Work

Accuracy Assessment of Unsupervised Land Cover Classification

Year: 2020
Type: article
Abstract: This paper shows the importance of combining remote sensing techniques and GIS tools to quantify the quality of unsupervised classification in addition to determining the land cover classes.In order to quantitative understanding of the allocation of different types of land, it's necessary to perform validation or assessment of the accuracy of the classification.The validation of the unsupervised land cover classification for the valley of river Treska (SJCE vol 9, issue 1) is performed by comparing the corresponding points placed on the reference surface (satellite image), from which the classification is obtained and the thematic map.The validation results showed an overall accuracy of the classification of 89%, while based on the Kappa coefficient -85% which is an indicator of high compatibility of the classified thematic map with the actual ground information.
Language: English
Open Access status: bronze
OpenAlex ID: W4285718436
FWCI: 0.078
Cites: 9
Cited by: 4
Related to: 10