In Detail: Land Cover Change Detection through Supervised Classification

Abstract: 

This Recommended Practice aims to (1) conduct a supervised land cover classification in QGIS using the SCP plugin and (2) to conduct change detection analysis. These skills are applicable to a vast range of disasters throughout the entire disaster management cycle. In this example, the remote sensing technique is applied to monitor deforestation in a part of the Amazon rainforest south of Santarém, Pará, in Brazil. However, it can be applied to any other study area. The required inputs are two or more satellite images of the same area at a different point in time. This will result in an output of a table showing the exact change in land cover in number of pixels as well as a visualization of the land cover change in the form of a shapefile.

Requirements: 

The data requirements for this analysis are at least two cloudless satellite images of the same area at a different point of time. This data will be used in order to uncover the land cover change between the two images.

Important: there should be no cloud coverage on the image, as this will distort the classification. This would greatly reduce the quality of the results.

Strengths and Limitations: 

This Recommended Practice explains how to conduct a supervised land cover classification followed by a change detection analysis. In this application, the method is applied for an area of rainforest in the Amazon to detect forest loss.

 

Zircon - This is a contributing Drupal Theme
Design by WeebPal.