DOI: 10.1016/j.isprsjprs.2020.02.001
论文题名: Geocoding of trees from street addresses and street-level images
作者: Laumer D. ; Lang N. ; van Doorn N. ; Mac Aodha O. ; Perona P. ; Wegner J.D.
刊名: ISPRS Journal of Photogrammetry and Remote Sensing
ISSN: 9242716
出版年: 2020
卷: 162 语种: 英语
英文关键词: City scale
; Deep learning
; Faster R-CNN
; Geocoding
; Global optimization
; Google Street View
; Image interpretation
; Object detection
; Street trees
; Tree inventories
; Urban areas
Scopus关键词: Combinatorial optimization
; Deep learning
; Ecosystems
; Global optimization
; Object detection
; City scale
; Faster R-CNN
; Geo coding
; Google Street View
; Image interpretation
; Street trees
; Tree inventories
; Urban areas
; Forestry
; artificial neural network
; detection method
; ecosystem service
; experimental study
; global change
; GPS
; image analysis
; inventory
; long-term change
; machine learning
; mortality
; optimization
; scale effect
; urban area
; California
; United States
英文摘要: We introduce an approach for updating older tree inventories with geographic coordinates using street-level panorama images and a global optimization framework for tree instance matching. Geolocations of trees in inventories until the early 2000s where recorded using street addresses whereas newer inventories use GPS. Our method retrofits older inventories with geographic coordinates to allow connecting them with newer inventories to facilitate long-term studies on tree mortality etc. What makes this problem challenging is the different number of trees per street address, the heterogeneous appearance of different tree instances in the images, ambiguous tree positions if viewed from multiple images and occlusions. To solve this assignment problem, we (i) detect trees in Google street-view panoramas using deep learning, (ii) combine multi-view detections per tree into a single representation, (iii) and match detected trees with given trees per street address with a global optimization approach. Experiments for >50000 trees in 5 cities in California, USA, show that we are able to assign geographic coordinates to 38% of the street trees, which is a good starting point for long-term studies on the ecosystem services value of street trees at large scale. © 2020 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS) Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/158464
Appears in Collections: 气候变化与战略
There are no files associated with this item.
作者单位: EcoVision Lab, Photogrammetry and Remote Sensing, ETH Zürich, Switzerland; Forest Service, US Department of Agriculture, United States; University of Edinburgh, United Kingdom; California Institute of Technology, United States
Recommended Citation:
Laumer D.,Lang N.,van Doorn N.,et al. Geocoding of trees from street addresses and street-level images[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2020-01-01,162