Increasing data accessibility by adding existing datasets and capabilities to a cutting-edge visualization app to enable cross-community use
We will collate and publish existing datasets from collaborators and ingest them into a visualization app to help researchers with machine learning model-building and hypothesis-making. These data collation and app development methods could help other researchers increase their data accessibility.
We will collect, collate, harmonize, and standardize existing hyperspectral datasets of major crops for selected agroecological zones throughout the world from known collaborators, ingesting them into a visualization app to help researchers gather crucial reference training, testing, and validation data for machine learning and artificial intelligence models for studying agricultural crops. Data collation and app development will provide easy search and visualization capabilities, increasing data accessibility and understanding. Anticipated deliverables include at least 6 data releases from collaborators including Gumma (India data), Huete (Australia, Turkey data), Marshall (California data), and Lamb/Hively (Maryland data); development and release of the app and source code; and presentations at two conferences to solicit feedback during development. This work reflects the CDI FY23 theme of “Increasing connection, readiness, and equity of climate related data and models” by connecting siloed data into our Global Hyperspectral Imaging Spectral-library of Agricultural crops (GHISA; www.usgs.gov/WGSC/GHISA) platform and our visualization app.
We will collate and publish existing datasets from collaborators and ingest them into a visualization app to help researchers with machine learning model-building and hypothesis-making. These data collation and app development methods could help other researchers increase their data accessibility.
We will collect, collate, harmonize, and standardize existing hyperspectral datasets of major crops for selected agroecological zones throughout the world from known collaborators, ingesting them into a visualization app to help researchers gather crucial reference training, testing, and validation data for machine learning and artificial intelligence models for studying agricultural crops. Data collation and app development will provide easy search and visualization capabilities, increasing data accessibility and understanding. Anticipated deliverables include at least 6 data releases from collaborators including Gumma (India data), Huete (Australia, Turkey data), Marshall (California data), and Lamb/Hively (Maryland data); development and release of the app and source code; and presentations at two conferences to solicit feedback during development. This work reflects the CDI FY23 theme of “Increasing connection, readiness, and equity of climate related data and models” by connecting siloed data into our Global Hyperspectral Imaging Spectral-library of Agricultural crops (GHISA; www.usgs.gov/WGSC/GHISA) platform and our visualization app.