Analyzing domain knowledge for big data analysis: A case study with urban tree type classification

Samantha Detor; Abigail Roh; Yiqun Xie; Shashi Shekhar

doi:10.1007/978-3-030-37188-3_11

Analyzing domain knowledge for big data analysis: A case study with urban tree type classification

Samantha Detor, Abigail Roh, Yiqun Xie, Shashi Shekhar

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The goals of this research were to create a labeled dataset of tree shadows and to test the feasibility of shadow-based tree type identification using aerial imagery. Urban tree big data that provides information about individual trees can help city planners optimize positive benefits of urban trees (e.g., increasing wellbeing of city residents) while managing potential negative impacts (e.g., risk to power lines). The continual rise of tree type specific threats, such as emerald ash borer, due to climate change has made this problem more pressing in recent years. However, urban tree big data are time consuming to create. This paper evaluates the potential of a new tree type identification method that utilizes shadows in aerial imagery to survey larger regions of land in a shorter amount of time. This work is challenging because there are structural variations across a given tree type and few verified tree type identification datasets exist. Related work has not explored how tree structure characteristics translate into a profile view of a tree’s shadow or quantified the feasibility of shadow-only based tree type identification. We created a consistent and accurate dataset of 4,613 tree shadows using ground truthing procedures and novel methods for ensuring consistent collection of spatial shadow data that take binary and spatial agreement between raters into account. Our results show that identifying trees from shadows in aerial imagery is feasible and merits further exploration in the future.

Original language	English (US)
Title of host publication	Big Data Analytics - 7th International Conference, BDA 2019, Proceedings
Editors	Sanjay Madria, Philippe Fournier-Viger, Sanjay Chaudhary, P. Krishna Reddy
Publisher	Springer
Pages	176-192
Number of pages	17
ISBN (Print)	9783030371876
DOIs	https://doi.org/10.1007/978-3-030-37188-3_11
State	Published - 2019
Event	7th International Conference on Big Data Analytics, BDA 2019 - Ahmedabad, India Duration: Dec 17 2019 → Dec 20 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11932 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	7th International Conference on Big Data Analytics, BDA 2019
Country/Territory	India
City	Ahmedabad
Period	12/17/19 → 12/20/19

Bibliographical note

Funding Information:
Acknowledgements. This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work.

Funding Information:
This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work.

Publisher Copyright:
© Springer Nature Switzerland AG 2019.

Keywords

Aerial imagery
Ground truthing
Labeled data creation
Smart cities
Tree disease
Urban tree big data
Urban trees

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access

10.1007/978-3-030-37188-3_11

OpenUrl availability

Full text

Cite this

Detor, S., Roh, A., Xie, Y., & Shekhar, S. (2019). Analyzing domain knowledge for big data analysis: A case study with urban tree type classification. In S. Madria, P. Fournier-Viger, S. Chaudhary, & P. K. Reddy (Eds.), Big Data Analytics - 7th International Conference, BDA 2019, Proceedings (pp. 176-192). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11932 LNCS). Springer. https://doi.org/10.1007/978-3-030-37188-3_11

Analyzing domain knowledge for big data analysis: A case study with urban tree type classification. / Detor, Samantha; Roh, Abigail; Xie, Yiqun et al.
Big Data Analytics - 7th International Conference, BDA 2019, Proceedings. ed. / Sanjay Madria; Philippe Fournier-Viger; Sanjay Chaudhary; P. Krishna Reddy. Springer, 2019. p. 176-192 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11932 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Detor, S, Roh, A, Xie, Y & Shekhar, S 2019, Analyzing domain knowledge for big data analysis: A case study with urban tree type classification. in S Madria, P Fournier-Viger, S Chaudhary & PK Reddy (eds), Big Data Analytics - 7th International Conference, BDA 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11932 LNCS, Springer, pp. 176-192, 7th International Conference on Big Data Analytics, BDA 2019, Ahmedabad, India, 12/17/19. https://doi.org/10.1007/978-3-030-37188-3_11

Detor S, Roh A, Xie Y, Shekhar S. Analyzing domain knowledge for big data analysis: A case study with urban tree type classification. In Madria S, Fournier-Viger P, Chaudhary S, Reddy PK, editors, Big Data Analytics - 7th International Conference, BDA 2019, Proceedings. Springer. 2019. p. 176-192. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-37188-3_11

Detor, Samantha ; Roh, Abigail ; Xie, Yiqun et al. / Analyzing domain knowledge for big data analysis : A case study with urban tree type classification. Big Data Analytics - 7th International Conference, BDA 2019, Proceedings. editor / Sanjay Madria ; Philippe Fournier-Viger ; Sanjay Chaudhary ; P. Krishna Reddy. Springer, 2019. pp. 176-192 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{8311a019f8374579bda18862acc858e1,

title = "Analyzing domain knowledge for big data analysis: A case study with urban tree type classification",

abstract = "The goals of this research were to create a labeled dataset of tree shadows and to test the feasibility of shadow-based tree type identification using aerial imagery. Urban tree big data that provides information about individual trees can help city planners optimize positive benefits of urban trees (e.g., increasing wellbeing of city residents) while managing potential negative impacts (e.g., risk to power lines). The continual rise of tree type specific threats, such as emerald ash borer, due to climate change has made this problem more pressing in recent years. However, urban tree big data are time consuming to create. This paper evaluates the potential of a new tree type identification method that utilizes shadows in aerial imagery to survey larger regions of land in a shorter amount of time. This work is challenging because there are structural variations across a given tree type and few verified tree type identification datasets exist. Related work has not explored how tree structure characteristics translate into a profile view of a tree{\textquoteright}s shadow or quantified the feasibility of shadow-only based tree type identification. We created a consistent and accurate dataset of 4,613 tree shadows using ground truthing procedures and novel methods for ensuring consistent collection of spatial shadow data that take binary and spatial agreement between raters into account. Our results show that identifying trees from shadows in aerial imagery is feasible and merits further exploration in the future.",

keywords = "Aerial imagery, Ground truthing, Labeled data creation, Smart cities, Tree disease, Urban tree big data, Urban trees",

author = "Samantha Detor and Abigail Roh and Yiqun Xie and Shashi Shekhar",

note = "Funding Information: Acknowledgements. This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work. Funding Information: This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work. Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019.; 7th International Conference on Big Data Analytics, BDA 2019 ; Conference date: 17-12-2019 Through 20-12-2019",

year = "2019",

doi = "10.1007/978-3-030-37188-3_11",

language = "English (US)",

isbn = "9783030371876",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "176--192",

editor = "Sanjay Madria and Philippe Fournier-Viger and Sanjay Chaudhary and Reddy, {P. Krishna}",

booktitle = "Big Data Analytics - 7th International Conference, BDA 2019, Proceedings",

}

TY - GEN

T1 - Analyzing domain knowledge for big data analysis

T2 - 7th International Conference on Big Data Analytics, BDA 2019

AU - Detor, Samantha

AU - Roh, Abigail

AU - Xie, Yiqun

AU - Shekhar, Shashi

N1 - Funding Information: Acknowledgements. This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work. Funding Information: This study is supported by the US NSF under Grants No. 1901099, 1737633, 1541876, 1029711, IIS-1320580, 0940818 and IIS-1218168, the USDOD under Grants HM0210-13-1-0005, USDA under Grant No. 2017-51181-27222, ARPA-E under Grant No. DE-AR0000795, NIH under Grant No. UL1 TR002494, KL2 TR002492 and TL1 TR0024-93, and the OVPR U-Spatial and Minnesota Supercomputing Institute at the University of Minnesota. Aerial imagery used in this work and shown in this paper was supplied by Hennepin and Ramsey County. Field surveys were provided by the City of St. Paul Forestry Unit and the University of Minnesota. We thank Kim Koffolt for improving the readability of this paper. We also thank the Spatial Computing Group for their feedback throughout this work. Publisher Copyright: © Springer Nature Switzerland AG 2019.

PY - 2019

Y1 - 2019

N2 - The goals of this research were to create a labeled dataset of tree shadows and to test the feasibility of shadow-based tree type identification using aerial imagery. Urban tree big data that provides information about individual trees can help city planners optimize positive benefits of urban trees (e.g., increasing wellbeing of city residents) while managing potential negative impacts (e.g., risk to power lines). The continual rise of tree type specific threats, such as emerald ash borer, due to climate change has made this problem more pressing in recent years. However, urban tree big data are time consuming to create. This paper evaluates the potential of a new tree type identification method that utilizes shadows in aerial imagery to survey larger regions of land in a shorter amount of time. This work is challenging because there are structural variations across a given tree type and few verified tree type identification datasets exist. Related work has not explored how tree structure characteristics translate into a profile view of a tree’s shadow or quantified the feasibility of shadow-only based tree type identification. We created a consistent and accurate dataset of 4,613 tree shadows using ground truthing procedures and novel methods for ensuring consistent collection of spatial shadow data that take binary and spatial agreement between raters into account. Our results show that identifying trees from shadows in aerial imagery is feasible and merits further exploration in the future.

AB - The goals of this research were to create a labeled dataset of tree shadows and to test the feasibility of shadow-based tree type identification using aerial imagery. Urban tree big data that provides information about individual trees can help city planners optimize positive benefits of urban trees (e.g., increasing wellbeing of city residents) while managing potential negative impacts (e.g., risk to power lines). The continual rise of tree type specific threats, such as emerald ash borer, due to climate change has made this problem more pressing in recent years. However, urban tree big data are time consuming to create. This paper evaluates the potential of a new tree type identification method that utilizes shadows in aerial imagery to survey larger regions of land in a shorter amount of time. This work is challenging because there are structural variations across a given tree type and few verified tree type identification datasets exist. Related work has not explored how tree structure characteristics translate into a profile view of a tree’s shadow or quantified the feasibility of shadow-only based tree type identification. We created a consistent and accurate dataset of 4,613 tree shadows using ground truthing procedures and novel methods for ensuring consistent collection of spatial shadow data that take binary and spatial agreement between raters into account. Our results show that identifying trees from shadows in aerial imagery is feasible and merits further exploration in the future.

KW - Aerial imagery

KW - Ground truthing

KW - Labeled data creation

KW - Smart cities

KW - Tree disease

KW - Urban tree big data

KW - Urban trees

UR - http://www.scopus.com/inward/record.url?scp=85077508609&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077508609&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-37188-3_11

DO - 10.1007/978-3-030-37188-3_11

M3 - Conference contribution

AN - SCOPUS:85077508609

SN - 9783030371876

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 176

EP - 192

BT - Big Data Analytics - 7th International Conference, BDA 2019, Proceedings

A2 - Madria, Sanjay

A2 - Fournier-Viger, Philippe

A2 - Chaudhary, Sanjay

A2 - Reddy, P. Krishna

PB - Springer

Y2 - 17 December 2019 through 20 December 2019

ER -

Analyzing domain knowledge for big data analysis: A case study with urban tree type classification

Abstract

Publication series

Conference

Bibliographical note

Keywords

UN SDGs

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this