Segmenting remote sensing imagery with box prompts

Qiusheng WuQiusheng Wu
1 min read

Segment-geospatial v0.9.0 is out📢It now supports segmenting remote sensing imagery with box prompts. You can create box prompts interactively or use existing vector data as box prompts 🌐

GitHub: https://github.com/opengeos/segment-geospatial
Notebook: https://samgeo.gishub.org/examples/box_prompts
Video tutorial: https://youtu.be/ayFgAj23BfI

10
Subscribe to my newsletter

Read articles from Qiusheng Wu directly inside your inbox. Subscribe to the newsletter, and don't miss out.

Written by

Qiusheng Wu
Qiusheng Wu

Dr. Qiusheng Wu is an Associate Professor in the Department of Geography & Sustainability at the University of Tennessee, Knoxville. He is also an Amazon Visiting Academic and a Google Developer Expert (GDE) for Earth Engine. His research interests include geospatial data science, remote sensing, and environmental modeling. Dr. Wu is particularly interested in utilizing big geospatial data and cloud computing (e.g., Google Earth Engine, Microsoft Planetary Computer, Amazon Web Services) to study environmental change, especially surface water and wetland inundation dynamics. His research has been funded by NASA, USDA, and the Department of Defense. Dr. Wu is an advocate of open science and reproducible research, having developed and published various open-source packages for advanced geospatial analysis and data visualization, such as geemap, leafmap, lidar, and segment-geospatial. Check out his open-source projects at https://open.gishub.org