By using speetrally stable targets, the empirical line （EL） method was tested to correct the multispectral IKONOS imagery acquired over Putuo Mountain, Zhejiang, China. A series of calibration targets, which were spectrally stable over time, were selected to establish the linear predicted equation. Subsequently, a series of spectrally stable validation targets were selected to assess the accuracy of the equations. And, validation targets, which were speetrally unstable over time, were used to test the feasibility of using the EL method to calibrate the archival remotely sensed data. Ground reflectance measurements for each target were made using an ASD FieldSpec spectroradiometer. A Trimble GeoXTTM GPS unit with sub-meter accuracy was used to estimate the target position accurately. Linear regression equations for four tKONOS bands were derived. The coefficients of determination for the blue, green, and red bands were all greater than 0.9800 and it was 0.9697 for the near infrared band. It was concluded that reasonable results could be obtained by using speetrally stable targets.
IKONOS image has been wildly used in city planning, precision agriculture and emergence response. However, the accuracy of IKONOS Geo product is limited due to distortion caused by terrain relief. Orthorectification was performed to remove the distortion and the impact of different DEM on orthorectification were evaluated. 38 ground control points (GCPs) and 25 independent check points (ICPs) were collected. DEMs were generated from 1∶10 000 and 1∶50 000 topographic maps. Results show that RMS error at the check points is 1.554 0 m using DEM generated from 1∶10 000 topographic map, which can meet the accuracy requirement of IKONOS Precision product (1.9 m RMSE). While RMS error is 2.572 4 m using DEM generated from 1∶50 000 topographic map.
