Skip to content

Image processing and dataset generation for training climate models

Notifications You must be signed in to change notification settings

adamconrad7/GeoData

Repository files navigation

Hyperspectral Satellite Image Translation

Overview

The goal of this project is to use deep image processing techniques to model weather conditions from hyperspectral satellite imagery.

Data

Input data consists of sixteen-channel hypersepectral readings from the Geostationary Operational Environmental Satellite (GOES) R-series Advanced Baseline Imager (ABI), sampled in five-minute increments. Ground truth data consists of images composed of readings from NOAA's U.S. Climate Reasearch Network (USCRN), a collection of 144 climate observing stations across the U.S.

Data Collection and Preprocessing

To Do: Organize input and groud truth rasters together for a given time.

GOES ABI

Many tools are available to retreieve the GOES ABI data, which is available on public cloud storage. For now I am using Goes2Go as it is very simple to use, but I may need to write my own for my desired functionality. Images are stored in NetCDF format which contains many metadata such as the satellite's position, angle of scan for each pixel, and more.

The first step is to downsample the image. alt text

Next, the pixel's coordinates are calculated from their scan angles using the algorithim found here.

The image is masked to include only the contigous U.S. alt text

To Do:

Crop image

Stack each band depth-wise

CRN

Data is downloaded from the sub-hourly USCRN Quality-controlled datasets directory. Each file is a list of readings for a single station over time.

First, all station files are concatenated.

Readings are then grouped by the date and time when they occured. (One file per five-minute interval with every station's reading at that time)

Using a copy of a georefernced GOES ABI image, all station readings for a given time are rasterized. alt text

A reading is interpolated over the entire raster using a radial basis function with a linear kernel. alt text

Model Selection

To Do: Select model

About

Image processing and dataset generation for training climate models

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published