Classification refers to categorizing or classifying parts of an image into different types, fo...
Public Lab is an open community which collaboratively develops accessible, open source, Do-It-Yourself technologies for investigating local environmental health and justice issues.
<a href="https://www.flickr.com/photos/worldworldworld/4918161225/" title="how much oil hit the coast ? by cesarharada.com, on Flickr"><img src="https://farm5.staticflickr.com/4074/4918161225_ac11e4ffa4.jpg" width="500" height="156" alt="how much oil hit the coast ?"></a>
Classification refers to categorizing or classifying parts of an image into different types, for example **pavement**, **water**, **soil** and **sand**. It often relies on examining an image pixel by pixel and using color or shape information to decide which "class" is the best match. The result is often an image such as this one, which is colored by class:
<img width="500px" src="https://publiclab.org/sites/default/files/USCB_GeniePro_Classification.jpg" />
If classification can be automated, using for example the software program [GeniePro](http://geniepro.lanl.gov/), large numbers of photos or large maps can be colored by class, helping to quantify for example "how much wetlands?", "how much oil?" or "how much _Spartina alterniflora?_"
Using the ratios of Red, Green, and Blue (and possibly Near-infrared), _spectral classification_ attempts to categorize regions of an image by land type.
Read more in [this post by Adam Griffith and students at University of South Carolina - Beaufort](http://publiclaboratory.org/notes/adam-griffith/9-10-2011/students-map-campus-university-south-carolina-beaufort-south)
##Open source classification software##
* [Spring](http://www.dpi.inpe.br/spring/) - desktop app, not super user-friendly (see [a tree canopy project using SPRING](http://depts.washington.edu/rsgalwrk/canopy/) by Diane Styers)
* [Clashifier](http://github.com/jywarren/clashifier) - web service, in planning stage, not functional