Robust Radiometric Normalization of the Near Equatorial Satellite Images Using Feature Extraction and Remote Sensing Analysis

Relative radiometric normalization (RRN) minimizes differences among images caused by inconsistencies of acquisition conditions rather than changes in surface. Scale invariant feature transform (SIFT) has the ability to automatically extract control points (CPs) and is commonly used for remote sensing images. However, its results are mostly inaccurate sometimes contain incorrect matching generating a small number false CP pairs. These pairs have high alarm matching. This paper presents modified method improve performance SIFT CPs applying sum absolute difference (SAD) different manner new optical satellite generation called near-equatorial orbit multi-sensor The proposed method, which significantly rate correct matches, improves data this study were obtained from RazakSAT near equatorial system. involves six steps: 1) reduction, 2) CPs, 3) refining using SAD algorithm with empirical threshold, 4) calculation true intensity values over all image’ bands, 5) preforming linear regression model between locate reverence sensed 6) conducting transformation functions. Different thresholds experimentally tested (50 70), followed it removed extracted be 775, 1125, 883, 804, 883 681 342, 424, 547, 706, 469 corrected matched pairs, respectively.