UVLens: Urban Village Boundary Identification and Population Estimation Leveraging Open ...
summary
A two-phase system identifies urban village polygons from satellite imagery by fusing road networks from taxi trajectories to produce road-guided patches and bike-sharing drop-off patterns to distinguish true urban villages from nearby residential blocks. Population is estimated by counting housing capacity from imagery and refining with human-activity features in a multi-view neural network, yielding IoU 0.827 for boundaries and R^2 0.92/0.94 for resident and floating populations and deployed on Xiamen’s open data platform.
AI-written summary · about
chapters (11)
- 0:00Intro
- 1:38Urban Villages: Renovation
- 2:01Traditional Methods
- 2:27New Idea: Heterogeneous Urban Data
- 3:03Framework Overview
- 3:15Urban Village Boundary Identification
- 4:37Evaluation: Results
- 5:06Urban Village Population Estimation
- 6:24Evaluation in Xiamen City
- 6:43Evaluation: Case Study #2
- 7:30Urban Village Information Service System
related talks
19mWhere's your hood at? Crowdsourcing Neighbourhoods using Open Source Tools
15mAustralia to Vanuatu: A visual exploration of the mapping challenges in Asia...
22mIdentifying Urban Makeshift Communities using satellite imagery and geo-coded data
18mDetection of solar panels based on aerial images of the city of...
13mDoes our choice of open data sets matter...
11mUrban Computing
43mPython at City Scale
11mScaling the Urban Forest: An Integrated Framework for Managing Cities by Fusing Raster(…)