GEOGRAPHICAL RESEARCH

Transcription

1 GEOGRAPHICAL RESEARCH Vol123, No12 Mar1, 2004 IKONOS,,, (, ) :, IKONOS, IKONOS,,, IKONOS 4, NDVI,,, : ; ; ; : TP79 ; TU : (2004) [1 ],,,, IKONOS, 1, [2 ] IKONOS, 1m, m, 4 m, : ( m) ( m) ( m) ( m) IKONOS,,,,, GIS, [3, 4 ],, [5 7 ] IKONOS,,, : ; : : (SILUP) : (19742),,

2 2 : IKONOS 275,,, IKONOS,,,,,,,,, [8 ],,,,,,,,, 2 IKONOS,, 4, IKONOS , 1m IKONOS, 4m IKONOS,,, 4 m 211,,, :,,,,,,,,,,, :, ;,,, 212,, 4 IKONOS 11 GeoTiff, 8,,,,

3 276 23,,, [9, 10 ], ( 4 1), , ,,,,,, 3,,,, IKONOS, [11 ] 2 311,, ( 3), 2, Fig12 Flow chart of information extraction,,,,,,,,,,,,,,, [12 ],,,,,,,,,

4 2 : IKONOS 277, 4, 312 (1) NDVI 3 Fig13 The spectral response of typical objects,, (NDVI), (DVI), (RVI) (PVI),, [13, 14 ] DVI Landsat MSS RVI,, RVI PVI,,,,, PVI, RVI NDVI,,, NDVI NDVI, : NDVI = ( IR2R) / ( IR + R) IKONOS,,, IR R,,,, 3, ( 1) NDVI, NDVI, ( 2) NDVI,NDVI,, NDVI,,,, NDVI, NDVI, NDVI, NDVI,NDVI,, IKONOS NDVI,,, ND2

5 VI 0118,,,, (2) ISODATA 4, 410, NDVI, IKONOS 4 > IKONOS 2, IKONOS 4 < IKONOS 2, NDVI, 3, 2, 2, , ,, , ,,, ISODATA,, IKONOS,,,, 3 3, 50 %,, , ISODATA ( 4 5),, ISO2 DATA [15 ],,,, ISODATA,,, ISODATA,,, 1 1 Tab11 Precision contrast of two research methods ( %) ( %) ISODATA

6 2 : IKONOS 279 4,,,,,, ISODATA,,, 16m 2,,,,,,,,,, [16 ] : [ 1 ], 1 1,2001,19(3) : [ 2 ],, 1 1,2000, 19(4) : [ 3 ],,, 1GIS RS 1,2001,6(3) : [ 4 ], 1 1, 1997, 16 (2) : [ 5 ],, 1 1, 2001, 20 (5) : [ 6 ] Yong Du, Philippe M1 Teillet, Josef Cihlar1 Radiometric normalization of multitemporal high2resolution satellite images with quality control for land cover change detection1 Remote Sensing of Environment,2002, 82 : [ 7 ],, 1 / 1, 2001, 20 (6) : [ 8 ],, 1 1 :, [ 9 ] Hutchinsion C F1 Techniques for combining landsat and ancillary data for digital classification improvement1 Photogrammetric En2 gineering & Remote Sensing, 1992,1 : [ 10 ] Miguel2Ayanz J, Bioging D1 An iterative classification approach for mapping natural resources from satellite imagery1 Int J Remote Sensing, 1996,7 (5) : [11 ], 1TM 1,2000,4 (2) : [12 ] Ben2Dor E, Levin N, Saaroni H1 Remote Sensing of an Urban Environment Using Hyperspectral Technology1 Proceedings of the Thirteenth International Conference on Applied Geologic Remote Sensing1123 March 1999, Vancouver, BC, Canada, II [13 ], 1 1, [14 ] Gilabert M A, Gonzalez2Piqueras J, Garcia2Haro F J, et al1 A generalized soil2adjusted vegetation index1 Remote Sensing of En2 vironment, 2002, 82 : [15 ] Zhang Y1 Texture2integrated classification of urban treed areas in high2resolution color2infrared imagery1 Photogrammetric Engi2 neering & Remote Sensing, 2001, 67 (12) : [16 ],, 1 1, 2002, 18 (2) :

7 Study on extraction of urban green space from IKONOS remote sensing images ZHANG You2shui, FENG Xue2zhi, DU Jin2kang, GU Guo2qin (Department of Urban and Resources Science, Nanjing University, Nanjing , China) Abstract :This paper discusses about the extraction of urban green space from an IKONOS image using a hierarchical classification technique1 Green space information was obtained based on the spectral characteristics of different objects with the help of available corresponding methods after the combina2 tion of IKONOS multi2spectral data1 Due to high resolution of IKONOS imagery, large amount of data and heterogeneous nature of spectrum, the extraction of urban green space was carried out on segments after image segmentation1 This would help much improving the accuracy of extraction of urban green space from the whole image1 In test area of the image, the spectral characteristics of different features in all 4 bands are ana2 lyzed1 The spectral characteristics of old urban area and asphalt road are similar to those of part of green space1 Moreover, it is difficult to extract green space under the shadow1 In order to extract infor2 mation from the mixed green space with non2green space, through enhancing NDVI values of a green space under the shadow, parts of green space are extracted (NDVI > 0118), then parts of non2green space are eliminated1 The next step is to extract green space from mixed green space and non2green space based on spectral knowledge and unsupervised ISODATA clustering1 Finally, green space infor2 mation of test area is obtained by aggregating different levels of green space1 The methodology is basi2 cally concerned with the object spectral features and noise due to the mixture of different land2use/ land2cover categories is significantly avoided1 To demonstrate the efficiency of proposed method, unsu2 pervised ISODATA clustering method was used to extract green space from the test area,then both re2 sults were compared to show accuracy1 The visual interpretation and ground truth checks of the test area have proved that the classification accuracy and productivity accuracy of the first method are higher than that of the latter1 Key words :information extraction ;green space ; normalized difference vegetation index ;mixed pixel

8