Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest

<i>Arecaceae</i> (palms) play a crucial role for native communities and wildlife in the Amazon region. This study presents a first-of-its-kind regional-scale spatial cataloging of palms using remotely sensed data for the country of Guyana. Using very high-resolution satellite images from...

Full description

Saved in:
Bibliographic Details
Main Authors: Matthew J. Drouillard, Anthony R. Cummings
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/24/4642
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846102937074401280
author Matthew J. Drouillard
Anthony R. Cummings
author_facet Matthew J. Drouillard
Anthony R. Cummings
author_sort Matthew J. Drouillard
collection DOAJ
description <i>Arecaceae</i> (palms) play a crucial role for native communities and wildlife in the Amazon region. This study presents a first-of-its-kind regional-scale spatial cataloging of palms using remotely sensed data for the country of Guyana. Using very high-resolution satellite images from the GeoEye-1 and WorldView-2 sensor platforms, which collectively cover an area of 985 km<sup>2</sup>, a total of 472,753 individual palm crowns are detected with F1 scores of 0.76 and 0.79, respectively, using a convolutional neural network (CNN) instance segmentation model. An example of CNN model transference between images is presented, emphasizing the limitation and practical application of this approach. A method is presented to optimize precision and recall using the confidence of the detection features; this results in a decrease of 45% and 31% in false positive detections, with a moderate increase in false negative detections. The sensitivity of the CNN model to the size of the training set is evaluated, showing that comparable metrics could be achieved with approximately 50% of the samples used in this study. Finally, the diameter of the palm crown is calculated based on the polygon identified by mask detection, resulting in an average of 7.83 m, a standard deviation of 1.05 m, and a range of {4.62, 13.90} m for the GeoEye-1 image. Similarly, for the WorldView-2 image, the average diameter is 8.08 m, with a standard deviation of 0.70 m and a range of {4.82, 15.80} m.
format Article
id doaj-art-7ccd4016b4334df48a8b596f0533c14b
institution Kabale University
issn 2072-4292
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-7ccd4016b4334df48a8b596f0533c14b2024-12-27T14:50:46ZengMDPI AGRemote Sensing2072-42922024-12-011624464210.3390/rs16244642Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese RainforestMatthew J. Drouillard0Anthony R. Cummings1Geospatial Information Sciences, School of Economic, Political and Policy Sciences, University of Texas at Dallas, Richardson, TX 75080, USADepartment of Earth and Environmental Sciences, Wesleyan University, Middletown, CT 06459, USA<i>Arecaceae</i> (palms) play a crucial role for native communities and wildlife in the Amazon region. This study presents a first-of-its-kind regional-scale spatial cataloging of palms using remotely sensed data for the country of Guyana. Using very high-resolution satellite images from the GeoEye-1 and WorldView-2 sensor platforms, which collectively cover an area of 985 km<sup>2</sup>, a total of 472,753 individual palm crowns are detected with F1 scores of 0.76 and 0.79, respectively, using a convolutional neural network (CNN) instance segmentation model. An example of CNN model transference between images is presented, emphasizing the limitation and practical application of this approach. A method is presented to optimize precision and recall using the confidence of the detection features; this results in a decrease of 45% and 31% in false positive detections, with a moderate increase in false negative detections. The sensitivity of the CNN model to the size of the training set is evaluated, showing that comparable metrics could be achieved with approximately 50% of the samples used in this study. Finally, the diameter of the palm crown is calculated based on the polygon identified by mask detection, resulting in an average of 7.83 m, a standard deviation of 1.05 m, and a range of {4.62, 13.90} m for the GeoEye-1 image. Similarly, for the WorldView-2 image, the average diameter is 8.08 m, with a standard deviation of 0.70 m and a range of {4.82, 15.80} m.https://www.mdpi.com/2072-4292/16/24/4642palmsremote sensingMask R-CNNGuyanasatellite imagery
spellingShingle Matthew J. Drouillard
Anthony R. Cummings
Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
Remote Sensing
palms
remote sensing
Mask R-CNN
Guyana
satellite imagery
title Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
title_full Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
title_fullStr Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
title_full_unstemmed Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
title_short Regional-Scale Detection of Palms Using VHR Satellite Imagery and Deep Learning in the Guyanese Rainforest
title_sort regional scale detection of palms using vhr satellite imagery and deep learning in the guyanese rainforest
topic palms
remote sensing
Mask R-CNN
Guyana
satellite imagery
url https://www.mdpi.com/2072-4292/16/24/4642
work_keys_str_mv AT matthewjdrouillard regionalscaledetectionofpalmsusingvhrsatelliteimageryanddeeplearningintheguyaneserainforest
AT anthonyrcummings regionalscaledetectionofpalmsusingvhrsatelliteimageryanddeeplearningintheguyaneserainforest