高光谱溢油图像分类算法研究
摘要
国内外海上重大溢油事故频繁发生,对各国经济的可持续发展形成相当大的威胁。
随着高光谱遥感技术的发展,以及该技术在溢油区检测、油种识别、油膜厚度估计等方面的独特优势,高光谱海洋油膜遥感检测已经成为国内外关注的热点。
论文以大连溢油和青岛溢油高光谱图像为研究背景,对现有的分类算法进行比较研究,并在此基础上针对地物特性对现有分类算法进行改进,最后,将改进的算法应用在溢油高光谱图像的分类处理中,得到了较好的分类效果。
算法分析过程中的实验结果证明了论文所提分类算法的有效性,高光谱溢油图像分类系统的实现和处理结果,进一步说明了论文所提方法在高光谱图像处理中的可行性。
The oil spoil events occurred frequently in the last years, which makes it more and more important for the nations to detect and control the events because they do influence the maintaining develop of the countries.
Hyperspectral detection of marine oil spill have been focused recently, because the development of hyperspectral technologies is amazing and it have absolutely advantages in region detecting, type identifying and thickness determining.
Firstly, the current methods for classification are compared and discussed. Then, the improvement is made based on the current method considering the property of object spectral. At the last, the prompted algorithm is realized in the classifying system for hyperspectral oil spill images, which are from the spill event in Dalian and Qingdao. It came that the algorithm is good in performance.
The experiments commit during the process of analysis shows the efficiency of the prompted algorithm. And the realization and results in the classifying system of hyperspectral oil spill image further the feasible of the method.
随着高光谱遥感技术的发展,以及该技术在溢油区检测、油种识别、油膜厚度估计等方面的独特优势,高光谱海洋油膜遥感检测已经成为国内外关注的热点。
论文以大连溢油和青岛溢油高光谱图像为研究背景,对现有的分类算法进行比较研究,并在此基础上针对地物特性对现有分类算法进行改进,最后,将改进的算法应用在溢油高光谱图像的分类处理中,得到了较好的分类效果。
算法分析过程中的实验结果证明了论文所提分类算法的有效性,高光谱溢油图像分类系统的实现和处理结果,进一步说明了论文所提方法在高光谱图像处理中的可行性。
The oil spoil events occurred frequently in the last years, which makes it more and more important for the nations to detect and control the events because they do influence the maintaining develop of the countries.
Hyperspectral detection of marine oil spill have been focused recently, because the development of hyperspectral technologies is amazing and it have absolutely advantages in region detecting, type identifying and thickness determining.
Firstly, the current methods for classification are compared and discussed. Then, the improvement is made based on the current method considering the property of object spectral. At the last, the prompted algorithm is realized in the classifying system for hyperspectral oil spill images, which are from the spill event in Dalian and Qingdao. It came that the algorithm is good in performance.
The experiments commit during the process of analysis shows the efficiency of the prompted algorithm. And the realization and results in the classifying system of hyperspectral oil spill image further the feasible of the method.
引文
[1]http://news.sina.com.cn/c/2010-08-02/014020803423.shtml
[2]http://news.sina.com.cn/w/p/2010-06-22/110320522087.shtml
[3]http://tech.sina.com.cn/d/2011-02-24/11345213929.shtml
[4]国家环境保护总局,《中国保护海洋环境免受陆源污染国家报告》,2006。
[5]Natural Resources Canada Website. http://ccrs.nrcan.gc.ca/resource/tutor/fundam/chapter5/28_e.php.
[6]陆应诚,田庆久等.海面油膜光谱响应实验研究.科学通报,2008(9)：1085-1088
[7]R.R.Coifman,M.V.Wickerhauser. Entropy-based algorithms for best basis selection. IEEE Transaction on Information Theory.1992,38(2):713-719
[8]P.S.Chavez,G.L.Berlin and L.B.Sowers.Statistical method for selecting Landsat MSS ratios.Joumal of Applied Photographic Engineering.1982,8:23-30
[9]Yang-Lang Chang,Jyh-Perng Fang,Jin-Nan Liu,Hsuan Ren,Wen-Yew Liang. A Simulated Annealing Feature Extraction approach for hyperspectral images. Geoscience and Remote Sensing Symposium. IGARSS 2007.2007:3190-3193
[10]周爽.蚁群算法在高光谱图像降维和分类中的应用研究.哈尔滨工业大学博士论文,2010.
[11]Jia Xiuping and J.A.Richards.Segmented Principal Components Transtrormation for Efficient Hyperspectral Remote-Sensing Image Display and Classification.IEEE Trans.on Geoscience and Remote Sensing.1999,37(1):538-542
[12]冯燕,何明一等.基于独立成分分析的高光谱图像数据降维及压缩.电子与信息学报,2007(12)：2871-2875
[13]L.M.Bruce,C.H.Koger,J.Li.Dimensionality reduction of hyperspectral data using discrete wavelet transform feature extraction.IEEE Trans.Geosci. Remote Sens.2002,40(10):2331-2338
[14]C.A.Shah,M.K.Arora,S.A.Robila,and P.K.Varshney. ICA Mixture Model based Unsupervised Classification of Hyperspectral Imagery, Proceedings of the 31st Applied Imagery Pattern Recognition Workshop(AIPR.02).2002:481-487
[15]杨倩倩.高光谱溢油图像特征提取在油种识别中的应用.大连海事大学硕士论文,2010.
[16]赵冬至,丛丕福.海面溢油的可见光波段地物光谱特征研究.遥感技术与应用,2000,15(3)：160-164
[17]O. Zielinski, T. Hengstermann, and N. Robbe, Detection of oil spills by airborne sensors, Marine Surface Films,2006:255-271.
[18]M. Lennon, S. Babichenko, N. Thomas, V. Mariette, G. Mercier, and A. Lisin, Detection and mapping of oil slicks in the sea by combined use of hyperspectral imagery and laser induced fluorescence, Remote Sens. Labs eProc.5,2006:120-128
[19]Svejkovsky J., J. Muskat and J. Mullin.2008. Mapping Oil Spill Thickness with a Portable Multispectral Aerial Imager. Proc. International Oil Spill Conference 2008,2008
[20]陈蕾,邓孺孺,谢健.基于AISA+机载高光谱的海上石油开发溢油监测研究.海洋湖沼通报,2009(1)：179-184
[21]颜春燕,刘强.机载高光谱数据提取冬小麦冠层叶绿素含量的模型分析及验证.遥感信息,2009(5)：8-14
[22]胡兴堂.高光谱水质遥感监测系统关键技术研究.中国科学院研究生院博士论文,2006.
[23]杨杭,张霞等OMIS-Ⅱ机载高光谱遥感图像边缘辐射畸变校正方法优选.国土资源遥感,2010(2)：17-21
[24]M.N. Jha, J. Levy and Y. Gao. Advances in Remote Sensing for Oil Spill Disaster Management:State-of-the-Art Sensors Technology for Oil Spill Surveillance. Sensors,2008(8): 236-255
[25]郑志刚,统计模式识别中的聚类方法,中科大,自动化系,2007.10[63]
[26]吕翠芬.基于独立成分分析的高光谱溢油图像聚类研究,大连海事大学硕士学位论文,2011
[27]Medvedovic M.,Sivaganesan S..Bayesian infinite mixture model based clustering of gene expression profiles,Bioinformatics,2002,18(9):1194-1206.
[28]Amir,B.,Friedman,N.,Yakhini,Z..Class discovery in gene expression data. In:RECOMB. 2001:31-38.
[29]Pierre B..Bioinformatics:the machine learning approach.Cambridge:MIT Press,2003,276-281.
[30]刘远超,王晓龙,徐志明等.文档聚类综述.中文信息学报,2006,20(3)：55-62.
[31]董连凤.高光谱图像预处理技术研究.长安大学硕士学位论文,2007
[32]Goetz A F H, Srivastava V. Mineralogical Mapping in the Curite Minning District, Nevada. Proceedings of Airborne Imaging Spectrometers(AIS) Data Analysis Workshop,8-10 April, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,22-31
[33]Conel J E. Calibration of Data Using Ground-based Spectral Reflectance Measurements. Proceedings of Airborne Imaging Spectrometer Data Analysis Workshop,15, September, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,184-185
[34]Green A A. and Craig M D. Analysis of Aircraft Spectrometer Data with Logarithmic Residuals. Proceedings of Airborne Imaging Spectrometer Workshop, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,111-119
[35]林德荤.资源一号卫星CCD相机辐射定标和图像辐射校正[J].航天返回与遥感,2000,21(2)：8-13
[36]王新刚,视频监控中多人体目标检测与跟踪方法的研究,大连海事大学硕士学位论文,2008
[2]http://news.sina.com.cn/w/p/2010-06-22/110320522087.shtml
[3]http://tech.sina.com.cn/d/2011-02-24/11345213929.shtml
[4]国家环境保护总局,《中国保护海洋环境免受陆源污染国家报告》,2006。
[5]Natural Resources Canada Website. http://ccrs.nrcan.gc.ca/resource/tutor/fundam/chapter5/28_e.php.
[6]陆应诚,田庆久等.海面油膜光谱响应实验研究.科学通报,2008(9)：1085-1088
[7]R.R.Coifman,M.V.Wickerhauser. Entropy-based algorithms for best basis selection. IEEE Transaction on Information Theory.1992,38(2):713-719
[8]P.S.Chavez,G.L.Berlin and L.B.Sowers.Statistical method for selecting Landsat MSS ratios.Joumal of Applied Photographic Engineering.1982,8:23-30
[9]Yang-Lang Chang,Jyh-Perng Fang,Jin-Nan Liu,Hsuan Ren,Wen-Yew Liang. A Simulated Annealing Feature Extraction approach for hyperspectral images. Geoscience and Remote Sensing Symposium. IGARSS 2007.2007:3190-3193
[10]周爽.蚁群算法在高光谱图像降维和分类中的应用研究.哈尔滨工业大学博士论文,2010.
[11]Jia Xiuping and J.A.Richards.Segmented Principal Components Transtrormation for Efficient Hyperspectral Remote-Sensing Image Display and Classification.IEEE Trans.on Geoscience and Remote Sensing.1999,37(1):538-542
[12]冯燕,何明一等.基于独立成分分析的高光谱图像数据降维及压缩.电子与信息学报,2007(12)：2871-2875
[13]L.M.Bruce,C.H.Koger,J.Li.Dimensionality reduction of hyperspectral data using discrete wavelet transform feature extraction.IEEE Trans.Geosci. Remote Sens.2002,40(10):2331-2338
[14]C.A.Shah,M.K.Arora,S.A.Robila,and P.K.Varshney. ICA Mixture Model based Unsupervised Classification of Hyperspectral Imagery, Proceedings of the 31st Applied Imagery Pattern Recognition Workshop(AIPR.02).2002:481-487
[15]杨倩倩.高光谱溢油图像特征提取在油种识别中的应用.大连海事大学硕士论文,2010.
[16]赵冬至,丛丕福.海面溢油的可见光波段地物光谱特征研究.遥感技术与应用,2000,15(3)：160-164
[17]O. Zielinski, T. Hengstermann, and N. Robbe, Detection of oil spills by airborne sensors, Marine Surface Films,2006:255-271.
[18]M. Lennon, S. Babichenko, N. Thomas, V. Mariette, G. Mercier, and A. Lisin, Detection and mapping of oil slicks in the sea by combined use of hyperspectral imagery and laser induced fluorescence, Remote Sens. Labs eProc.5,2006:120-128
[19]Svejkovsky J., J. Muskat and J. Mullin.2008. Mapping Oil Spill Thickness with a Portable Multispectral Aerial Imager. Proc. International Oil Spill Conference 2008,2008
[20]陈蕾,邓孺孺,谢健.基于AISA+机载高光谱的海上石油开发溢油监测研究.海洋湖沼通报,2009(1)：179-184
[21]颜春燕,刘强.机载高光谱数据提取冬小麦冠层叶绿素含量的模型分析及验证.遥感信息,2009(5)：8-14
[22]胡兴堂.高光谱水质遥感监测系统关键技术研究.中国科学院研究生院博士论文,2006.
[23]杨杭,张霞等OMIS-Ⅱ机载高光谱遥感图像边缘辐射畸变校正方法优选.国土资源遥感,2010(2)：17-21
[24]M.N. Jha, J. Levy and Y. Gao. Advances in Remote Sensing for Oil Spill Disaster Management:State-of-the-Art Sensors Technology for Oil Spill Surveillance. Sensors,2008(8): 236-255
[25]郑志刚,统计模式识别中的聚类方法,中科大,自动化系,2007.10[63]
[26]吕翠芬.基于独立成分分析的高光谱溢油图像聚类研究,大连海事大学硕士学位论文,2011
[27]Medvedovic M.,Sivaganesan S..Bayesian infinite mixture model based clustering of gene expression profiles,Bioinformatics,2002,18(9):1194-1206.
[28]Amir,B.,Friedman,N.,Yakhini,Z..Class discovery in gene expression data. In:RECOMB. 2001:31-38.
[29]Pierre B..Bioinformatics:the machine learning approach.Cambridge:MIT Press,2003,276-281.
[30]刘远超,王晓龙,徐志明等.文档聚类综述.中文信息学报,2006,20(3)：55-62.
[31]董连凤.高光谱图像预处理技术研究.长安大学硕士学位论文,2007
[32]Goetz A F H, Srivastava V. Mineralogical Mapping in the Curite Minning District, Nevada. Proceedings of Airborne Imaging Spectrometers(AIS) Data Analysis Workshop,8-10 April, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,22-31
[33]Conel J E. Calibration of Data Using Ground-based Spectral Reflectance Measurements. Proceedings of Airborne Imaging Spectrometer Data Analysis Workshop,15, September, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,184-185
[34]Green A A. and Craig M D. Analysis of Aircraft Spectrometer Data with Logarithmic Residuals. Proceedings of Airborne Imaging Spectrometer Workshop, JPL Publication 85-41, Jet Propulsion Laboratory, Pasadena, California,1985,111-119
[35]林德荤.资源一号卫星CCD相机辐射定标和图像辐射校正[J].航天返回与遥感,2000,21(2)：8-13
[36]王新刚,视频监控中多人体目标检测与跟踪方法的研究,大连海事大学硕士学位论文,2008