RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation

Supervised methods for object delineation in remote sensing require labeled ground-truth data. Gathering sufficient high quality ground-truth data is difficult, especially when targets are of irregular shape or difficult to distinguish from background or neighboring objects. Tree crown delineation p...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Dylan Stewart, Alina Zare, Sergio Marconi, Ben G. Weinstein, Ethan P. White, Sarah J. Graves, Stephanie A. Bohlman, Aditya Singh
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Imprecise labels quantitative evaluation remote sensing tree crown delineation Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/b3004e22a27c4369a7331fd04e2a81c8
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b3004e22a27c4369a7331fd04e2a81c8
record_format	dspace
spelling	oai:doaj.org-article:b3004e22a27c4369a7331fd04e2a81c82021-11-17T00:00:13ZRandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation2151-153510.1109/JSTARS.2021.3122345https://doaj.org/article/b3004e22a27c4369a7331fd04e2a81c82021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9585420/https://doaj.org/toc/2151-1535Supervised methods for object delineation in remote sensing require labeled ground-truth data. Gathering sufficient high quality ground-truth data is difficult, especially when targets are of irregular shape or difficult to distinguish from background or neighboring objects. Tree crown delineation provides key information from remote sensing images for forestry, ecology, and management. However, tree crowns in remote sensing imagery are often difficult to label and annotate due to irregular shape, overlapping canopies, shadowing, and indistinct edges. There are also multiple approaches to annotation in this field (e.g., rectangular boxes vs. convex polygons) that further contribute to annotation imprecision. However, current evaluation methods do not account for this uncertainty in annotations, and quantitative metrics for evaluation can vary across multiple annotators. In this article, we address these limitations by developing an adaptation of the Rand index (RI) for weakly labeled crown delineation that we call RandCrowns (RC). Our new RC evaluation metric provides a method to appropriately evaluate delineated tree crowns while taking into account imprecision in the ground-truth delineations. The RC metric reformulates the RI by adjusting the areas over which each term of the index is computed to account for uncertain and imprecise object delineation labels. Quantitative comparisons to the commonly used intersection over union method show a decrease in the variance generated by differences among multiple annotators. Combined with qualitative examples, our results suggest that the RC metric is more robust for scoring target delineations in the presence of uncertainty and imprecision in annotations that are inherent to tree crown delineation.Dylan StewartAlina ZareSergio MarconiBen G. WeinsteinEthan P. WhiteSarah J. GravesStephanie A. BohlmanAditya SinghIEEEarticleImprecise labelsquantitative evaluationremote sensingtree crown delineationOcean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11229-11239 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Imprecise labels quantitative evaluation remote sensing tree crown delineation Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Imprecise labels quantitative evaluation remote sensing tree crown delineation Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Dylan Stewart Alina Zare Sergio Marconi Ben G. Weinstein Ethan P. White Sarah J. Graves Stephanie A. Bohlman Aditya Singh RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
description	Supervised methods for object delineation in remote sensing require labeled ground-truth data. Gathering sufficient high quality ground-truth data is difficult, especially when targets are of irregular shape or difficult to distinguish from background or neighboring objects. Tree crown delineation provides key information from remote sensing images for forestry, ecology, and management. However, tree crowns in remote sensing imagery are often difficult to label and annotate due to irregular shape, overlapping canopies, shadowing, and indistinct edges. There are also multiple approaches to annotation in this field (e.g., rectangular boxes vs. convex polygons) that further contribute to annotation imprecision. However, current evaluation methods do not account for this uncertainty in annotations, and quantitative metrics for evaluation can vary across multiple annotators. In this article, we address these limitations by developing an adaptation of the Rand index (RI) for weakly labeled crown delineation that we call RandCrowns (RC). Our new RC evaluation metric provides a method to appropriately evaluate delineated tree crowns while taking into account imprecision in the ground-truth delineations. The RC metric reformulates the RI by adjusting the areas over which each term of the index is computed to account for uncertain and imprecise object delineation labels. Quantitative comparisons to the commonly used intersection over union method show a decrease in the variance generated by differences among multiple annotators. Combined with qualitative examples, our results suggest that the RC metric is more robust for scoring target delineations in the presence of uncertainty and imprecision in annotations that are inherent to tree crown delineation.
format	article
author	Dylan Stewart Alina Zare Sergio Marconi Ben G. Weinstein Ethan P. White Sarah J. Graves Stephanie A. Bohlman Aditya Singh
author_facet	Dylan Stewart Alina Zare Sergio Marconi Ben G. Weinstein Ethan P. White Sarah J. Graves Stephanie A. Bohlman Aditya Singh
author_sort	Dylan Stewart
title	RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
title_short	RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
title_full	RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
title_fullStr	RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
title_full_unstemmed	RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation
title_sort	randcrowns: a quantitative metric for imprecisely labeled tree crown delineation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/b3004e22a27c4369a7331fd04e2a81c8
work_keys_str_mv	AT dylanstewart randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT alinazare randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT sergiomarconi randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT bengweinstein randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT ethanpwhite randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT sarahjgraves randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT stephanieabohlman randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation AT adityasingh randcrownsaquantitativemetricforimpreciselylabeledtreecrowndelineation
_version_	1718426069127659520

RandCrowns: A Quantitative Metric for Imprecisely Labeled Tree Crown Delineation

Ejemplares similares