Open Source Archives | Page 3 of 10

OAR map overlaid on top of hands sewing a garment.

Connecting Data and People with the Open Apparel Registry

By Deb Boyer on August 7th, 2019

Since its launch on March 28, the Open Apparel Registry (OAR) has grown to include over 18,300 facilities in 92 countries. We’ve already heard of a few fascinating use cases where data from the OAR contributed to decision making by brands and facilities.

Ground truth imperfectly overlaying machine learning predictions.

Software Engineering

Using Noisy Labels to Train Deep Learning Models on Satellite Imagery

By Lewis Fishgold on August 5th, 2019

How do noisy labels affect the accuracy of a deep learning model? We added different amounts of noise to the SpaceNet Vegas buildings dataset and trained some models to find out.

Azavea employee presenting GIS training in Guyana.

Company

Developing Countries, Capacity Building, and SDGs

By Fernando Ramirez on May 30th, 2019

Utilizing earth imagery to achieve all 17 UN SDGs by 2030 will take considerate effort. In order to do this effectively, we must include training and capacity building in our solutions.

Parts of map highlighted overlaid on a photograph of a bee.

Software Engineering

Using Cloud-Optimized GeoTIFFs (COGs) for More Efficient Web App Architecture

By Jenny Fung on April 23rd, 2019

Cloud-Optimized GeoTIFFs (COGs) are geoTIFFs hosted on a cloud or file server, and are optimized for remote reads. They proved useful in a recent project.

Districts of Philadelphia overlaid on top of a photo of a child swinging with the sky and playground in the background.

Spatial Analysis

Scoring Philadelphia’s City Council Districts for Children’s Health and Well-being

By Daniel McGlone on April 16th, 2019

Scoring Philadelphia City Council districts on assets and risks using a weighted spatial analysis model in R and Python.

Raster Vision applied to the city of Georgetown, Guyana.

Software Engineering

Raster Vision 0.9 Release Candidate

By Lewis Fishgold on April 15th, 2019

This release of Raster Vision includes bug fixes, an easier setup, improved performance, and the ability to train models off of labels in OSM.