论文信息 - Water body extraction in urban region from high resolution satellite imagery with near-infrared spectral analysis

Water body extraction in urban region from high resolution satellite imagery with near-infrared spectral analysis

Water body information is very important for urban planning and environment improving. Extracting water body information from the satellite images such as MODIS images, Spot images, Radarsat SAR images and LANDSAT TM images has been explored by some researchers. However, extracting water body information from high resolution satellite images is still a problem because of the shade effect of buildings in urban area. The color of the calm water surface is similar to the shadows and it is difficult to distinguish between water body and shadows from satellite imagery. With Near-Infrared Spectral Analysis of IKONOS imagery, we present a new method to extract water body and make distinctions between the water body and the shadows. We employ existing knowledge for extracting water body information by image analysis at first, then the dark objects is extracted by a object-oriented operation from the high resolution satellite image. Near-infrared spectral analysis is developed to remove the building shadows from the dark objects extracted. Two NIR indexes are combined to remove the shadows. The experiments show that the main water body, such as artificial lake and rivers can be extracted effectively from IKONOS imagery with the method presented in this paper.

[1] L. Mertes,et al. Remote sensing of riverine landscapes , 2002 .

[2] S. K. McFeeters. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features , 1996 .

[3] S. Goetz,et al. IKONOS imagery for resource management: Tree cover, impervious surfaces, and riparian buffer analyses in the mid-Atlantic region , 2003 .

[4] David H. Douglas,et al. ALGORITHMS FOR THE REDUCTION OF THE NUMBER OF POINTS REQUIRED TO REPRESENT A DIGITIZED LINE OR ITS CARICATURE , 1973 .

[5] Nicholas J. Redding,et al. Implementation of a Fast Algorithm for Segmenting SAR Imagery , 2002 .

[6] Mingjun Song,et al. A competitive pixel-object approach for land cover classification , 2005 .

[7] B. Forsberg,et al. Spatial patterns of hydrology, geomorphology, and vegetation on the floodplain of the Amazon River in Brazil from a remote sensing perspective , 1995 .

[8] D. M. McKeown,et al. Building knowledge-based systems for detecting man-made structures from remotely sensed imagery , 1988 .

[9] Demetre Argialas,et al. Geomorphological feature extraction from a digital elevation model through fuzzy knowledge-based classification , 2003, SPIE Remote Sensing.