I need to analyze differences (historical) in coverage of some species (seagrass Posidonia). Although its a marine species, subject area is very shallow and even from common satellite and aerial imagery, species can be observed significantly (screenshot attached).enter image description here

I need to create 2 polygons (old - new imagery) that encompass coverage of that species and by vector overlay method need to to create third polygon which will show only gain/loss.

Is there possibility to automatically (by a GIS software) create polygon based on color differences?

1 Answer 1


There are several ways you can do this, but the most appropriate for you, in my opinion, is to do classification on your data, Image classification techniques group pixels to represent different features based on different DN Values. You should be able to classify your data if the features show a consistent difference. There are three main image classification techniques.

This is a tutorial that shows how to use the Semi Automatic Classification Plugin for QGIS, the tutorial shows how to classify landcover, but it should give you an idea of how you can do this.

If your study area is small, then you might be better off just digitizing the feature polygons on your own.

P.S: Take a look at this as well,it might give you some ideas. What you are looking to do is basically change detection, take a look at this, as well.

  • 1
    Thank you for answering. I found some (biological) paper method that I need to do, it would be Unsupervised image classification - to transform a digital image into a thematic map by defining groups of pixels. Mar 24, 2016 at 10:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.