Hwan-Hee Yoo

Accurate classification of water area with fusion of RADARSAT and SPOT satellite imagery

We fused RADARSAT images and SPOT panchromatic images by a wavelet transform in order to improve the accuracy of classification of water areas. Fused images not only maintained the characteristic of the SAR image (low pixel value) but also had improved boundary information. This leads to an accurate method to classify water areas.

Surface reconstruction using terrain-dependent rational function model

This study reports on the surface reconstruction from high resolution satellite imagery with the terrain-dependent RFM (rational function model). The study area covers 17 km/spl times/14 km with two pairs of stereo images. The overlapped area is taken from SPOT panchromatic and KOMPSAT (Korea Multi-Purpose Satellite) stereo images. The ground resolution of each image is 10 m and 6.6 m, respectively. We improved the accuracy of RFM using correlation analysis for both images. We then successfully created surface DEM (digital elevation model) for the overlapped areas of both images.

Precise geometric modeling for KOMPSAT-1 satellite imagery

The geometric modeling for 3-D positioning was performed using the strip stereo images with the ancillary data of KOMPSAT-1. As a result, the satellite position vector offset was corrected by only one ground control point and the accuracy for the geometric modeling improved to 25.7 m RMSE.