基于CBERSO2B和SAR的喀斯特地区土地利用景观格局研究
|
摘要
贵州大部分地区属典型的岩溶环境,环境容量小,生态环境脆弱。为了能更好地研究喀斯特地区生态环境,本文以毕节地区的毕节市为典型样区,首次选用CBERS02B和SAR数据的配合,探讨两类型遥感影像在喀斯特山区图像处理上的技术实用性以及土地利用较理想的分类方法。
在遥感影像的处理技术上,本研究对SAR和CBERS02B分别采用了斑点噪声压缩、边缘增强、Wallis自适应滤波以及直方图均衡化和去霾处理。在SAR的处理中发现均值滤波和Lee滤波压缩斑点噪声的效果较好,Robinson 3-level模型和Prewitt Gradient模型较大程度地增强了地物的边缘效果,Wallis自适应滤波在不同的窗口大小下对影像的增强效果明显不一样,窗口越大,亮度信息越集中;在对CBERS02B图像的处理上,直方图均衡化较好地增强了影像的总体亮度,经过低频模糊度和高频模糊度处理之后增强了交通用地的纹理信息。而SAR和CBERS02B的融合方法主成份中的直方图匹配法显示了较好的效果,而且还较好地去除了原多光谱的云层信息。
利用监督分类对CBERS02B进行土地利用的分类,并在此基础上,辅以雷达单波段的纹理分类,对前后分类精度进行比较;最后对研究区和在不同岩组背景下的土地利用成果数据进行景观格局研究。通过不断地试验及比较,最终得出以下结论:
(1)在CBERS02B影像纯光谱分类的基础上,辅以雷达单波段的纹理分类,在一定程度上提高了研究区土地利用的分类精度。
(2)在整个研究区中,水域的分维值最大,裸岩石砾地的分维值最小,显示了人类活动对裸岩石砾地的干扰强度较大;整个研究区域显示了相对丰富度和破碎度较低,均匀度较小,异质性较大以及较高的优势度,其中以耕地、林地占主导地位。
(3)不论是在喀斯特区域、半喀斯特区域还是非喀斯特区域,均以林地和耕地占主导地位,且园地较其他地类斑块都较破碎;草地、园地和水域在喀斯特较为破碎;人为景观中的耕地和建设用地在半喀斯特最为破碎;自然景观中的林地则在非喀斯特整体性较好。裸岩石砾地在喀斯特的面积百分比远大于其余两类。非喀斯特区域内嵌块体类型的相对丰富度和破碎度较低,其受一种或少数几种景观支配的现象较喀斯特和半喀斯特较明显;而半喀斯特区域内的嵌块体类型分布较均匀,异质性小,优势度小,但破碎度较高。
Most parts of Guizhou is the model of the karst environment, the environmental capacity is small and the ecological environment is frangible,In order to study the ecological environment better in the karst area,this article took Bijie city in Bijie area as the typical type area,selecting the CBERS02B and the SAR datas for the first time,discussing the imagery processing technical usability about the two types of remote sensing images in the karst mountainous area as well as the landuse ideal Classifing approaches.
On the processing technology of the remote sensing images, this research had used methods of Speckle Suppression,the Edge Enhancement,the Wallis Adaptive Filtering,the Histogram Equalization and the Haze Reduction to the SAR and CBERS02B datas.During the processing to the SAR,it discovered that the Mean filter and the Lee filter reduced the speckle noise better; the Robinson the 3-level model and Prewitt the Gradient model strengthened the edge effect of the images greatly;the Wallis Adaptive Filter enhanced the images in obviously different effect under the different window sizes,when the window was bigger,the luminance information was more centralized.AND during the processing to the CBERS02B,it also discovered that the Histogram Equalization strengthened the brightness of the whole images very well,and after the low frequency ambiguity and high frequency ambiguity processing,the transportation land texture information had been strengthened.The method of the PCI histogram matching had demonstrated the good effect in resoluting merge of the SAR and in the CBERS02B and had also removed the original multi-spectrum cloudy layer information very well.
This article performed supervised classification to the CBERS02B,and basing on the supervised classification, plusing the texture classification of the radar single wave band,comparing the precision of the classifications,studied the landscape pattern about the landuse in the area and under the different rocks,finally it drew conclusions as follows:
(1),The pure spectrum classified precision of the CBERS02B had been improved to a certain extent after plusing the texture classification of the radar.
(2),The fractal dimension value of the water was the biggest,and the bare rock gravel's was the smallest,which had demonstrated the human activity acted greatly to the bare rock gravel;the entire research region had also demonstrated that the relative rich degree and broken degree were lower, evenest was smaller,complex was bigger as well as higher superiority,with the farmland,the forest land occupying the dominant position.
(3),The forest and the farmland had occupied the dominant position whatever in the karst,half-karst and the non-karst region,and the garden landuse was much more broken than others; The lawn,the garden landuse and the waters was much more broken in the karst;The farmland and the construction land of the artificial landscape were much more broken in the half-karst, and the integrity of the natural landscape's forest was good in the non-karst;The bare rock gravel was too much bigger than other two kinds in the karst area percentage,the non-karst region had demonstrated that the relative rich degree and broken degree were lower,it fell under the one kind or the minority several kind of landscape control's phenomenon was much more obvious than the karst and half-karst;The half-karst region had demonstrated that the evenest was bigger,complex was smaller as well as lower superiority,with higher broken degree
在遥感影像的处理技术上,本研究对SAR和CBERS02B分别采用了斑点噪声压缩、边缘增强、Wallis自适应滤波以及直方图均衡化和去霾处理。在SAR的处理中发现均值滤波和Lee滤波压缩斑点噪声的效果较好,Robinson 3-level模型和Prewitt Gradient模型较大程度地增强了地物的边缘效果,Wallis自适应滤波在不同的窗口大小下对影像的增强效果明显不一样,窗口越大,亮度信息越集中;在对CBERS02B图像的处理上,直方图均衡化较好地增强了影像的总体亮度,经过低频模糊度和高频模糊度处理之后增强了交通用地的纹理信息。而SAR和CBERS02B的融合方法主成份中的直方图匹配法显示了较好的效果,而且还较好地去除了原多光谱的云层信息。
利用监督分类对CBERS02B进行土地利用的分类,并在此基础上,辅以雷达单波段的纹理分类,对前后分类精度进行比较;最后对研究区和在不同岩组背景下的土地利用成果数据进行景观格局研究。通过不断地试验及比较,最终得出以下结论:
(1)在CBERS02B影像纯光谱分类的基础上,辅以雷达单波段的纹理分类,在一定程度上提高了研究区土地利用的分类精度。
(2)在整个研究区中,水域的分维值最大,裸岩石砾地的分维值最小,显示了人类活动对裸岩石砾地的干扰强度较大;整个研究区域显示了相对丰富度和破碎度较低,均匀度较小,异质性较大以及较高的优势度,其中以耕地、林地占主导地位。
(3)不论是在喀斯特区域、半喀斯特区域还是非喀斯特区域,均以林地和耕地占主导地位,且园地较其他地类斑块都较破碎;草地、园地和水域在喀斯特较为破碎;人为景观中的耕地和建设用地在半喀斯特最为破碎;自然景观中的林地则在非喀斯特整体性较好。裸岩石砾地在喀斯特的面积百分比远大于其余两类。非喀斯特区域内嵌块体类型的相对丰富度和破碎度较低,其受一种或少数几种景观支配的现象较喀斯特和半喀斯特较明显;而半喀斯特区域内的嵌块体类型分布较均匀,异质性小,优势度小,但破碎度较高。
Most parts of Guizhou is the model of the karst environment, the environmental capacity is small and the ecological environment is frangible,In order to study the ecological environment better in the karst area,this article took Bijie city in Bijie area as the typical type area,selecting the CBERS02B and the SAR datas for the first time,discussing the imagery processing technical usability about the two types of remote sensing images in the karst mountainous area as well as the landuse ideal Classifing approaches.
On the processing technology of the remote sensing images, this research had used methods of Speckle Suppression,the Edge Enhancement,the Wallis Adaptive Filtering,the Histogram Equalization and the Haze Reduction to the SAR and CBERS02B datas.During the processing to the SAR,it discovered that the Mean filter and the Lee filter reduced the speckle noise better; the Robinson the 3-level model and Prewitt the Gradient model strengthened the edge effect of the images greatly;the Wallis Adaptive Filter enhanced the images in obviously different effect under the different window sizes,when the window was bigger,the luminance information was more centralized.AND during the processing to the CBERS02B,it also discovered that the Histogram Equalization strengthened the brightness of the whole images very well,and after the low frequency ambiguity and high frequency ambiguity processing,the transportation land texture information had been strengthened.The method of the PCI histogram matching had demonstrated the good effect in resoluting merge of the SAR and in the CBERS02B and had also removed the original multi-spectrum cloudy layer information very well.
This article performed supervised classification to the CBERS02B,and basing on the supervised classification, plusing the texture classification of the radar single wave band,comparing the precision of the classifications,studied the landscape pattern about the landuse in the area and under the different rocks,finally it drew conclusions as follows:
(1),The pure spectrum classified precision of the CBERS02B had been improved to a certain extent after plusing the texture classification of the radar.
(2),The fractal dimension value of the water was the biggest,and the bare rock gravel's was the smallest,which had demonstrated the human activity acted greatly to the bare rock gravel;the entire research region had also demonstrated that the relative rich degree and broken degree were lower, evenest was smaller,complex was bigger as well as higher superiority,with the farmland,the forest land occupying the dominant position.
(3),The forest and the farmland had occupied the dominant position whatever in the karst,half-karst and the non-karst region,and the garden landuse was much more broken than others; The lawn,the garden landuse and the waters was much more broken in the karst;The farmland and the construction land of the artificial landscape were much more broken in the half-karst, and the integrity of the natural landscape's forest was good in the non-karst;The bare rock gravel was too much bigger than other two kinds in the karst area percentage,the non-karst region had demonstrated that the relative rich degree and broken degree were lower,it fell under the one kind or the minority several kind of landscape control's phenomenon was much more obvious than the karst and half-karst;The half-karst region had demonstrated that the evenest was bigger,complex was smaller as well as lower superiority,with higher broken degree
引文
[1].贵州省国土资源厅编制,贵州省地图集[M],成都地图出版社印刷厂,2005:8-179
[2].王建军,吕喜玺,周跃,基于Landsat ETM+卫星影像的长江上游混浊河水悬沙浓度提取[J],科学通报,2007,(52):234-240
[3].郑长春,王秀珍,黄敬峰,基于特征波段的SPOT-5卫星影像水稻面积信息自动提取的方法研究[J],遥感技术与应用,2008,23(3):294-299
[4].刘克,赵文吉,胡德勇,基于SPOT5和北京一号小卫星数据的北京北部山区土壤侵蚀变化研究[J],国土资源遥感,2008,(3):78-82
[5].周佳,赵冬玲,何宇华等,基于中巴卫星影像的土地利用分类精度评价[J],测绘科学,2007,32(6):27-29
[6].张微,黄勤,李启敏,卫星热红外遥感技术在地震监测预报中的应用[J],杭州科技,2008,(1):63-64.
[7].陈江,王安建,利用ASTER热红外遥感数据开岩石化学成分填图的初步研究[J],遥感学报,2007,11,(4):601-608.
[8].刘汉湖,杨武年,夏涛,高精度卫星遥感图像在岩溶地区岩溶漏斗调查中的应用[J],工程勘察,2007(7):68-72
[9].张文,周立江,潘发明等,利用CBERS02B进行汶川地震区森林资源损失快速评估山地学报[J],2008,26(6):748-754
[10].胡娟,熊康宁,安裕伦,CBERS-02数据在喀斯特石漠化遥感调查中的应用——以贵州黔南布依族自治州为例[J],贵州师范大学学报(自然科学版),2008,26(2):39-42
[11].赵柏林,微波遥感水面油污状况的研究[J],中国科学,1983,26(3):283-283
[12].杨龙江,阎满存,李华梅,探地雷达在隐伏岩溶区工程地质勘探中的应用[J],物探化探计算技术,1999,21(2):172-176
[13].陈劲松,邵芸,林晖,全极化SAR数据在地表覆盖/利用监测中的应用[J],国土资源遥感,2004,2:39-42
[14].曹银璇,张继贤,燕琴等,RADARSAT影像在土地利用动态监测中的应用[J],测绘科学2005.8(30):57-60
[15].李红军,李进忠,多普勒天气雷达在沙尘天气监测中的应用[C],,研究论文技术报告:2006,29(6):11-14
[16].池泓,汪小钦,Envisat ASAR数据在沿海防护林提取中的应用[J],测绘科学,2008,33(3):138-140
[17].XU Qing,LIN Hui,ZHENG Quanan,Evaluation of ENVISAT ASAR data for sea surface wind retrieval in HongKong coastal waters of China[J],ActaOceanologica Sinica 2008,27(4):57-62
[18].HellmannM,CloudeSR,PapathanassiouKP.Classification Using Polarimetri cand Interferometric SAR Data[C].IEEEIGARSS.97,1997:1411-1413.
[19].秦其明,陆荣建,分形与神经网络方法在卫星数字图像分类中的应用[J],北京大学学报(自然科学版),2000.11(36):858-864
[20].LeeJS,GrunesMR,KwokR.Classification of Multi-look Polarimetric SAR Imagery Based on ComplexWishartDistri2bution[J].InternationalJournalofRemoteSensing,1994,15(11):2299-2311.
[21].LiuHao,ShaoYun.An Improved Learning Vector Quantization Neural Network for LandCover ClassificationwithmultitemporalRadarsatImages[C].IEEEIGRASS.98,1998,1787-1789.
[22].VanZylJJ,BurnetteCE.Bayesian Classification of Polarimetric SAR Images Using Adaptivea PrioriProbabilities[J].In2ternationalJournalofRemoteSensing,1992,13(5):835-840.
[23].邵怀勇,仙巍,周万村,基于3S技术的三峡库区不同高程带景观格局研究—以巫山县为例[J],水土保持通报,2005,25(3):54-57
[24].仙巍,邵怀勇,周万村,基于3S技术的三峡库区不同坡度带与坡向带景观格局研究—以巫溪县为例[J],中国生态农业学报,2007,15(1):140-144
[25].魏丹,唐代生,唐嘉锴等,基于3S技术的土地利用景观格局研究[J],森林工程,2008,24(2):5-8
[26].肖丹,CBERS数字图像处理技术及其在喀斯特山区土地利用动态监测中的应用[D],贵州师范大学硕士学位论文,2006
[27].贵州省农业区划丛书编辑委员会,毕节县综合农业区划[M],贵州:贵州人民出版社:1989:4-10
[28].赵英时,遥感应用分析原理与方法[M].北京:科学出版社,2003:194-277
[29].曾琪明,焦健,合成孔径雷达遥感原理及应用简介(1-4)[J],遥感信息,1998,2:41-47
[30].曾桂香,基于多尺度纹理分析的SAR图像地物分类[D],武汉大学硕士学位论文,2004
[31].匡燕,李安,李子扬等,ENVISAT卫星综述[J],遥感数据,2006,6:90-92
[32].党安荣,王晓栋,陈晓峰等,ERDAS IMAGINE遥感图象处理方法[M],北京:清华大学出版社,2002:186-317
[33].胡娟,“3S”技术支持下喀斯特景观解译与格局研究—以清镇市为例[D],贵州师范大学硕士学位论文,2006
[34].贾永红,李德仁,孙家柄.多源遥感影像数据融合[J].遥感技术与应用,2000,15(1):41-44.
[35].党安荣,毛其智.人居环境研究中多源遥感信息融合试验[J].清华大学学报:自然科学版,2000,40(S1):7-10.
[36].张炳智,张继贤,张丽,土地利用动态遥感监测中多源遥感影像融合方法比较研究[J],测绘科学,2000,25(3):46-50
[37].邹敏,吴泉源,曲伟,SPOT-5遥感影像自身融合方法的比较研究[J],山东师范大学学报(自然科学版),2007,22(1):95-97
[38].孙丹峰,IKONOS全色与多光谱数据融合方法的比较研究[J],遥感技术与应用,2002,17(1):41-45
[39].王智均,李德仁,基于小波理论的IKONOS卫星全色影像和多光谱影像的融合[J].测绘学报,2001,30(2):112-116
[40].刘桂芳,SPOT5遥感影像不同融合方法对地物识别精度的影响[J],河南大学学报(自然科学版)2006,36(3):58-62
[41].贾海峰,Haifeng Jia,Xuehua Liu,环境遥感原理与应用[M],清华大学出版社出版,2006:101
[42].王人潮,史舟,胡月明.浙江红壤资源信息系统的研制与应用[M].北京:中国农业出版社,1999
[43].RosenfeldA,KakA.Digital Picture Processing(2nd edition)[M].Academic Press,1982.
[44].G.R.Cross and A.K.Jain.Markov random field texture models.IEEE Trans.Pattern Analysis Machine Intelligence.1983,5(1):25-39.
[45].R.L.Kashyap and P.M.Lapsa.Synthesis and Estimation of Random Fields Using Long-Correlation Models.IEEE Transactions on Pattern Analysis and Machine Intelligence.1984,6(6):800-809.
[46].Zhang Q,Wang J,Gong P,et al.Study of Urban Spatial Patterns SPOT Pan chromatic Imagery Using Texture Analysis[J].International Journalof Remote Sensing,2003,24(21):4137-4160.
[47].杨淑莹,胡军,曹作良,基于图像纹理分析的目标物体识别方法[J],天津理工学院学,2001,17(4):31-33
[48].彭光雄,李京,何宇华等,用纹理分析方法提取CBERS02星CCD图像土地覆盖信息[J],遥感技术与应用,2007,22(1):8-12
[49].薛重生,王霞.基于分形几何的遥感图像纹理分析方法及应用[J].地质科技情报,1997,16:199-105.
[50].章孝灿,黄智才,赵元洪.遥感数字图像处理[M].浙江:浙江大学出版社,1997.
[51].章毓晋.图像工程(上册)[M].北京:清华大学出版社,2000
[52].贾永红编著.数字图像处理[M],武汉大学出版社,2003:178.
[53].王娟,基于分形纹理的遥感影像土地覆盖类型识别方法研究[D],辽宁工程技术大学硕士学位论文,2006
[54].A.Rosenfeld,Digital Picture Analysis,Springer-Verlag,1976.
[55].Haralick.R.M,Shanmugam.K,Dinstein.Texture Features for Image Classification.IEEE Transaction of System,Man,and Cybernetics.1973,3(6):610-621
[56].于海鹏,刘一星,王金满等.木材表面纹理的灰度共生矩阵分析[C].中国林学会木材科学分会第九次学术研讨会论文集.2008:314-322
[57].A.B and F.P.An investigation of the Textural Characteristics Associated with Gray Level Co-occurrence Matrix Statistical Parameters.IEEE Transactions OnGeoscience and Reomte Sensing.1995,32(2):293-304.
[58].P.Gong,J.D.Marceau,and P.J.Howarth.A comparison of spatial feature extraction algorithms for land-use classification with SPOTHRD data.Remote Sensing Environment.1992.40:137-151.
[59].龚声蓉,刘纯平 王强等,数字图像处理与分析[M],清华大学出版社出版,2006:232
[60].孙艳霞,纹理分析在遥感图像识别中的应用[D],新疆大学硕士学位论文,2005
[61].孙栋,基于纹理分析的目标图像识别技术研究[D],南京理工大学硕士学位论文,2005
[62].李俊杰,何隆华,戴锦芳等,基于遥感影像纹理信息的湖泊围网养殖区提取[J],湖泊科学,2006,18(4):337-342
[63].吴樊,王超,张红,基于纹理特征的高分辨率SAR影像居民区提取[J],遥感技术与应用:2005,20(1):149-152
[64].舒宁,卫星遥感影像纹理分析与分形分维方法[J].武汉测绘科技大学报,1998,23(4):370-373.
[65].Anne P,Jacky H,Christiane W.The Utility of Texture Analysis to Improve Per-pixel Classification for High to Very High Spatial Resolution Imagery [J].International Journal of Remote Sensing,2005,26(4):733-745.
[66].Steven E.Franklin and Bradley A.Wilson.A three-stage classifier for remote sensing of mountain environments.PE&RS.1992,58(4):449-454.
[67].布仁仓,常禹,胡远满等,基于Kappa系数的景观变化测度—以辽宁省中部城市群为例[J],生态学报,2005,25(4):778-783
[68].闫岩,华琳,张建,对诊断一致性kappa系数及评价指标的探讨[C],中国卫生统计2007,24(3):313-315
[69].黄杏元,马劲松,汤勤等,地理信息系统概论[M],北京:高等教育出版社,2001,2:14-17
[70].肖笃宁,李秀珍,高峻.景观生态学[J].北京:科学出版社,2003:1-10.
[71].邬建国.景观生态学—格局、过程、尺度与等级[M].北京:高等教育出版社,2000.
[72].Foman R,Godron M.Landscape Ecology[M].New York John Wiley&Sons,1986
[73].卢远等.基于RS和GIS的喀斯特山区景观生态格局[J].山地学报,2002,20(6):727-731.
[74].徐建华.现代地理学中的数学方法[M].北京:高等教育出版社.2002:412-413
[75].李阳兵.不同石漠化程度岩溶峰丛洼地系统景观格局的比较[J].地理研究.2005,24(3)
[76].丁圣彦,张明亮.1988-2002年开封市景观动态变化[J].地理研究.2005,24(1):28-36
[77].仝川,金蕾,周延林.内蒙古托克地区景观格局特征分析[J].应用生态学报,2003,14(1):56-89
[78].陈彩虹,胡锋,张落成.南京市城乡交错带景观格局研究[J].应用生态学报,2003, 14(8):1363-1368
[79].赵羿,李月辉.实用景观生态学[M].北京:科学出版社,2001
[80].FuB J.1995.Landscape diversity analysis and its design.Acta EcolSin,15(4):345-349
[81].贵州省区域地理信息项目领导小组编.贵州省地理信息数据集[M].贵州:贵州人民出版社,1996.275.
[82].安裕伦,吕涛,熊康宁等,“3S”在贵州喀斯特石漠化现状研究中的应用探讨[M].北京:中国水利水电出版社,2001.144-152.
[83].傅伯杰,陈顶利,马克明等,景观生态学原理及应用[M]。北京:科学出版社.2002
[84].周兴,广西石灰岩山地受害生态系统的改建[J],山地研究(现《山地学报》).1995.13(4):241-247
[2].王建军,吕喜玺,周跃,基于Landsat ETM+卫星影像的长江上游混浊河水悬沙浓度提取[J],科学通报,2007,(52):234-240
[3].郑长春,王秀珍,黄敬峰,基于特征波段的SPOT-5卫星影像水稻面积信息自动提取的方法研究[J],遥感技术与应用,2008,23(3):294-299
[4].刘克,赵文吉,胡德勇,基于SPOT5和北京一号小卫星数据的北京北部山区土壤侵蚀变化研究[J],国土资源遥感,2008,(3):78-82
[5].周佳,赵冬玲,何宇华等,基于中巴卫星影像的土地利用分类精度评价[J],测绘科学,2007,32(6):27-29
[6].张微,黄勤,李启敏,卫星热红外遥感技术在地震监测预报中的应用[J],杭州科技,2008,(1):63-64.
[7].陈江,王安建,利用ASTER热红外遥感数据开岩石化学成分填图的初步研究[J],遥感学报,2007,11,(4):601-608.
[8].刘汉湖,杨武年,夏涛,高精度卫星遥感图像在岩溶地区岩溶漏斗调查中的应用[J],工程勘察,2007(7):68-72
[9].张文,周立江,潘发明等,利用CBERS02B进行汶川地震区森林资源损失快速评估山地学报[J],2008,26(6):748-754
[10].胡娟,熊康宁,安裕伦,CBERS-02数据在喀斯特石漠化遥感调查中的应用——以贵州黔南布依族自治州为例[J],贵州师范大学学报(自然科学版),2008,26(2):39-42
[11].赵柏林,微波遥感水面油污状况的研究[J],中国科学,1983,26(3):283-283
[12].杨龙江,阎满存,李华梅,探地雷达在隐伏岩溶区工程地质勘探中的应用[J],物探化探计算技术,1999,21(2):172-176
[13].陈劲松,邵芸,林晖,全极化SAR数据在地表覆盖/利用监测中的应用[J],国土资源遥感,2004,2:39-42
[14].曹银璇,张继贤,燕琴等,RADARSAT影像在土地利用动态监测中的应用[J],测绘科学2005.8(30):57-60
[15].李红军,李进忠,多普勒天气雷达在沙尘天气监测中的应用[C],,研究论文技术报告:2006,29(6):11-14
[16].池泓,汪小钦,Envisat ASAR数据在沿海防护林提取中的应用[J],测绘科学,2008,33(3):138-140
[17].XU Qing,LIN Hui,ZHENG Quanan,Evaluation of ENVISAT ASAR data for sea surface wind retrieval in HongKong coastal waters of China[J],ActaOceanologica Sinica 2008,27(4):57-62
[18].HellmannM,CloudeSR,PapathanassiouKP.Classification Using Polarimetri cand Interferometric SAR Data[C].IEEEIGARSS.97,1997:1411-1413.
[19].秦其明,陆荣建,分形与神经网络方法在卫星数字图像分类中的应用[J],北京大学学报(自然科学版),2000.11(36):858-864
[20].LeeJS,GrunesMR,KwokR.Classification of Multi-look Polarimetric SAR Imagery Based on ComplexWishartDistri2bution[J].InternationalJournalofRemoteSensing,1994,15(11):2299-2311.
[21].LiuHao,ShaoYun.An Improved Learning Vector Quantization Neural Network for LandCover ClassificationwithmultitemporalRadarsatImages[C].IEEEIGRASS.98,1998,1787-1789.
[22].VanZylJJ,BurnetteCE.Bayesian Classification of Polarimetric SAR Images Using Adaptivea PrioriProbabilities[J].In2ternationalJournalofRemoteSensing,1992,13(5):835-840.
[23].邵怀勇,仙巍,周万村,基于3S技术的三峡库区不同高程带景观格局研究—以巫山县为例[J],水土保持通报,2005,25(3):54-57
[24].仙巍,邵怀勇,周万村,基于3S技术的三峡库区不同坡度带与坡向带景观格局研究—以巫溪县为例[J],中国生态农业学报,2007,15(1):140-144
[25].魏丹,唐代生,唐嘉锴等,基于3S技术的土地利用景观格局研究[J],森林工程,2008,24(2):5-8
[26].肖丹,CBERS数字图像处理技术及其在喀斯特山区土地利用动态监测中的应用[D],贵州师范大学硕士学位论文,2006
[27].贵州省农业区划丛书编辑委员会,毕节县综合农业区划[M],贵州:贵州人民出版社:1989:4-10
[28].赵英时,遥感应用分析原理与方法[M].北京:科学出版社,2003:194-277
[29].曾琪明,焦健,合成孔径雷达遥感原理及应用简介(1-4)[J],遥感信息,1998,2:41-47
[30].曾桂香,基于多尺度纹理分析的SAR图像地物分类[D],武汉大学硕士学位论文,2004
[31].匡燕,李安,李子扬等,ENVISAT卫星综述[J],遥感数据,2006,6:90-92
[32].党安荣,王晓栋,陈晓峰等,ERDAS IMAGINE遥感图象处理方法[M],北京:清华大学出版社,2002:186-317
[33].胡娟,“3S”技术支持下喀斯特景观解译与格局研究—以清镇市为例[D],贵州师范大学硕士学位论文,2006
[34].贾永红,李德仁,孙家柄.多源遥感影像数据融合[J].遥感技术与应用,2000,15(1):41-44.
[35].党安荣,毛其智.人居环境研究中多源遥感信息融合试验[J].清华大学学报:自然科学版,2000,40(S1):7-10.
[36].张炳智,张继贤,张丽,土地利用动态遥感监测中多源遥感影像融合方法比较研究[J],测绘科学,2000,25(3):46-50
[37].邹敏,吴泉源,曲伟,SPOT-5遥感影像自身融合方法的比较研究[J],山东师范大学学报(自然科学版),2007,22(1):95-97
[38].孙丹峰,IKONOS全色与多光谱数据融合方法的比较研究[J],遥感技术与应用,2002,17(1):41-45
[39].王智均,李德仁,基于小波理论的IKONOS卫星全色影像和多光谱影像的融合[J].测绘学报,2001,30(2):112-116
[40].刘桂芳,SPOT5遥感影像不同融合方法对地物识别精度的影响[J],河南大学学报(自然科学版)2006,36(3):58-62
[41].贾海峰,Haifeng Jia,Xuehua Liu,环境遥感原理与应用[M],清华大学出版社出版,2006:101
[42].王人潮,史舟,胡月明.浙江红壤资源信息系统的研制与应用[M].北京:中国农业出版社,1999
[43].RosenfeldA,KakA.Digital Picture Processing(2nd edition)[M].Academic Press,1982.
[44].G.R.Cross and A.K.Jain.Markov random field texture models.IEEE Trans.Pattern Analysis Machine Intelligence.1983,5(1):25-39.
[45].R.L.Kashyap and P.M.Lapsa.Synthesis and Estimation of Random Fields Using Long-Correlation Models.IEEE Transactions on Pattern Analysis and Machine Intelligence.1984,6(6):800-809.
[46].Zhang Q,Wang J,Gong P,et al.Study of Urban Spatial Patterns SPOT Pan chromatic Imagery Using Texture Analysis[J].International Journalof Remote Sensing,2003,24(21):4137-4160.
[47].杨淑莹,胡军,曹作良,基于图像纹理分析的目标物体识别方法[J],天津理工学院学,2001,17(4):31-33
[48].彭光雄,李京,何宇华等,用纹理分析方法提取CBERS02星CCD图像土地覆盖信息[J],遥感技术与应用,2007,22(1):8-12
[49].薛重生,王霞.基于分形几何的遥感图像纹理分析方法及应用[J].地质科技情报,1997,16:199-105.
[50].章孝灿,黄智才,赵元洪.遥感数字图像处理[M].浙江:浙江大学出版社,1997.
[51].章毓晋.图像工程(上册)[M].北京:清华大学出版社,2000
[52].贾永红编著.数字图像处理[M],武汉大学出版社,2003:178.
[53].王娟,基于分形纹理的遥感影像土地覆盖类型识别方法研究[D],辽宁工程技术大学硕士学位论文,2006
[54].A.Rosenfeld,Digital Picture Analysis,Springer-Verlag,1976.
[55].Haralick.R.M,Shanmugam.K,Dinstein.Texture Features for Image Classification.IEEE Transaction of System,Man,and Cybernetics.1973,3(6):610-621
[56].于海鹏,刘一星,王金满等.木材表面纹理的灰度共生矩阵分析[C].中国林学会木材科学分会第九次学术研讨会论文集.2008:314-322
[57].A.B and F.P.An investigation of the Textural Characteristics Associated with Gray Level Co-occurrence Matrix Statistical Parameters.IEEE Transactions OnGeoscience and Reomte Sensing.1995,32(2):293-304.
[58].P.Gong,J.D.Marceau,and P.J.Howarth.A comparison of spatial feature extraction algorithms for land-use classification with SPOTHRD data.Remote Sensing Environment.1992.40:137-151.
[59].龚声蓉,刘纯平 王强等,数字图像处理与分析[M],清华大学出版社出版,2006:232
[60].孙艳霞,纹理分析在遥感图像识别中的应用[D],新疆大学硕士学位论文,2005
[61].孙栋,基于纹理分析的目标图像识别技术研究[D],南京理工大学硕士学位论文,2005
[62].李俊杰,何隆华,戴锦芳等,基于遥感影像纹理信息的湖泊围网养殖区提取[J],湖泊科学,2006,18(4):337-342
[63].吴樊,王超,张红,基于纹理特征的高分辨率SAR影像居民区提取[J],遥感技术与应用:2005,20(1):149-152
[64].舒宁,卫星遥感影像纹理分析与分形分维方法[J].武汉测绘科技大学报,1998,23(4):370-373.
[65].Anne P,Jacky H,Christiane W.The Utility of Texture Analysis to Improve Per-pixel Classification for High to Very High Spatial Resolution Imagery [J].International Journal of Remote Sensing,2005,26(4):733-745.
[66].Steven E.Franklin and Bradley A.Wilson.A three-stage classifier for remote sensing of mountain environments.PE&RS.1992,58(4):449-454.
[67].布仁仓,常禹,胡远满等,基于Kappa系数的景观变化测度—以辽宁省中部城市群为例[J],生态学报,2005,25(4):778-783
[68].闫岩,华琳,张建,对诊断一致性kappa系数及评价指标的探讨[C],中国卫生统计2007,24(3):313-315
[69].黄杏元,马劲松,汤勤等,地理信息系统概论[M],北京:高等教育出版社,2001,2:14-17
[70].肖笃宁,李秀珍,高峻.景观生态学[J].北京:科学出版社,2003:1-10.
[71].邬建国.景观生态学—格局、过程、尺度与等级[M].北京:高等教育出版社,2000.
[72].Foman R,Godron M.Landscape Ecology[M].New York John Wiley&Sons,1986
[73].卢远等.基于RS和GIS的喀斯特山区景观生态格局[J].山地学报,2002,20(6):727-731.
[74].徐建华.现代地理学中的数学方法[M].北京:高等教育出版社.2002:412-413
[75].李阳兵.不同石漠化程度岩溶峰丛洼地系统景观格局的比较[J].地理研究.2005,24(3)
[76].丁圣彦,张明亮.1988-2002年开封市景观动态变化[J].地理研究.2005,24(1):28-36
[77].仝川,金蕾,周延林.内蒙古托克地区景观格局特征分析[J].应用生态学报,2003,14(1):56-89
[78].陈彩虹,胡锋,张落成.南京市城乡交错带景观格局研究[J].应用生态学报,2003, 14(8):1363-1368
[79].赵羿,李月辉.实用景观生态学[M].北京:科学出版社,2001
[80].FuB J.1995.Landscape diversity analysis and its design.Acta EcolSin,15(4):345-349
[81].贵州省区域地理信息项目领导小组编.贵州省地理信息数据集[M].贵州:贵州人民出版社,1996.275.
[82].安裕伦,吕涛,熊康宁等,“3S”在贵州喀斯特石漠化现状研究中的应用探讨[M].北京:中国水利水电出版社,2001.144-152.
[83].傅伯杰,陈顶利,马克明等,景观生态学原理及应用[M]。北京:科学出版社.2002
[84].周兴,广西石灰岩山地受害生态系统的改建[J],山地研究(现《山地学报》).1995.13(4):241-247