Automated detection of geological landforms on Mars using Convolutional Neural Networks

Leon F. Palafox, Christopher W. Hamilton, Stephen P. Scheidt, Alexander M. Alvarez

Research output: Research - peer-reviewArticle

  • 1 Citations

Abstract

The large volume of high-resolution images acquired by the Mars Reconnaissance Orbiter has opened a new frontier for developing automated approaches to detecting landforms on the surface of Mars. However, most landform classifiers focus on crater detection, which represents only one of many geological landforms of scientific interest. In this work, we use Convolutional Neural Networks (ConvNets) to detect both volcanic rootless cones and transverse aeolian ridges. Our system, named MarsNet, consists of five networks, each of which is trained to detect landforms of different sizes. We compare our detection algorithm with a widely used method for image recognition, Support Vector Machines (SVMs) using Histogram of Oriented Gradients (HOG) features. We show that ConvNets can detect a wide range of landforms and has better accuracy and recall in testing data than traditional classifiers based on SVMs.

LanguageEnglish (US)
Pages48-56
Number of pages9
JournalComputers and Geosciences
Volume101
DOIs
StatePublished - Apr 1 2017

Fingerprint

landform
Mars
detection
Landforms
Neural networks
support vector machine
Support vector machines
Classifiers
image resolution
histogram
crater
method
Image recognition
Image resolution
Cones
Testing

Keywords

  • Convolutional neural networks
  • Mars
  • Support vector machines
  • Transverse aeolian ridges
  • Volcanic rootless cones

ASJC Scopus subject areas

  • Information Systems
  • Computers in Earth Sciences

Cite this

Automated detection of geological landforms on Mars using Convolutional Neural Networks. / Palafox, Leon F.; Hamilton, Christopher W.; Scheidt, Stephen P.; Alvarez, Alexander M.

In: Computers and Geosciences, Vol. 101, 01.04.2017, p. 48-56.

Research output: Research - peer-reviewArticle

Palafox, Leon F. ; Hamilton, Christopher W. ; Scheidt, Stephen P. ; Alvarez, Alexander M./ Automated detection of geological landforms on Mars using Convolutional Neural Networks. In: Computers and Geosciences. 2017 ; Vol. 101. pp. 48-56
@article{c576eea0f69c4954a8938eccd7758444,
title = "Automated detection of geological landforms on Mars using Convolutional Neural Networks",
abstract = "The large volume of high-resolution images acquired by the Mars Reconnaissance Orbiter has opened a new frontier for developing automated approaches to detecting landforms on the surface of Mars. However, most landform classifiers focus on crater detection, which represents only one of many geological landforms of scientific interest. In this work, we use Convolutional Neural Networks (ConvNets) to detect both volcanic rootless cones and transverse aeolian ridges. Our system, named MarsNet, consists of five networks, each of which is trained to detect landforms of different sizes. We compare our detection algorithm with a widely used method for image recognition, Support Vector Machines (SVMs) using Histogram of Oriented Gradients (HOG) features. We show that ConvNets can detect a wide range of landforms and has better accuracy and recall in testing data than traditional classifiers based on SVMs.",
keywords = "Convolutional neural networks, Mars, Support vector machines, Transverse aeolian ridges, Volcanic rootless cones",
author = "Palafox, {Leon F.} and Hamilton, {Christopher W.} and Scheidt, {Stephen P.} and Alvarez, {Alexander M.}",
year = "2017",
month = "4",
doi = "10.1016/j.cageo.2016.12.015",
volume = "101",
pages = "48--56",
journal = "Computers and Geosciences",
issn = "0098-3004",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Automated detection of geological landforms on Mars using Convolutional Neural Networks

AU - Palafox,Leon F.

AU - Hamilton,Christopher W.

AU - Scheidt,Stephen P.

AU - Alvarez,Alexander M.

PY - 2017/4/1

Y1 - 2017/4/1

N2 - The large volume of high-resolution images acquired by the Mars Reconnaissance Orbiter has opened a new frontier for developing automated approaches to detecting landforms on the surface of Mars. However, most landform classifiers focus on crater detection, which represents only one of many geological landforms of scientific interest. In this work, we use Convolutional Neural Networks (ConvNets) to detect both volcanic rootless cones and transverse aeolian ridges. Our system, named MarsNet, consists of five networks, each of which is trained to detect landforms of different sizes. We compare our detection algorithm with a widely used method for image recognition, Support Vector Machines (SVMs) using Histogram of Oriented Gradients (HOG) features. We show that ConvNets can detect a wide range of landforms and has better accuracy and recall in testing data than traditional classifiers based on SVMs.

AB - The large volume of high-resolution images acquired by the Mars Reconnaissance Orbiter has opened a new frontier for developing automated approaches to detecting landforms on the surface of Mars. However, most landform classifiers focus on crater detection, which represents only one of many geological landforms of scientific interest. In this work, we use Convolutional Neural Networks (ConvNets) to detect both volcanic rootless cones and transverse aeolian ridges. Our system, named MarsNet, consists of five networks, each of which is trained to detect landforms of different sizes. We compare our detection algorithm with a widely used method for image recognition, Support Vector Machines (SVMs) using Histogram of Oriented Gradients (HOG) features. We show that ConvNets can detect a wide range of landforms and has better accuracy and recall in testing data than traditional classifiers based on SVMs.

KW - Convolutional neural networks

KW - Mars

KW - Support vector machines

KW - Transverse aeolian ridges

KW - Volcanic rootless cones

UR - http://www.scopus.com/inward/record.url?scp=85011700169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011700169&partnerID=8YFLogxK

U2 - 10.1016/j.cageo.2016.12.015

DO - 10.1016/j.cageo.2016.12.015

M3 - Article

VL - 101

SP - 48

EP - 56

JO - Computers and Geosciences

T2 - Computers and Geosciences

JF - Computers and Geosciences

SN - 0098-3004

ER -