面向图像差异特征融合的基于弗里德曼检验的小波基分类研究
|
摘要
对小波基分类是图像融合中依据不同融合需求选择小波基的基础,可以提高图像融合的智能化水平。针对现有的小波基分类方法仅根据小波自身特性进行分类,没有从统计角度有效建立小波基和图像差异特征之间的联系,本文提出了面向图像差异特征融合的基于弗里德曼检验的小波基分类方法。首先,选择典型的差异特征和小波基用于分类研究;其次,选择针对差异特征的评价指标,以评价指标结果作为标记量并进行分类实验的区组设计;然后,采用弗里德曼检验对不同区组数据进行处理及执行相应的后续检验和分类步骤,形成面向图像差异特征的小波基类集;最后,设计对比实验对分类方法的有效性进行验证和分析。试验结果表明,该分类方法能有效把对图像差异特征融合效果相近的小波基归为一类,能根据融合需求选择较好的小波基。
The classification of wavelet bases is in accordance with different fusion requirements in the image fusion process, which can improve the intelligence level of image fusion. In the existing wavelet base classification methods, based only on the wavelet's inherent characteristics, there is no effective connection between the wavelet bases and the image difference features from the statistical perspective. This paper proposes a wavelet base classification method based on Friedman test for the image feature fusion combining different features. Initially, typical difference features and wavelet bases for classification research are selected, and then evaluation indices to evaluate difference features are selected. The evaluation indices results are utilized as mark measures to conduct block design of the classification experiment. Subsequently, the Friedman test to process different data blocks is utilized and the corresponding test procedures are implemented. Furthermore, classification procedures are performed to construct a set of wavelet bases that can fuse image difference features satisfactorily. Finally, a comparative experiment is designed to verify and analyze the effectiveness of the classification method. The experimental results demonstrate that the classification method can effectively classify the wavelet bases with similar fusion performance on image difference features into subsets and can choose a better wavelet base according to fusion requirements.
The classification of wavelet bases is in accordance with different fusion requirements in the image fusion process, which can improve the intelligence level of image fusion. In the existing wavelet base classification methods, based only on the wavelet's inherent characteristics, there is no effective connection between the wavelet bases and the image difference features from the statistical perspective. This paper proposes a wavelet base classification method based on Friedman test for the image feature fusion combining different features. Initially, typical difference features and wavelet bases for classification research are selected, and then evaluation indices to evaluate difference features are selected. The evaluation indices results are utilized as mark measures to conduct block design of the classification experiment. Subsequently, the Friedman test to process different data blocks is utilized and the corresponding test procedures are implemented. Furthermore, classification procedures are performed to construct a set of wavelet bases that can fuse image difference features satisfactorily. Finally, a comparative experiment is designed to verify and analyze the effectiveness of the classification method. The experimental results demonstrate that the classification method can effectively classify the wavelet bases with similar fusion performance on image difference features into subsets and can choose a better wavelet base according to fusion requirements.
引文
[1]杨风暴.红外物理与技术[M].北京:电子工业出版社,2014.YANG Fengbao.Infrared Physics and Technology[M].Beijing:Publishing House of Electronics Industry,2014.
[2]张彬,杨风暴.小波分析方法及其应用[M].北京:国防工业出版社,2011.ZHANG Bin,YANG Fengbao.Wavelet Analysis Method and Its Application[M].Beijing:National Defense Industry Press,2011.
[3]Amolins K,Zhang Y,Dare P.Wavelet based image fusion techniquesAn introduction,review and comparison[J].Isprs Journal of Photogrammetry&Remote Sensing,2007,62(4):249-263.
[4]黄奎.基于小波变换的图像处理综述[J].价值工程,2015,34(8):255-256.HUANG Kui.Overview of Image Processing Based on Wavelet Transform[J].Value Engineering,2015,34(8):255-256.
[5]龚建周,刘彦随,夏北成,等.小波基及其参数对遥感影像融合图像质量的影响[J].地理与地理信息科学,2010,26(2):6-10.GONG Jianzhou,LIU Yansui,XIA Beicheng,et al.Effect of Wavelet Basis and Decomposition Levels on Performance of Fusion Images from Remotely Sensed Data[J].Geography and Geo-Information Science,2010,26(2):6-10.
[6]安富,杨风暴,蔺素珍,等.特征差异驱动的红外偏振与光强图像融合[J].中国科技论文,2014(1):96-102.AN Fu,YANG Fengbao,LIN Suzhen,et al.Feature differences driven fusion of infrared polarization and intensity images[J].China Sciencepaper,2014(1):96-102.
[7]牛涛,杨风暴,王肖霞,等.差异特征与融合算法的集值映射关系的建立[J].红外与激光工程,2015,44(3):1073-1079.NIU Tao,YANG Fengbao,WANG Xiaoxia,et al.Establishment of set-valued mapping between difference characteristics and fusion algorithms[J].Infrared and Laser Engineering,2015,44(3):1073-1079.
[8]方丽萍,张馨方,罗启姣,等.管道泄漏信号去噪中小波基及分解尺度的选取--基于小波分析[J].钦州学院学报,2017,32(5):18-22.FANG Liping,ZHANG Xinfang,LUO Qijiao,et al.Wavelet Basis and Decomposition Scale Selection in Pipeline Leakage Signal Denoising--Based on wavelt[J].Journal of Qinzhou University,2017,32(5):18-22.
[9]Sappa A D,Carvajal J A,Aguilera C A,et al.Wavelet-Based Visible and Infrared Image Fusion:A Comparative Study[J].Sensors,2016,16(6):861-876.
[10]杨艳春,李娇,王阳萍.图像融合质量评价方法研究综述[J].计算机科学与探索,2018,12(7):6-20.YANG Yanchun,LI Jiao,WANG Yangping.Review of Image Fusion Quality Evaluation Methods[J].Journal of Frontiers of Computer Science and Technology,2018,12(7):6-20.
[11]王跃华,陶忠祥.红外与可见光图像融合质量评价方法综述[J].红外,2012,33(6):7-11.WANG Yuehua,TAO Zhongxiang.Overview of Quality Evaluation Methods of Fused Infrared and Visible Images[J].Infrared,2012,33(6):7-11.
[12]LIU Z,Blasch E,XUE Z,et al.Objective Assessment of Multiresolution Image Fusion Algorithms for Context Enhancement in Night Vision:AComparative Study[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2011,34(1):94-109.
[13]Ayd,Taner Z O,Alpayd,et al.Cost-conscious comparison of supervised learning algorithms over multiple data sets[M].Elsevier Science Inc.2012.
[14]Trawiński B,Sm?tek M,Telec Z,et al.Nonparametric statistical analysis for multiple comparison of machine learning regression algorithms[J].International Journal of Applied Mathematics&Computer Science,2012,22(4):449-471.
[15]Derrac J,García S,Molina D,et al.A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms[J].Swarm&Evolutionary Computation,2011,1(1):3-18.
[16]Garc,Salvador A,Ndez A,et al.Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining:experimental analysis of power[J].Information Sciences,2010,180(10):2044-2064.
[17]Noble W S.How does multiple testing correction work?[J].Nature Biotechnology,2009,27(12):1135-1137.
[18]Bergmann B,Hommel P G.Improvements of General Multiple Test Procedures for Redundant Systems of Hypotheses[M]//Multiple Hypothesenprüfung/Multiple Hypotheses Testing.Springer Berlin Heidelberg,1988:3-7.
[2]张彬,杨风暴.小波分析方法及其应用[M].北京:国防工业出版社,2011.ZHANG Bin,YANG Fengbao.Wavelet Analysis Method and Its Application[M].Beijing:National Defense Industry Press,2011.
[3]Amolins K,Zhang Y,Dare P.Wavelet based image fusion techniquesAn introduction,review and comparison[J].Isprs Journal of Photogrammetry&Remote Sensing,2007,62(4):249-263.
[4]黄奎.基于小波变换的图像处理综述[J].价值工程,2015,34(8):255-256.HUANG Kui.Overview of Image Processing Based on Wavelet Transform[J].Value Engineering,2015,34(8):255-256.
[5]龚建周,刘彦随,夏北成,等.小波基及其参数对遥感影像融合图像质量的影响[J].地理与地理信息科学,2010,26(2):6-10.GONG Jianzhou,LIU Yansui,XIA Beicheng,et al.Effect of Wavelet Basis and Decomposition Levels on Performance of Fusion Images from Remotely Sensed Data[J].Geography and Geo-Information Science,2010,26(2):6-10.
[6]安富,杨风暴,蔺素珍,等.特征差异驱动的红外偏振与光强图像融合[J].中国科技论文,2014(1):96-102.AN Fu,YANG Fengbao,LIN Suzhen,et al.Feature differences driven fusion of infrared polarization and intensity images[J].China Sciencepaper,2014(1):96-102.
[7]牛涛,杨风暴,王肖霞,等.差异特征与融合算法的集值映射关系的建立[J].红外与激光工程,2015,44(3):1073-1079.NIU Tao,YANG Fengbao,WANG Xiaoxia,et al.Establishment of set-valued mapping between difference characteristics and fusion algorithms[J].Infrared and Laser Engineering,2015,44(3):1073-1079.
[8]方丽萍,张馨方,罗启姣,等.管道泄漏信号去噪中小波基及分解尺度的选取--基于小波分析[J].钦州学院学报,2017,32(5):18-22.FANG Liping,ZHANG Xinfang,LUO Qijiao,et al.Wavelet Basis and Decomposition Scale Selection in Pipeline Leakage Signal Denoising--Based on wavelt[J].Journal of Qinzhou University,2017,32(5):18-22.
[9]Sappa A D,Carvajal J A,Aguilera C A,et al.Wavelet-Based Visible and Infrared Image Fusion:A Comparative Study[J].Sensors,2016,16(6):861-876.
[10]杨艳春,李娇,王阳萍.图像融合质量评价方法研究综述[J].计算机科学与探索,2018,12(7):6-20.YANG Yanchun,LI Jiao,WANG Yangping.Review of Image Fusion Quality Evaluation Methods[J].Journal of Frontiers of Computer Science and Technology,2018,12(7):6-20.
[11]王跃华,陶忠祥.红外与可见光图像融合质量评价方法综述[J].红外,2012,33(6):7-11.WANG Yuehua,TAO Zhongxiang.Overview of Quality Evaluation Methods of Fused Infrared and Visible Images[J].Infrared,2012,33(6):7-11.
[12]LIU Z,Blasch E,XUE Z,et al.Objective Assessment of Multiresolution Image Fusion Algorithms for Context Enhancement in Night Vision:AComparative Study[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2011,34(1):94-109.
[13]Ayd,Taner Z O,Alpayd,et al.Cost-conscious comparison of supervised learning algorithms over multiple data sets[M].Elsevier Science Inc.2012.
[14]Trawiński B,Sm?tek M,Telec Z,et al.Nonparametric statistical analysis for multiple comparison of machine learning regression algorithms[J].International Journal of Applied Mathematics&Computer Science,2012,22(4):449-471.
[15]Derrac J,García S,Molina D,et al.A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms[J].Swarm&Evolutionary Computation,2011,1(1):3-18.
[16]Garc,Salvador A,Ndez A,et al.Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining:experimental analysis of power[J].Information Sciences,2010,180(10):2044-2064.
[17]Noble W S.How does multiple testing correction work?[J].Nature Biotechnology,2009,27(12):1135-1137.
[18]Bergmann B,Hommel P G.Improvements of General Multiple Test Procedures for Redundant Systems of Hypotheses[M]//Multiple Hypothesenprüfung/Multiple Hypotheses Testing.Springer Berlin Heidelberg,1988:3-7.