Raster Vision is the interface between the fields of earth observation and deep learning, making it easier to apply novel computer vision techniques to geospatial imagery of all types. Joe lays out how it can be implemented in your organization and give you a competitive advantage.
Raster Vision is the interface between the fields of earth observation and deep learning, making it easier to apply novel computer vision techniques to geospatial imagery of all types. Joe lays out how it can be implemented in your organization and give you a competitive advantage.
We calculated which congressional districts have become over or underpopulated since they were last drawn.
Without taking on copious outside investment, we have to support our products with only a share of our available time and resources. Here are five strategies we use to build successful bootstrapped products.
Mining the knowledge and expertise of your data labeling team will improve the data quality of your machine learning project and increase your team’s productivity.
Remote sensing instruments like NASA’s GEDI, which is mounted to the Japanese Experiment Module – Exposed Facility (JEM-EF) on the International Space Station (ISS), produce a massive amount of data and one of the tools scientists use when working with those data is a geo-locator–a geographic coordinates search engine. GEDI, launched in 2018, is the…
We’re passionate about Earth Science Data and helping to solve the challenges in that field. At this year’s ESIP Summer Meeting, we have four E84 team members speaking and moderating across multiple sessions. Here’s where you can find us if you want to chat! STAC and Sentinel-2 Cloud-Optimized GeoTIFFs SpatioTemporal Asset Catalogs (STAC) is an…
We refactored the Raster Vision codebase from the ground up to make it simpler, more consistent, and more flexible. Check out Raster Vision 0.12.
CurbLR is a promising new open data specification for curb regulation. We used it to visualize and analyze Philadelphia’s curb management approaches affect on traffic.
Incorporate high-resolution satellite imagery into your labeling projects for free.
We pulled data from disparate hospital data sources to create a comprehensive national dataset of the hospital system for the COVID-19 response, using geocoding, proximity matching, and fuzzy string matching.