Scribd
Upload
16 views3 pages

Chapter 1

This study focuses on developing a machine learning model using deep learning algorithms to automate core logging processes in mineral exploration, aiming to enhance the efficiency and accuracy of geological data collection. It seeks to utilize publicly available core images to predict rock types and calculate rock quality designation, thereby assisting geologists in their work. The research addresses the challenges of traditional core logging methods and aims to improve the understanding of mineral deposits through advanced data analysis.

Uploaded by

Tchouleko judicael
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
16 views3 pages

Chapter 1

This study focuses on developing a machine learning model using deep learning algorithms to automate core logging processes in mineral exploration, aiming to enhance the efficiency and accuracy of geological data collection. It seeks to utilize publicly available core images to predict rock types and calculate rock quality designation, thereby assisting geologists in their work. The research addresses the challenges of traditional core logging methods and aims to improve the understanding of mineral deposits through advanced data analysis.

Uploaded by

Tchouleko judicael
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

CHAPTER 1 INTRODUCTION

1.1 Background of study

With the advent of Deep Neural network architectures the need for rapid and
accurate data collection in core logging has risen, with machine learning being
brought as a tool to work hand in hand with geologists to optimize logging
processes, generate more accurate and organized geodatabases, and help them
better understand the configuration and dynamics of mineral deposits. Systems
based on this have been developed around the world in recent years and are
being used , but a lot of skeptics arise at the level of robustness of such methods,
their accessibility and whether or not the can be effectively incorporated in the
working environment of the geologist. Over the past decades , machine learning
algorithms have been applied in various industries to carry rapid and automatic
processes saving time , money, ensuring the acquisition of accurate data
therefore increasing productivity. In mineral exploration , machine learning can be
an asset , to be used in automating core logging processes since they are based
on visual observations; These algorithms are trained from large datasets
extracted from cores to learn how to make predictions from core images. With
artificial intelligence constantly evolving and big data too, outdated core analysis
methods( manual) are being left out to embrace digital logging using advanced
machine learning algorithms. To this end, this study aims therefore is aimed at
developing a Machine learning model capable of predicting geological
features( rock types) and calculate the rock quality designation of cores from core
images in other to help geologist to better understand the geology of sites under
investigation all these using publicly available Core images from the National
Virtual Core library of the Australian government, with future aim to optimize,
meet and surpass present industry standards.
1.2 Statement of the problem and rationale

Due to the fact that Core logging has always been a tedious and laborious process
often covering considerable amounts of time and energy in exploration phases,
this study aims make use of a Deep learning algorithms, convolutional Neural
Network Architecture to rapidly and efficiently collect geological/geotechnical data
making use of core data from south Australian exploration and mining companies
in order to aid geologists in their working environment to accomplish their job
efficiently.
The collected data will help mining engineers and geologist to have an insight of
the site under investigation, by identifying different rock types and extracting
geotechnical data necessary in the design processes.

1.3 Research questions/hypotheses


1.3.1 Main research question

Can machine learning help in automating and facilitating core logging processes?

1.3.2 Specific research questions

Where is data gotten and how can it be prepared, used to automate logging?

How is the model prepared and trained to predict lithologies and calculate the
Rock quality designation?

After evaluation of the obtained results/predictions, how can the models


performance be improved?

How can the obtained results from such predictions help in understanding the
configuration and dynamics of mineral deposits?
1.4 Objectives
1.4.1 Overall Objective
Was to develop a machine learning model capable of identifying various rock type
and perform RQD calculation in a fully automated manner.

1.4 Specific Objectives


Were to:

 Generate a dataset from core images

 Preprocess the data by applying various augmentations methods on the


data to increase the ability of the generated data to be able to identify
features anf performed the desired calculations on cores

 Generate a dataset splitted into training, testing and validation datasets


respectively

 train the model to classify rock types and perform RQD calculations

 Test unseen core images to evaluate the models performance to identify


new or unseen image

1.5 Significance of study

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy