Extra cting the Boundary of Forests from High-resolution Remote SensingImagery Using Texture Feature in Contour let Doma in

LI Xiao-feng; ZHANG Shu-qing; NA Xiao-dong; YU Huan; KONG Bo; LIU Chun-yue

Contourlet is a new 22D signa l ana lysis tool as well as wavelet ana lysis. It provides a flexible multiresolution local and directiona l image expansion, and ismore suitable for image processing than othermath tools. Inth is paper, an IKONOS imagewith forests is decomposed by Contourlet transformation, then amethod for extractingthe boundary of forests from high-resolution remote sensing imagery is proposed associated watershed transformationand texture analysis. The experiment showed that themethod could extract the boundary of forests accurately.

1. North east Institute ofGeography and Agroecology, Ch inese Academy of Sciences, Changchun 130012, Jilin, Ch ina

2. GraduateUn ivers ity ofCh inese Academy of Sc iences, Beijing 100039, China

Received Date: 2008-12-10

Abstract: Contourlet is a new 22D signa l ana lysis tool as well as wavelet ana lysis. It provides a flexible multiresolution local and directiona l image expansion, and ismore suitable for image processing than othermath tools. Inth is paper, an IKONOS imagewith forests is decomposed by Contourlet transformation, then amethod for extractingthe boundary of forests from high-resolution remote sensing imagery is proposed associated watershed transformationand texture analysis. The experiment showed that themethod could extract the boundary of forests accurately.

HTML

Reference (15)

[1]	杨永恬, 田昕, 冯仲科, 等. 遥感混和分类算法及其在森林分类中的应用[J]. 测绘科学, 2004, 29 ( 4): 55- 57
[2]	李春干, 李崇贵, 景海涛. 森林小班多边形的自动聚合[J]. 北京林业大学学报, 2005, 27 ( 1): 55 - 58
[3]	Woodcock C E, Strah ler A H. The factor of scale in remote sen sing[J]. R emote Sen sE nvironment, 1987, 21: 311- 332
[4]	H ayG J, Castilla G, Wu lderM A, et al. An automated ob ject-basedapproach for the mu ltiscale image segmen tation of forest scenes[J].In ternat ional Journal of Appl ied Earth Observation and Geo informat ion, 2005, 7: 339- 359
[5]	Marceau D J, H owarth P J, Dubois JM M, et al. Evaluation of theGrey-LevelCo2Occu rrenceMatrixMethod For Land2Cover Classificat ion U sing SPOT Imagery[J]. IEEE T ran saction on Geoscience andRemote Sens ing, 1990, 28 ( 4 ): 513- 519
[6]	H ayG J, N iemannK O, McLeanG F. An ob ject-specific image-texture analysis ofH -resolu tion forest imagery [J]. Remote Sen singE nviron, 1996, 55: 108 122
[7]	Wang Zuyu an, Boesch R. Color- andTex ture2Based Image Segmentation for Imp roved Forest Delin eation[J]. IEEE T ran saction on G eoscience and Remote Sen sing, 2007, 45 ( 10): 3055- 3062
[8]	Bouzid iA, BaazizN. Con tourlet Domain Featu re Extract ion for Image Conten tAu th ent icat ion Mu lt imed ia Signal Processing [J]. 2006IEEE 8 thWorkshop on, 2006: 202- 206
[9]	沙宇恒, 丛琳, 孙强, 等. 基于Con tourlet域HMT模型的多尺度图像分割[J]. 红外与毫米波学报, 2005, 24( 6): 472- 476
[10]	侯彪, 刘芳, 焦李成. 基于小波变换的高分辨SAR港口目标自动分割[J]. 红外与毫米波学报, 2002, 21( 5): 385- 389
[11]	DonohoM N, VetterliM. Contou rlets: a new d irectionalmu ltiresolut ion image representation[C]. Sign als, Sys tems and Computers,Conference Record of the Th irty-Sixth As ilomar Conference, Roches ter, 2002, 1: 3- 6
[12]	焦李成, 谭山. 图像的多尺度几何分析: 回顾和展望[J]. 电子学报, 2003, 31 ( 12): 1975- 1981
[13]	杨丽君, 王保保. 一种基于小波和分水岭变换的图像分割方法[J]. 计算机应用, 2005, 25: 253- 254
[14]	H arisK, E fstrat iad is S N, Mag laveras N, et a l. H ybrid image segmen tation u sing watersh eds and fas t reg ion merg ing [J]. IEEETransact ions on Image Process ing, 1998, 7( 12 ): 1684 - 1699
[15]	Vander S C J, Jong SM, RooA P J. A Segmen tat ion and Classificat ion Approach of IKONOS22 lmagery for Land CoverMapp ing toAss ist F lood R isk and F lood Damage A ssessmen t[J]. In ternat ionalJourn al of App lied E arth Observation and Geoin format ion, 2003( 4 ): 217 - 229

Extra cting the Boundary of Forests from High-resolution Remote SensingImagery Using Texture Feature in Contour let Doma in

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Proportional views