USING REMOTE SENSING DATA TO CONSTRUCT LAND COVER MAPS OF CHO DON DISTRICT, BAC KAN PROVINCE

Abstract

The study represents the approach to remote sensing data to land cover classification of Cho Don district, Bac Kan province based on the neural network classification methods. The study is a combination of the supervised classification method based on the training sample file, with the methods of document collection and field investigation. Through the process of remote sensing image interpretation, the land cover system of Cho Don district is divided into 5 layers: crops, rice; residential; street; forest; water. The result of the study is to construct the current land cover map of Cho Don district using Landsat 8 images, The accuracy of the classification results is evaluated based on 2 parameters: overall accuracy and Kappa coefficient: overall accuracy is 99.0458% and Kappa coefficient is 0.988. The research results show the advantages of remote sensing data in the assessment of land cover classification and is an important document in land management, design and land use planning.