AAU Student Projects - visit Aalborg University's student projects portal
A master thesis from Aalborg University

Estimation of illegal small-scale mining in Ghana using a machine learning method and open-source satellite imagery

Author(s)

Term

4. term

Education

Publication year

2022

Submitted on

2022-06-03

Pages

53 pages

Abstract

Ghana is considered one of the largest gold-producing countries in the world and is ranked first in the African continent. This nation produces vast quantities of gold through Galamsey (illegal small-scale gold mining), which is a very popular but unregulated technique for mineral extraction in southern Ghana and is the main source of income for many Ghanaians. Over the past decade, Galamsey has grown tremendously, causing a concerning and noticeable degradation of the environment and posing a real threat to people’s lives. Attempts have been made to detect and map these illicit mining operations in the past, but a high-quality map identifying the distribution patterns of Galamsey has not yet been conducted. In this study, a highly detailed Galamsey identification map of the entire country of Ghana was produced using an image recognition deep learning method [1]. More specifically, the regression ML approach to detect Galamsey features followed a Convolutional Neural Network (CNN) algorithm with an Inception- ResNet -like architecture. Predictions were computed using the current best available resolution for open-source satellite images (10m), and Sentinel-1 and -2 products were processed to train the ML model for a three months period (November 1- 2021 to February 1- 2022). Even though further studies should include more algorithm testing, this pixel-based method delivered good results, achieving a binary accuracy of nearly 90%. Model predictions have shown that illegal mining is concentrated in four main regions in Ghana, being Western Ghana the hotspot for unlicensed artisanal miners. Galamsey spatial distribution is characterized by clusters along the ramification of main water streams of the country, degrading forest reserves that are protected at a national level. The ML method presented in this study serves as a valuable tool for identifying unauthorized gold mining activities and is therefore of considerable significance for government decision-makers and stakeholders that are involved in law-making practices against Galamsey.

Documents


Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.

If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.