Press 'o' to toggle the slide overview and 'f' for full-screen mode.

Choose the theme in which to view this presentation:

Black - White - League - Sky - Beige - Simple
Serif - Blood - Night - Moon - Solarized



Copyright © John Lindsay, 2015

• In the georelational data model used for vectors, each feature is assigned a unique key or identifer.
• With raster data, points, lines, and polygons are stored as one or more pixels but there is nothing that explicitly identifies a cell as belonging to a certain feature
• What can you do when you need to perform operations on features?

Clumping (Grouping)

• Each contiguous group of cells with the same value in the input image will be assigned the same unique identifier in the output image


• Called Region Group in ArcGIS, r.clump in GRASS, Group in Idrisi, and Clump (Group) in Whitebox GAT.


• Only performed on Boolean or categorical raster images
  • Why not on continuous data images?

Clumping (Grouping)

Spatial Filtering

• A very widely used analysis technique in raster geospatial analysis; perhaps the most common raster neighbourhood operation
• Based on convolution techniques of image processing
• Involves visiting each grid cell in an image and examining the neighbouring cells within a kernel, also called a window or filter
• Works on Boolean, categorical, and continuous images

Spatial Filtering

• Filters are used for all kinds of applications


• Some filters are used to smooth surfaces


• Some are used to emphasize the high-frequency noise


• Others are used to find edges in an image

Varrying kernel size

Varrying kernel size

• The most common kernel size is 3 cell × 3 cell


• Any other kernel size is possible, though it should be an odd number so that there is a centre cell to the kernel


• The number of calculations that are needed to perform a filter increase exponentially with increased kernel size


• Repeating a filter several times is equivalent to using a larger filter window

Varrying kernel shapes

Varrying kernel shapes

• Other kernel shapes are possible
  • e.g. rectangles, ovals, etc.
  • The shape must be approximated by a grid


• These are usually used to give a directional preference to the filter known as anisotropy


• Isotropy = homogeneity in all directions


• Anisotropy is the opposite, i.e. pronounced directionality