0

I want to detect avalanche debris of a known avalanche using Sentinel-2 optical imagery. I've calculated the Normalized Difference Snow Index and from there I've extracted all the values above 0.4 which gives me a raster file that contains all the snow for the scene. There, I can easily recognise the avalanche debris that lie below the snow line based on their distinctive elongated tongue shape.

Basically what I want to end up with is a file that contains nothing but the avalanche debris without manually selecting them. Is it possible for ArcGIS to identify and extract these elongated tongue shapes?

I'm rather new to ArcGIS so I have no idea if this is super trivial or practically impossible.

2
  • Welcome to GIS SE! As a new user please take the tour to learn about our focused Q&A format. I have edited your question slightly, removing the "or any other application for that matter" as asking for solutions for multiple GIS software packages would make this question too broad.
    – Midavalo
    May 6 '17 at 16:35
  • Welcome to gis.stackexchange! Please note that a good question on this site is expected to show some degree of research on your part, i.e. what you have tried and - if applicable - code so far. For more info, you can check our faq.
    – underdark
    May 6 '17 at 18:50
2

ArcGIS does not currently have advanced feature extraction capabilities. The most advanced image classification tools available in ArcGIS would be the segmentation and classification tools available with the Spatial Analyst extension. However, keep in mind that this approach would not be able to distinguish between shapes to extract debris fields.

I would recommend using Python libraries such as scikit-learn, scikit-image, or OpenCV--all of which have many examples. If you prefer a graphic user interface for extracting features from imagery, the gold standard is eCognition Developer.

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.