Links
- More information: Publisher Index Page (html)
- Download citation as: RIS | Dublin Core
Abstract
Large geospatial datasets must often be generalized for analysis and display at reduced scales. Automated methods including artificial intelligence and deep learning are being applied to this problem, but the results are often analyzed on the basis of limited and subjective measures. To better support automation, a project is underway to develop a robust Python toolkit for computing objective metrics of the quality of generalized vector data. Six metrics are currently under development: Hausdorff and average distance between polylines, feature density difference, polyline legibility, change in topological adjacency, and change in sinuosity. This paper discusses the development progress on tools to compute the Hausdorff and average distance between polylines, generating quality metrics to quantify displacement.
| Publication type | Conference Paper |
|---|---|
| Publication Subtype | Conference Paper |
| Title | Generalization quality metrics to support multiscale mapping: Hausdorff and average distance between polylines |
| Year Published | 2023 |
| Language | English |
| Publisher | ICA Commission on Multiscale Cartography |
| Contributing office(s) | Center for Geospatial Information Science (CEGIS) |
| Description | 3 p. |
| Conference Title | GIScience 2023 Workshop on CartoAI: AI for cartography |
| Conference Location | Leeds, United Kingdom |
| Conference Date | September 12-15, 2023 |
| Google Analytic Metrics | Metrics page |