Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery
Deep learning (DL) convolutional neural networks (CNNs) have been rapidly adapted in very high spatial resolution (VHSR) satellite image analysis. DLCNN-based computer visions (CV) applications primarily aim for everyday object detection from standard red, green, blue (RGB) imagery, while earth scie...
Published in: | Journal of Imaging |
---|---|
Main Authors: | , , , , , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2020
|
Subjects: | |
Online Access: | https://doi.org/10.3390/jimaging6090097 |
id |
ftmdpi:oai:mdpi.com:/2313-433X/6/9/97/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2313-433X/6/9/97/ 2023-08-20T04:04:29+02:00 Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery Md Abul Ehsan Bhuiyan Chandi Witharana Anna K. Liljedahl Benjamin M. Jones Ronald Daanen Howard E. Epstein Kelcy Kent Claire G. Griffin Amber Agnew 2020-09-17 application/pdf https://doi.org/10.3390/jimaging6090097 EN eng Multidisciplinary Digital Publishing Institute https://dx.doi.org/10.3390/jimaging6090097 https://creativecommons.org/licenses/by/4.0/ Journal of Imaging; Volume 6; Issue 9; Pages: 97 deep learning tundra ice-wedge polygons Mask R-CNN satellite imagery permafrost Arctic Text 2020 ftmdpi https://doi.org/10.3390/jimaging6090097 2023-08-01T00:07:41Z Deep learning (DL) convolutional neural networks (CNNs) have been rapidly adapted in very high spatial resolution (VHSR) satellite image analysis. DLCNN-based computer visions (CV) applications primarily aim for everyday object detection from standard red, green, blue (RGB) imagery, while earth science remote sensing applications focus on geo object detection and classification from multispectral (MS) imagery. MS imagery includes RGB and narrow spectral channels from near- and/or middle-infrared regions of reflectance spectra. The central objective of this exploratory study is to understand to what degree MS band statistics govern DLCNN model predictions. We scaffold our analysis on a case study that uses Arctic tundra permafrost landform features called ice-wedge polygons (IWPs) as candidate geo objects. We choose Mask RCNN as the DLCNN architecture to detect IWPs from eight-band Worldview-02 VHSR satellite imagery. A systematic experiment was designed to understand the impact on choosing the optimal three-band combination in model prediction. We tasked five cohorts of three-band combinations coupled with statistical measures to gauge the spectral variability of input MS bands. The candidate scenes produced high model detection accuracies for the F1 score, ranging between 0.89 to 0.95, for two different band combinations (coastal blue, blue, green (1,2,3) and green, yellow, red (3,4,5)). The mapping workflow discerned the IWPs by exhibiting low random and systematic error in the order of 0.17–0.19 and 0.20–0.21, respectively, for band combinations (1,2,3). Results suggest that the prediction accuracy of the Mask-RCNN model is significantly influenced by the input MS bands. Overall, our findings accentuate the importance of considering the image statistics of input MS bands and careful selection of optimal bands for DLCNN predictions when DLCNN architectures are restricted to three spectral channels. Text Arctic Ice permafrost Tundra wedge* MDPI Open Access Publishing Arctic Journal of Imaging 6 9 97 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
deep learning tundra ice-wedge polygons Mask R-CNN satellite imagery permafrost Arctic |
spellingShingle |
deep learning tundra ice-wedge polygons Mask R-CNN satellite imagery permafrost Arctic Md Abul Ehsan Bhuiyan Chandi Witharana Anna K. Liljedahl Benjamin M. Jones Ronald Daanen Howard E. Epstein Kelcy Kent Claire G. Griffin Amber Agnew Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
topic_facet |
deep learning tundra ice-wedge polygons Mask R-CNN satellite imagery permafrost Arctic |
description |
Deep learning (DL) convolutional neural networks (CNNs) have been rapidly adapted in very high spatial resolution (VHSR) satellite image analysis. DLCNN-based computer visions (CV) applications primarily aim for everyday object detection from standard red, green, blue (RGB) imagery, while earth science remote sensing applications focus on geo object detection and classification from multispectral (MS) imagery. MS imagery includes RGB and narrow spectral channels from near- and/or middle-infrared regions of reflectance spectra. The central objective of this exploratory study is to understand to what degree MS band statistics govern DLCNN model predictions. We scaffold our analysis on a case study that uses Arctic tundra permafrost landform features called ice-wedge polygons (IWPs) as candidate geo objects. We choose Mask RCNN as the DLCNN architecture to detect IWPs from eight-band Worldview-02 VHSR satellite imagery. A systematic experiment was designed to understand the impact on choosing the optimal three-band combination in model prediction. We tasked five cohorts of three-band combinations coupled with statistical measures to gauge the spectral variability of input MS bands. The candidate scenes produced high model detection accuracies for the F1 score, ranging between 0.89 to 0.95, for two different band combinations (coastal blue, blue, green (1,2,3) and green, yellow, red (3,4,5)). The mapping workflow discerned the IWPs by exhibiting low random and systematic error in the order of 0.17–0.19 and 0.20–0.21, respectively, for band combinations (1,2,3). Results suggest that the prediction accuracy of the Mask-RCNN model is significantly influenced by the input MS bands. Overall, our findings accentuate the importance of considering the image statistics of input MS bands and careful selection of optimal bands for DLCNN predictions when DLCNN architectures are restricted to three spectral channels. |
format |
Text |
author |
Md Abul Ehsan Bhuiyan Chandi Witharana Anna K. Liljedahl Benjamin M. Jones Ronald Daanen Howard E. Epstein Kelcy Kent Claire G. Griffin Amber Agnew |
author_facet |
Md Abul Ehsan Bhuiyan Chandi Witharana Anna K. Liljedahl Benjamin M. Jones Ronald Daanen Howard E. Epstein Kelcy Kent Claire G. Griffin Amber Agnew |
author_sort |
Md Abul Ehsan Bhuiyan |
title |
Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
title_short |
Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
title_full |
Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
title_fullStr |
Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
title_full_unstemmed |
Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery |
title_sort |
understanding the effects of optimal combination of spectral bands on deep learning model predictions: a case study based on permafrost tundra landform mapping using high resolution multispectral satellite imagery |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2020 |
url |
https://doi.org/10.3390/jimaging6090097 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic Ice permafrost Tundra wedge* |
genre_facet |
Arctic Ice permafrost Tundra wedge* |
op_source |
Journal of Imaging; Volume 6; Issue 9; Pages: 97 |
op_relation |
https://dx.doi.org/10.3390/jimaging6090097 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/jimaging6090097 |
container_title |
Journal of Imaging |
container_volume |
6 |
container_issue |
9 |
container_start_page |
97 |
_version_ |
1774714852541988864 |