APPLICATIONS OF ARTIFICIAL NEURAL NETWORK IN LAND COVER CLASSIFICATION Case study for the coastal area of Hoai Nhon district, Binh Dinh province

Abstract

Land cover classification has a significantrole in natural resource management. The artificial neutral network has been successfully applied in the identification and classification of some objects including land covers. In this paper, the artificial neutral network was applied toclassify land covers of the coastal area in HoaiNhon district, BinhDinh province. The research results illustrated that the land covers in the research area are classified into 9 different types. The classification results of this method were compared with two other popular classification methods namely maximum likelihood and minimum distance.The comparison shown that classification by artificial neural network has higher accuracy than by maximum likelihood and minimum distance with the appropriate number of neural-network training repetitions.