Tags:
, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Remote Sensing QGIS: Semi-Automatic-Classification Plugin (SCP)

Semi-Automatic Classification Plugin

 

Today I’m going to take a quick look at one of the remote sensing plugins for QGIS.  Let’s have a look at what I think is one of the more useful plugins for digital image processing and is referred to as the Semi-Automatic-Classification Plugin (SCP).

SCP aims to make image-classification accessible to anyone, not just Remote Sensing professionals. It accomplishes this goal by streamlining the most popular image processing techniques and presenting them in an user-friendly format. This is a good time to start using the SCP, as Version 5.3.6 (released Jan. 27, 2017) has improved significantly since the release of Version 1.2 in 2013.

The SCP allows direct download of Remote Sensing open data. There are tabs in the main dialogue window for downloading satellite images from Landsat, Sentinel-2, Aster, and Modis. Additionally, it’s possible to download spectral reference materials from the USGS Spectral Library.

The SCP includes many features for classification and editing of rasters. There are many resources available to help you become familiarized with the features of the SCP,  for example the online community and Luca Congedo’s blog From GIS to Remote Sensing. Please watch Lucas Congedo introduce the features of the SCP Version 5:

 

What is Supervised Classification?

 

Supervised Classification is an alternative to the Unsupervised Classification method of using computer software to group pixels according to common characteristics.  In Supervised Classification, a human-guided technique is used to group pixels according to common characteristics.  This is done by the user selecting specific pixels in an image and assigning them to a category of land cover. Each of these land cover categories are referred to as an ‘Input Class’ or ‘Training Site’.  The user specifies the algorithm for how similar adjacent pixels must be in order to be added to one of the Input Classes.

Here are a few examples of where classification analysis proves useful. With the QGIS SCP and the right data we can differentiate different types of land cover, conduct water body delineation, classify cropland, and conduct forest monitoring analysis to name just a few useful things.

Getting Started with the SCP Plugin:

 

To try the SCP Plugin for yourself, please check out Luca Congedo’s blog From GIS to Remote Sensing.  Also, read through the very useful SCP user’s manual.

 

2 comments on "Remote Sensing QGIS: Semi-Automatic-Classification Plugin (SCP)"

  1. you are doing great work

  2. Thokozane says:

    Hi, I am a PhD student in South Africa, I would like to know if I can use SCP as method to my papers that I will publish at the end of my project.

Comments are closed.

Related Articles

Google Earth Engine – What is a Data Cube and Why Canadians Should Care

What is a data cube and how is it used for big data analytics?  How does a data cube work…

GIS in Action: How Does BC Pinot Noir Express Terroir?

Editors note: In this previously published article, Karl Kliparchuk, BCIT GIS Program Head & Instructor uses GIS to explore the…

Eastern Canada

Grand Banks to the Gulf of Mexico: My Canadian Geomatics Career

I often contemplate what it would have been like working at the same job for an entire career. Mind you,…