Element 84 has developed near real-time edge processing of drone and aerial imagery for human identification that leverages machine learning and AWS Snowcone edge capabilities during austere operations for search and rescue applications.
We outline recent projects tackling complex challenges through the lens of Machine Learning and discuss how our past experience will shape future work.
Responsible for the data labeling for a machine learning project? Here are some insights we’ve developed while managing data labeling for machine learning.
From predicting outbreaks of infectious disease to predicting the likelihood of an asthma attack, Geospatial AI and medicine are transforming healthcare.
Today, the availability of satellite imagery still far outpaces our capacity to analyze it, but machine learning and tools like Raster Vision are helping.
In a previous post we showed how the E84 R&D team used RoboSat by Mapbox to prepare training data, train a machine learning model, and run predictions on new satellite imagery. In this example, we’re going to use the same imagery source and label data as a proxy for data produced by our AWS disaster […]
Recently, the E84 R&D team has been experimenting with machine learning pipelines and identifying potential use cases. There are a lot of new and exciting tools out there and we’re interested in exploring what’s available, particularly tools related to satellite and aerial imagery (one of our specialties). Mapbox‘s RoboSat was released earlier this year and […]
Evaluating machine learning models in R with a focus on how to handle biased and imperfect data, specifically volunteer collected marine debris data.