Object Based Classification for Land Cover Mapping using Sentinel -1Ain Yogyakarta

摘要

Information on tte latest land cover can be obtained from remote sensing data processing, but many do still use optical imagery. In fact, Indonesia is the next country tropical so that the possibility of disrupted cloud cover throughout the year is enormous. Therefore, it is necessary that imagery can penetrate the cloud cover of the radar image. Behind the excess, there is a weakness of radar imagery is a salt and paper disorder that appears bitnik-black and white spots on imagery. This Salt and paper can have a pixel value that is therefore less precise when a pixel-based classification is performed. It is necessary to do a classification that not only considers the pixel value of object-based classification. This research aims to segment objects in the form of land cover and calculate the accuracy of the segmentation results.This research was conducted using a radar image, namely Citra Sentinel-1 A with 10m×10m resolution. The segmentation process carried out using a multiresolution segmentation algorism. Based on the results of the study, the best segmentation has an input channel parameter weight of 1, 0.5, 1, output parameter weight 25, shape parameter weight 0.3 and compactness parameter weight 0.9. The value of segmentation accuracy produced by considering five parameters in the shape of oversegmentation (OSeg), undersegmentation (USeg), root mean square error (D), area fit index (AFI), and quality rate (Qr) is 57%. Low accuracy value because radar images focus on object morphology in the shape of altitude and surtace conditions. Whereas in a land cover the objects morphology can vary and surtace roughness can vary.