Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems

Tropical rainforests worldwide are negatively impacted from a variety of human-caused threats. Unfortunately, our ability to study these rainforests is impeded by logistical problems such as their physical inaccessibility, expensive aerial imagery, and/or coarse satellite data. One solution is the u...

Full description

Bibliographic Details
Main Author: Epperson, Matthew
Format: Doctoral or Postdoctoral Thesis
Language:English
Published: eScholarship, University of California 2018
Subjects:
CNN
Online Access:http://www.escholarship.org/uc/item/5rh083rc
id ftcdlib:qt5rh083rc
record_format openpolar
spelling ftcdlib:qt5rh083rc 2023-05-15T15:11:29+02:00 Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems Epperson, Matthew 57 2018-01-01 application/pdf http://www.escholarship.org/uc/item/5rh083rc en eng eScholarship, University of California http://www.escholarship.org/uc/item/5rh083rc qt5rh083rc public Epperson, Matthew. (2018). Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems. UC San Diego: Electrical Engineering (Intelsys, Robotics and Cont). Retrieved from: http://www.escholarship.org/uc/item/5rh083rc Computer science Electrical engineering Wildlife conservation CNN computer vision conservation deep learning ecology machine learning dissertation 2018 ftcdlib 2018-05-18T22:51:55Z Tropical rainforests worldwide are negatively impacted from a variety of human-caused threats. Unfortunately, our ability to study these rainforests is impeded by logistical problems such as their physical inaccessibility, expensive aerial imagery, and/or coarse satellite data. One solution is the use of low-cost, Unmanned Aerial Vehicles (UAV), commonly referred to as drones. Drones are now widely recognized as a tool for ecology, environmental science, and conservation, collecting imagery that is superior to satellite data in resolution. We asked: Can we take advantage of the sub-meter, high-resolution imagery to detect specific tree species or groups, and use these data as indicators of rainforest functional traits and characteristics? We demonstrate a low-cost method for obtaining high-resolution aerial imagery in a rainforest of Belize using a drone over three sites in two rainforest protected areas. We built a workflow that uses Structure from Motion (SfM) on the drone images to create a large orthomosaic and a Deep Convolutional Neural Network (CNN) to classify indicator tree species. We selected: 1) Cohune Palm (Attalea cohune) as they are indicative of past disturbance and current soil condition; and, 2) the dry-season deciduous tree group since deciduousness is an important ecological factor of rainforest structure and function. This framework serves as a guide for tackling difficult ecological challenges and we show two additionally examples of how a similar architecture can help count wildlife populations in the Arctic. Doctoral or Postdoctoral Thesis Arctic University of California: eScholarship Arctic
institution Open Polar
collection University of California: eScholarship
op_collection_id ftcdlib
language English
topic Computer science
Electrical engineering
Wildlife conservation
CNN
computer vision
conservation
deep learning
ecology
machine learning
spellingShingle Computer science
Electrical engineering
Wildlife conservation
CNN
computer vision
conservation
deep learning
ecology
machine learning
Epperson, Matthew
Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
topic_facet Computer science
Electrical engineering
Wildlife conservation
CNN
computer vision
conservation
deep learning
ecology
machine learning
description Tropical rainforests worldwide are negatively impacted from a variety of human-caused threats. Unfortunately, our ability to study these rainforests is impeded by logistical problems such as their physical inaccessibility, expensive aerial imagery, and/or coarse satellite data. One solution is the use of low-cost, Unmanned Aerial Vehicles (UAV), commonly referred to as drones. Drones are now widely recognized as a tool for ecology, environmental science, and conservation, collecting imagery that is superior to satellite data in resolution. We asked: Can we take advantage of the sub-meter, high-resolution imagery to detect specific tree species or groups, and use these data as indicators of rainforest functional traits and characteristics? We demonstrate a low-cost method for obtaining high-resolution aerial imagery in a rainforest of Belize using a drone over three sites in two rainforest protected areas. We built a workflow that uses Structure from Motion (SfM) on the drone images to create a large orthomosaic and a Deep Convolutional Neural Network (CNN) to classify indicator tree species. We selected: 1) Cohune Palm (Attalea cohune) as they are indicative of past disturbance and current soil condition; and, 2) the dry-season deciduous tree group since deciduousness is an important ecological factor of rainforest structure and function. This framework serves as a guide for tackling difficult ecological challenges and we show two additionally examples of how a similar architecture can help count wildlife populations in the Arctic.
format Doctoral or Postdoctoral Thesis
author Epperson, Matthew
author_facet Epperson, Matthew
author_sort Epperson, Matthew
title Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
title_short Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
title_full Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
title_fullStr Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
title_full_unstemmed Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems
title_sort empowering conservation through deep convolutional neural networks and unmanned aerial systems
publisher eScholarship, University of California
publishDate 2018
url http://www.escholarship.org/uc/item/5rh083rc
op_coverage 57
geographic Arctic
geographic_facet Arctic
genre Arctic
genre_facet Arctic
op_source Epperson, Matthew. (2018). Empowering Conservation through Deep Convolutional Neural Networks and Unmanned Aerial Systems. UC San Diego: Electrical Engineering (Intelsys, Robotics and Cont). Retrieved from: http://www.escholarship.org/uc/item/5rh083rc
op_relation http://www.escholarship.org/uc/item/5rh083rc
qt5rh083rc
op_rights public
_version_ 1766342326857236480