Assessing of land cover classification methods for Ha Giang province using Sentinel-2 satellite imagery

Abstract

    The study aims to compare the classification accuracy of vegetation cover in Hà Giang province using Sentinel-2 satellite imagery 2024 through two classification methods: Support Vector Machine (SVM) and Maximum Likelihood Classification (MLC). A comparison of the two methods showed that SVM (OA = 83,59%, K = 0,815) exhibited superior accuracy compared to MLC (OA = 78,36%, K = 0,756) when classifying vegetation cover using Sentinel-2 satellite imagery. The classification results for Hà Giang province in 2024, with 9 vegetation cover types, indicated that the majority of the area is forest land, followed by agricultural land (annual crops and perennial crops), bare land, residential land, and finally water bodies.