Mapping the Individual Trees in Urban Orchards by Incorporating Volunteered Geographic Information and Very High Resolution Optical Remotely Sensed Data: A Template Matching-Based Approach

被引:23
作者
Vahidi, Hossein [1 ,2 ]
Klinkenberg, Brian [2 ]
Johnson, Brian A. [3 ]
Moskal, L. Monika [4 ]
Yan, Wanglin [1 ,5 ]
机构
[1] Keio Univ, Grad Sch Media & Governance, EcoGIS Lab, Fujisawa, Kanagawa 2520882, Japan
[2] Univ British Columbia, Lab Adv Spatial Anal, Vancouver, BC V6T 1Z2, Canada
[3] Inst Global Environm Strategies, Nat Resources & Ecosyst Serv Area, Hayama, Kanagawa 2400115, Japan
[4] Univ Washington, Remote Sensing & Geospatial Anal Lab, Seattle, WA 98195 USA
[5] Keio Univ, Fac Environm & Informat Studies, Fujisawa, Kanagawa 2520882, Japan
关键词
volunteered geographic information; very high resolution imagery; collective sensing; data quality; template matching; individual tree detection; urban orchard; NORMALIZED CROSS-CORRELATION; CITIZEN SCIENCE DATA; CROWN DETECTION; DATA QUALITY; IMAGERY; FOREST; LIDAR; OPENSTREETMAP; VALIDATION; COVER;
D O I
10.3390/rs10071134
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
This paper presents a collective sensing approach that integrates imperfect Volunteered Geographic Information (VGI) obtained through Citizen Science (CS) tree mapping projects with very high resolution (VHR) optical remotely sensed data for low-cost, fine-scale, and accurate mapping of trees in urban orchards. To this end, an individual tree crown (ITC) detection technique utilizing template matching (TM) was developed for extracting urban orchard trees from VHR optical imagery. To provide the training samples for the TM algorithm, remotely sensed VGI about trees including the crowdsourced data about ITC locations and their crown diameters was adopted in this study. A data quality assessment of the proposed approach in the study area demonstrated that the detected trees had a very high degree of completeness (92.7%), a high thematic accuracy (false discovery rate (FDR) = 0.090, false negative rate (FNR) = 0.073, and F-1 score (F-1) = 0.918), and a fair positional accuracy (root mean squareerror (RMSE) = 1.02 m). Overall, the proposed approach based on the crowdsourced training samples generally demonstrated a promising ITC detection performance in our pilot project.
引用
收藏
页数:42
相关论文
共 191 条
[51]  
Foster A, 2017, URBAN SCI, V1, DOI 10.3390/urbansci1030030
[52]  
Fouda Y., 2015, TRENDS APPL SCI RES, V10, P195, DOI [DOI 10.3923/tasr.2015.195.206, 10.3923/tasr.2015.195.206]
[53]  
Fouda Y. M., 2014, International Journal of Computer Science & Information Technology, V6, P47, DOI 10.5121/ijcsit.2014.6404
[54]  
Fritz S., 2017, MAPPING CITIZEN SENS
[55]  
Fritz S, 2011, LECT NOTES COMPUT SC, V6783, P39
[56]   3D Tree Dimensionality Assessment Using Photogrammetry and Small Unmanned Aerial Vehicles [J].
Gatziolis, Demetrios ;
Lienard, Jean F. ;
Vogs, Andre ;
Strigul, Nikolay S. .
PLOS ONE, 2015, 10 (09)
[57]   Joint use of remote sensing data and volunteered geographic information for exposure estimation: evidence from Valparaiso, Chile [J].
Geiss, Christian ;
Schauss, Anne ;
Riedlinger, Torsten ;
Dech, Stefan ;
Zelaya, Cecilia ;
Guzman, Nicolas ;
Hube, Mathias A. ;
Arsanjani, Jamal Jokar ;
Taubenboeck, Hannes .
NATURAL HAZARDS, 2017, 86 :81-105
[58]   Integrating Crowdsourced Data with a Land Cover Product: A Bayesian Data Fusion Approach [J].
Gengler, Sarah ;
Bogaert, Patrick .
REMOTE SENSING, 2016, 8 (07)
[59]  
Gibson J. J., 1979, ECOLOGICAL APPROACH, DOI DOI 10.4324/9781315740218
[60]  
Gomes M.F., 2013, P 2013 SPIE REM SENS