摘要
本文选择青海省玛多县为土地沙化监测的典型研究区,影像数据来自美国的QuickBird遥感卫星,经度范围98°04′12.9″~98°10′42.28″,纬度范围34°42′5.9″~34°47′29″,面积为100平方公里,影像的过境时间是2007年8月,全色段空间分辨率为0.61m,多光谱空间分辨率为2.44m,影像数据共有四个多光谱波段,分别为:B(450~520nm)、G (520~600nm)、R (630~690nm)、NIR (760~900nm)波段。由于分辨率较高的全色波段数据为黑白影像,包含的光谱信息量较少,本文选择全色和多光谱波段融合进行研究。经辐射定标、大气校正图像融合、几何纠正处理后再进行面向对象分类。在研究中采用了QuickBird高分辨率遥感影像,应用面向对象技术方法对研究区进行了土地沙化分类分级,提取沙化信息,达到了较好的分类精度。此方法与人工解译相比,大大提高了工作效率,具有重要的推广应用价值。
In this paper, the author chooses MaDuo county of Qinghai province for Sandy Desertification monitoring.It is the typical study area, and the data is the QuickBird that the American satellite image data of remote sensing.
The Longitude range is 98°04′12.9″—98°10′42.28″.The latitudes is 34°42′5.9″—34°47′29″. The area is 100 square kilometers; the transit time is in August 2007. Full-color segment space is 0.61 m, and the Multi-spectral resolution is 2.44 m. Image data have four Multi-spectral bands,the respectively regions is: B (450 ~ 520nm), G (520 ~ 600nm), R (6.3 ~ 690nm), NIR (7.6 ~ 900nm) bands. Due to the high resolution full-color band data is black and white images, including less spectral information. This paper merge more full-color and Multi-spectral bands. Through mergging of radiation calibration, atmospheric correction, mixing bands of the image, geometric correction, after that process the object-oriented classification.
In the study used the QuickBird high-resolution remote sensing image,and apply the object-oriented technology method in the study area .ECognitiong classification software on the Sandy Desertification classification, extracted the desertification information classification, and obtain a good classification accuracy. This method compared with artificially interpretation, which greatly improved the extracted by remote sensing image the speed of desertification information.
引文
[1]石玉芳,徐东瑞.河北坝上草原风蚀沙化遥感监测及防治对策[J].河北师范大学学报,1996,20(4):96-101.
[2]张玉贵.TM影像的计算机屏幕解译和荒漠化监测[J].林业科学研究,1998,11(6):599-606.
[3]李金桐,镨拉提,纪良等.基于GIS的MODIS环境荒漠化监测中的应用方法研究[J]新疆气象,2003,26(2):20-23.
[4]黄慧萍,吴炳方,李苗苗等.高分辨率影像城市绿地快速提取技术与应用[J].遥感学报,2004,8(l):68-74.
[5]张路,廖明生.一种顾及上下文的遥感影像模糊聚类[J].遥感学报,2006,10(l):58-65.
[6]ZhangY.OPtimisation of Building Deteetionin Satellite Images by Combining MultisPectral Classification and Texture Filtering[J].PE&RS,54(1999):50-60.
[7]BlasehkeT , stroblJ.what is Wrong with pixels?some Recent Developments Interfacing Remote Sensing and GIS[J].GeoBIT/GIS,2001,(6):12-17.
[8]ZhangJX,FoodyGM.AFuZZy Classifcation of Sub-urban Land Cover from Remotely Sensed Imagery [J].IntJ.RemoteSensing,1998,19(14):2721-2738
[9]程承旗,马廷.高分辨率卫星影像上地物线性特征的自动识别[J].遥感学报,2003,7(l):26-30.
[10]贾永红,李德仁.多源遥感影像像素级融合分类与决策级分类融合法的研究[J].武汉大学学报信息科学版,2001,26(5):430-434.
[11]许磊,李朝峰,杨蒙召,SVM结合模糊方法在遥感图像分类中的应用[J].计算机工程与应用,2005,41(36):79-82.
[12]MuralH,Omatu s.Remote Sensing Image Analysis Using a Neural Network and kowledge-based Proeessing[J].Remote Sensing,1997,18(4):811-828.
[13]李厚强,王宜卞,刘政凯.一种适用于多类别遥感图像分类的新方法-复合神经网络分类方法[J].遥感学报,1997,l(4):257-261
[14]刘修国,罗小波.自组织神经网络在遥感影像分类中的应用研究[J].国土资源遥感,2004,(4):14-18.
[15]张友水,冯学智,阮仁宗等.Kohonen神经网络在遥感影像分类中的应用研究[J].遥感学报.2004,8(2):178-184.
[16]DefiniensImageGmbH.eCognitionUserGuide[M].German,2004,60-165.
[17]Ketting R J,Landgrebe D A.Classifieation of Multi-sPectral lmage Data by Extractionand Classifieation of Homogeneous objects [J].IEEE Transactions on Geoseience and Electronies,1976,14(l):19-26.
[18]B Kartikeyan,K L Majumder,A R DasguPta.An ExPert System for LandCover Classifieation[J].IEEET Transactions on Geoseience and remotesensing,1995,33(l):59-66.
[19]Lobo A,Chie O,Casterad A.Classification of Mediterranean Crops with Multi sensor Data:Per-Pixel Versus Per-object Statistics and Image Segmentation [J].International Journal of RemoteSensing 17(1996):2358-2400.
[20]Baatz M,Sehape A.Multiresoluti on Segmentation:An OPtimiZation APProach for High Quality Multi·scale Image Segmentaion [J]. 2000.
[21]骆剑承,周成虎,杨艳.遥感地学智能图解模型支持下的土地覆盖/土地利用分类[J].自然资源学报,2001,16(2):179-183.
[22]Hay G J,Niemann K 0.Visualizing3-D Texture:A Three-Dimensional Structural Approach to Model Forest Texture[J].Candian Joumai of Remote Sensing,1994,20(2):90-101.
[23]Hay G J,Niemann K O,Goodenough DG.Spatial Thiresholds,Imageo-bjects and Upscaling:A Multi-Scale Evaluation [J]. Remote Sensing of Environment, 62 (1997) : l-19.?
[24]Hay G J,Marceau D J,Dube P,et al.A Multiscale Framework for Landscape Analysis: object-specific analysis and Upscaling [J].Landscape Ecology.2001,Vol.16.
[25]Castleman K R.Digital Image Processing[M].Prentice Hall,1996.
[26]章毓晋.图象分割[M].北京:科学出版社,2001.
[27]范九伦.模糊聚类新算法与聚类有效性问题研究田[J].西安:西安电子科技大煞博士学位论文1998.
[28]朱述龙,张占睦.遥感图像获取与分析[M].北京:科学技术出版社,2000.
[29]文贡坚,王润生.一种有效的实现分裂合并算法的数据结构[J].中国图像图形学报.1998,2(7):544-548.
[30]青海省统计局.青海省统计年鉴,青海省统计出版社.2008.
[31]明冬萍,骆剑承,沈占锋.高分辨率遥感影像信息提取与目标识别技术研究[J].测绘科学,2005,Voll30(13):18-21.
[32]黄慧萍.面向对象影像分析中的尺度问题研究[D].北京:中国科学院研究生院遥感应用所,博士学位论文,2003.
[33]李霖,吴凡.空间数据多尺度表达模型及其可视化[M].北京:科学出版社,2005,2.
[34]邬建国.景观生态学一格局、过程、尺度与等级[M].北京:高等教育出版社,2000.11-l2.
[35]孙即样.图像分析[M].北京:科学出版社,2005.
[36]Milan Sonka.Vaclav Hlavac,Roger Boyle.Image processing,Analysis and Machine Vision (SeeondEdition)[M].Intemational Thomson Publishing,2002.
[37]杜凤兰,田庆久,夏学齐等.面向对象的地物分类法分析与评价[J].遥感技术与应用,2004,19(1):20-23
[38]秦其明.遥感图像自动解译面临的问题与解决的途径[J].测绘科学. 2000, 25(2):21-24
[39]苏伟,李京,陈云浩等.基于多尺度影像分割的面向对象城市土地覆被分类研究-以马来西亚吉隆坡市城市中心区为例[J].遥感学报,2007,11(4):521-530.
[40]周春艳,王萍,张振勇等.基于面向对象信息提取技术的城市用地分类[J].遥感技术与应用,2008,23(1):31-35
[41]邓劲松,王珂,邓艳华等. SPOT-5卫星影像中水体信息自动提取的一种有效方法[J].上海交通大学学报,2005,23(2):198-201
[42]李云鹏,洪金益.浅论基于面向对象的遥感图像分类[J].西部探矿工程.2008(2):103-105
[43]JANSSEN L. L. F., VAN DER WEL F. J. M. Accuracy Assessment of Satellite-derived Land-cover Data: A Review [J]. Photogrammetric Engineering and Remote Sensing, 1994,60 (4) : 419-426.
[44]Kalkhen, M.A., R. Reich, and T.J. Stohlgren. Assessing the accuracy of Landsat Thematic Mapper classification using double sampling [J].International Journal of Remote Sensing, 1998, 19 (11) : 2049-2060.
[45]Congalton R G. A Review of Assessing the Accuracy of Classifications of Remotely Sensed Data [J]. Remote Sensing of Environment, 1991, 37 (1) : 35-46.
[46]Hay A M. The Derivation of Global Estimation from a Confusion Matrix [J]. International Journal of Remote Sensing, 1988,9 (8) : 1395-1398.
[47]Congalton R G. A Comparison of Sampling Schemes Used in Generating Error atrices for Assessing the Accuracy of Maps Generated from Remotely Sensing Data [J]. Photogrammetric Engineering and Remote Sensing, 1988, 54 (5): 593-600.