In order to benchmark efficiency, we take a deep dive into Zarr and Parquet data retrieval to compare performance on various time scales.
In order to benchmark efficiency, we take a deep dive into Zarr and Parquet data retrieval to compare performance on various time scales.
This blog explores the direct classification approach to change detection using our open-source geospatial deep learning framework, Raster Vision, and the publicly available Onera Satellite Change Detection (OSCD) dataset.
As one of seven pilot programs to address environmental issues in Africa, Azavea trained student workers to label satellite imagery using GroundWork and created a machine learning model to identify tree canopy.
Azavea is releasing a dataset consisting of 32 unique Sentinel-2 tiles with cloud labels produced by humans.
SAR imagery is having a moment. In this blog we explore what exactly it is, why it is so special, and tips for labeling it.
This release presents a major jump in Raster Vision’s power and flexibility. The newly added features allow for finer control of the model training as well as greater flexibility in ingesting data.
Need to create a training dataset that contains multiple images? GroundWork launches “Campaigns” to help you handle large datasets for machine learning.
We compare the results of two machine learning models that detect clouds in Sentinel-2 satellite imagery and share pointers about models you can try for yourself.
Hear from labeling competition participants on why this diverse group of people came together, from across the world, to collaborate on a hallmark training dataset compliant with the STAC standard.
Responsible for the data labeling for a machine learning project? Here are some insights we’ve developed while managing data labeling for machine learning.