| Abstrak/Abstract |
The atmospheric disturbance in remote sensing imagery greatly influences the object's spectral response in the imagery. This, in turn, will impact the object characterization. The atmospheric effects on remote sensing imagery can be reduced through atmospheric correction. There are various types of atmospheric correction methods and each of them has its own working principles. Daerah Istimewa Yogyakarta (DIY) Province, Indonesia, was chosen to be study area for this research. The research objectives are to evaluate the atmospheric correction method, which consist of Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH), Quick Atmospheric Correction (QUAC), Dark Object Subtraction (DOS), Second Simulation of the Satellite Signal in the Solar Spectrum (6S), Atmospheric Correction (ATCOR2), and Landsat 8 Surface Reflectance Code (LaSRC) by NASA. The compared objects consist of water, vegetation, and soil objects. The evaluation was based on Standard Error of Estimate (SEE), accuracy, and curve pattern. The result shows that the best atmospheric correction varies on each object. The spectral response curve pattern shows similarity but each object has its own accurate atmospheric method based on SEE result. The FLAASH, 6s, and ATCOR2 method show the lowest SEE result for mature vegetation leaves, beach sand, sand suns, and, lagoon, while QUAC method shows the lowest SEE result for young vegetation leaves, paddy plants, grass, and reservoir. |
