Artificial Intelligence Remote Sensing for Open-pit Mining Detection in the Tropical Environment of Indonesia

Main Article Content

Fajar Yulianto, Parwati Sofan, Gatot Nugroho, Suwarsono

Abstract

The purpose of this study is to build an Artificial Intelligence Remote Sensing model for Open-pit Mining Detection in the Tropical Environment of Indonesia. This is done based on the characteristics of the climate and environmental conditions in Indonesia which are humid or tropical and have vegetation. There are various challenges and variations of surface objects in the field that are similar to mining objects, as well as following the development of satellite image classification methods that are currently being developed. In this study, the classification process for Open-pit Mining Detection is carried out by applying the Random Forest (RF) algorithm. The result of accuracy assessment based on the reference availability of SPOT 6/7 data which consists of Procedure Accuracy (PA), User Accuracy (UA), and Overall Accuracy (OA) are 90.41%; 84.21%, and 72.03% respectively.

Article Details

Section
Articles