Raster processing using Python Tools: Reference

Key Points

Introduction: Working with Raster Data
  • Gridded data that vary in space and time are common in many geospatial applcations

  • Specialized tools are needed to accommodate the complexity and size of many raster datasets

  • GIS tools (e.g. QGIS, SAGA GIS) are usually needed to visualize these datasets

Using GDAL
  • GDAL is extremely useful for reading/writing/transforming raster datasets

  • GDAL usually ships with useful command-line utilities

  • Errors can be handled pythonically

  • Pixel values can be extracted to a numpy array

  • Raster attributes can usually be read without reading pixel values

Efficient raster computation with PyGeoProcessing
  • PyGeoProcessing provides programmable operations for efficient raster computations

  • PyGeoProcessing is most useful for large but not big data

  • PyGeoProcessing can help you to align inputs so they overlap perfectly

  • If your rasters may be too large to fit into main memory, PyGeoProcessing may be faster than matrix operations

FIXME: more reference material.