Translator Disclaimer
10 October 2018 Comparison of scoring, matching, SMCE and geographically weighted regression In malaria vulnerability spatial modelling using satellite imagery: an Indonesian example
Author Affiliations +
Abstract
Malaria is one of deadly infectious diseases commonly found in tropical countries, and until now its preventive efforts are still going on. From a spatial-analytical perspective, the preventive efforts can be done by developing malaria vulnerability maps, which can be used as a basis for risk management. Remotely sensed imagery is a powerful source for collecting relevant spatial data for that purpose. Among various models, there are four analysis methods for generating such maps, i.e. scoring, matching, spatial multi-criteria evaluation (SMCE) and geographically weighted regression (GWR), which have been compared according to their effectiveness and accuracies. The authors tested those methods in Purworejo Regency, Central Java, Indonesia, which has been recognized as a malaria endemic area. This study used Landsat-8 OLI imagery as a basis for deriving spatial parameters closely related to malaria vulnerability . Each vulnerability spatial model’s accuracy was then evaluated by calculating the number of cases found in the field, with respect to each vulnerability class, and then compiling all values using cross tabulation. It was found that, among other methods, the SMCE-based malaria vulnerability map statistically delivered the best result.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Prima Widayani and Projo Danoedoro "Comparison of scoring, matching, SMCE and geographically weighted regression In malaria vulnerability spatial modelling using satellite imagery: an Indonesian example", Proc. SPIE 10783, Remote Sensing for Agriculture, Ecosystems, and Hydrology XX, 107832H (10 October 2018); https://doi.org/10.1117/12.2325680
PROCEEDINGS
6 PAGES


SHARE
Advertisement
Advertisement
Back to Top