Marius Appel wrote this piece about processing satellite image collections. This blog post introduces the gdalcubes R package, aiming at making the work with collections and time series of satellite imagery easier and more interactive.
Scientists working with collections and time series of satellite imagery quickly run into some of the following problems:
- Images from different areas of the world have different spatial reference systems (e.g., UTM zones)
- The pixel size of a single image sometimes differs among its spectral bands / variables
- Spatially adjacent image tiles often overlap
- Time series of images are often irregular when the area of interest covers spatial areas larger than the extent of a single image
- Images from different data products or different satellites are distributed in diverse data formats and structures
The article then explains how to overcome these problemss when using gdalcubes R package. The author uses demo dataset, a collection of 180 Landsat 8 surface reflectance images, covering a small part of the Brazilian Amazon forest. Nice one!
[Read More]