The application of Sentinel 2B satellite imagery using Supervised Image Classification of Maximum Likelihood Algorithm in Landcover Updating of the Mamminasata Metropolitan Area, South Sulawesi

Abstract

Abstract Mamminasata Metropolitan Area includes the City of Makassar, some sub-Districts of Maros, Gowa and Takalar Regencies. This metropolitan area were formed based on the Governor of South Sulawesi Province Decree in 2003 with a total area of 246,230 ha. Sentinel-2B is a European optical imaging satellite that was launched on 7 March 2017. It is the second Sentinel-2 satellite launched as part of the European Space Agency’s Copernicus Programme, and its orbit phases 180° degrees against Sentinel-2A. The satellite carries a wide swath high-resolution multispectral imager with 13 spectral bands. It provides information for agriculture and forestry, among others allowing for prediction of crop yields. Multispectral classification process can be divided into two types namely supervised and unsupervised classification. In this research, supervised classification was applied which includes a set of algorithms that are based on the entry of sample objects and geographic location. Maximum likelihood algorithm is the statistically most established algorithm. Other algorithm is based on the measurement of the distance between the coordinates of the sample group and the pixel coordinates candidate while the maximum likelihood algorithm using probability calculation. In this algorithm, pixels were classified as specific objects not because of their Euclidean distance, but by the shape, size and orientation of the sample on the feature space. Maximum likelihood algorithm works the following way, program briefly marks each pixel which has a measurement of the pattern or appearance of X into class i whose unit is most likely to be grouped as a vector X. Foundation of the maximum likelihood algorithm uses a probability value of a pixel X to be a member of a particular class or a particular label.

Access to Document

Other files and links

• Link to publication in Scopus
• Open Access Version Available

Fingerprint