Machine Learning and Computer Vision Techniques in Geosciences: Laboratory and Field Applications

A special issue of Minerals (ISSN 2075-163X). This special issue belongs to the section "Crystallography and Physical Chemistry of Minerals & Nanominerals".

Deadline for manuscript submissions: closed (13 May 2022) | Viewed by 3237

Special Issue Editors


E-Mail Website
Guest Editor
Laboratorio de Geotecnia, CEDEX, Madrid, Spain
Interests: rock mechanics; mining; machine-learning techniques; image processing; petrography; arduino/raspberry Pi in geosciences
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Department of Applied Mathematics I, University of Vigo, Vigo, Spain
Interests: machine learning techniques: applications and new algorithms; functional statistics: outliers detection and quality control; image processing;
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Laboratorio de Geotecnia, CEDEX, Madrid, Spain
Interests: rock mechanics; geology; slope stability; crystallography; arduino/raspberry Pi in geosciences;

Special Issue Information

Dear Colleagues,

We kindly invite you to contribute to a new Special Issue for Minerals, which will be focused on applications of machine learning and computer vision in geosciences, with particular dedication to the fields of mineralogy and petrology.

Mineralogy and petrology stand as the basis of geosciences and are fundamental to a proper understanding of the formation, development, and exploitation of mineral deposits, as well as oil, gas, and geothermal resources. Features such as texture, internal structure, mechanical properties, and composition may affect, at different scales, the behavior of minerals and rocks, which are essential resources in building and manufacturing processes.

The ability to recognize structures and patterns at different scales is fundamental, not only for a correct identification of minerals and rocks, but also for the determination of parameters that influence the exploitation of natural resources, such as porosity or granulometry. Even though the role of experts may never be replaced, it is true that there are many techniques developed so far that have helped and optimized classically manual processes in geosciences, such as mineral and rock identification through thin-section microscopy or rock-mass structure recognition. These advances are backboned by machine learning and computer vision techniques.

Moreover, the availability of 3D resources has proved to be very useful in the field of education in geosciences, where access to real-life environments is not always easy due to time and cost limitations. The combination of computer vision techniques with virtual reality allows the possibility of having available 3D virtual replicas of spaces and materials (like mineral and rock specimens) that represent an interesting potential for educational purposes.

In this Special Issue, we invite researchers to contribute with developments related to the application of machine learning and computer vision techniques in the field of geosciences, with particular emphasis on mineralogy and petrology.

Dr. Ignacio Pérez-Rey
Prof. Dr. Javier Martínez
Dr. Mauro Muñiz-Menéndez
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Minerals is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • mineralogy
  • petrology
  • petrography
  • machine learning
  • computer vision
  • mining engineering
  • 3D point cloud

Published Papers (1 paper)

Order results
Result details
Select all
Export citation of selected articles as:

Research

16 pages, 2447 KiB  
Communication
Mineral Photos Recognition Based on Feature Fusion and Online Hard Sample Mining
by Liqin Jia, Mei Yang, Fang Meng, Mingyue He and Hongmin Liu
Minerals 2021, 11(12), 1354; https://doi.org/10.3390/min11121354 - 30 Nov 2021
Cited by 9 | Viewed by 1898
Abstract
Mineral recognition is of importance in geological research. Traditional mineral recognition methods need professional knowledge or special equipment, are susceptible to human experience, and are inconvenient to carry in some conditions such as in the wild. The development of computer vision provides a [...] Read more.
Mineral recognition is of importance in geological research. Traditional mineral recognition methods need professional knowledge or special equipment, are susceptible to human experience, and are inconvenient to carry in some conditions such as in the wild. The development of computer vision provides a possibility for convenient, fast, and intelligent mineral recognition. Recently, several mineral recognition methods based on images using a neural network have been proposed for this aim. However, these methods do not exploit features extracted from the backbone network or available information of the samples in the mineral dataset sufficiently, resulting in low recognition accuracy. In this paper, a method based on feature fusion and online hard sample mining is proposed to improve recognition accuracy by using only mineral photo images. This method first fuses multi-resolution features extracted from ResNet-50 to obtain comprehensive information of mineral photos, and then proposes the weighted top-k loss to emphasize the learning of hard samples. Based on a dataset consisting of 14,986 images of 22 common minerals, the proposed method with 10-fold cross-validation achieves a Top1 accuracy of 88.01% on the validation image set, surpassing those of Inception-v3 and EfficientNet-B0 by a margin of 1.88% and 1.29%, respectively, which demonstrates the good prospect of the proposed method for convenient and reliable mineral recognition using mineral photos only. Full article
Show Figures

Figure 1

Back to TopTop