It is well known that remote sensed scenes could be affected by many factors and, for optimum change detection, these unwanted effects must be removed. In this study a new algorithm is proposed for PIF (Pseudo Invariant Features) extraction and relative radiometric normalization. The new algorithm can be labeled as a supervised one and combines three methods for the detection of PIFs: Moment distance index (MDI), Normalized Difference Vegetation Index (NDVI) masks morphological erosion and dilate operators. In order to prove its effectiveness, the algorithm was tested by using Landsat 8 scenes of the “Mar de Plstico” landscape of the Andalusian Almería. Many tests were performed in order to provide a set of valid input parameters for the chosen environments. Lastly, the results were statistically assessed with parametric and non-parametric tests showing very good and stable results in the four different study areas.
A new threshold relative radiometric correction algorithm (TRRCA) of multiband satellite data / Novelli, Antonio; Aguilar, Manuel A.; Tarantino, Eufemia (SMART INNOVATION, SYSTEMS AND TECHNOLOGIES). - In: Intelligent interactive multimedia systems and services 2017 / [a cura di] Giuseppe De Pietro, Luigi Gallo, Robert J. Howlett, Lakhmi C. Jain. - ELETTRONICO. - Cham, CH : Springer, 2018. - ISBN 978-3-319-59479-8. - pp. 41-50 [10.1007/978-3-319-59480-4_5]
A new threshold relative radiometric correction algorithm (TRRCA) of multiband satellite data
NOVELLI, Antonio;TARANTINO, Eufemia
2018
Abstract
It is well known that remote sensed scenes could be affected by many factors and, for optimum change detection, these unwanted effects must be removed. In this study a new algorithm is proposed for PIF (Pseudo Invariant Features) extraction and relative radiometric normalization. The new algorithm can be labeled as a supervised one and combines three methods for the detection of PIFs: Moment distance index (MDI), Normalized Difference Vegetation Index (NDVI) masks morphological erosion and dilate operators. In order to prove its effectiveness, the algorithm was tested by using Landsat 8 scenes of the “Mar de Plstico” landscape of the Andalusian Almería. Many tests were performed in order to provide a set of valid input parameters for the chosen environments. Lastly, the results were statistically assessed with parametric and non-parametric tests showing very good and stable results in the four different study areas.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
