemote Sensing Image Retrieval[A].Proceed ings of the IEEEConference on System s,Man and Cybernetics[C].2004.
[6]Zhang Y J.Content-based V ision Information Retrieval[M].Beijing:Science Press,2003.[章毓晋.基于内容的视觉信息检索[M].北京:科学出版社,2003.]
[7]Ku llback S.Information Theory and Statistics[M].New York,1968.
[8]Cover TM,Thomas J A.E lem ents of Information Theory,W ileySeries in Telecommun ications[M].John W iley&Sons,NewYork,USA,1991.
[9]Swain M J,Ballard D H.Color Indexing[J].InternationalJournal ofComputer Vision,1991,7(1):11—32.
[10]Rafael C Gonzalez,R ichard E W oods.D igital Image Processing[M].2nd ed.Add ison-W esley Longman Pub lish ing Co.,Inc.USA,1992.
[11]Bozkaya T,Oasoyoglu M.D istance-based Indexing for H igh-d im ensionalM etric Spaces[J].ACM Special Interest Group onManagem ent ofData,1997,26(2):357—368.
[12]B ian Z Q,Zhang X G.Pattern Recogn ition[M].2nd ed.,Beijing:Tsinghua Un iversity Press,2000.[边肇祺,张学工.模式识别[M].第二版.北京:清华大学出版社,2000.]
[13]X ie H.Som e Strategies to Improve the Ob ject Profile-based ImageRetrieval Model and Its Prototype[D].M.S.thesis,JinanUn iv.,Guangzhou,Ch ina,2001.[谢红.基于目标形状的图像检索模型的改进策略及其原型系统[D].暨南大学硕士学位论文,中国广州,2001.]
[14]Hu G S.D igital S ignal Process-Theory,A lgorithm andRealization[M].2nd ed,Beijing:Tsinghua Un iversity Press,2003.[胡广书.数字信号处理———理论、算法与实现[M].第二版,北京:清华大学出版社,2003.]
[15]Sergios Theodorid is,Konstantinos Koutroumbas.Patternrecogn ition[M].2nd ed.,USA:Academ ic Press,2003.
[16]ChengM D,Shen X C.Introduction to Image Recogn ition[M].Shanghai:Shanghai Science Technology Press,1983.[程民德,沈燮昌.图像识别导论[M].上海:上海科学技术出版社,1983.]
[17]Jeffrey Bach,Santanu Pau l,Ram esh Jain.A V isual InformationManagem ent System for the Interactive Retrieval of Faces[J].IEEE Transactions on Knowledge and Data Engineering,1993,5:619—628.
[18]W angW H.Content-based Image Retieval Techn ique Research[D].Ph.D.d issertation,Changsha,Ch ina,2001.[王文惠.基于内容的图像检索技术研究[D].国防科学技术大学博士学位论文,2001.]
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心