Skip to content

This Jupyter notebook studies classification of land types from EOS SAT Sentinel 2A imagery with random forest, SVM, Naive Bayes, Decision Tree (CART) and other methods. QGIS software is used to generate training data for the classifier.

Notifications You must be signed in to change notification settings

GreenspaceGroup/Topo-Feature-Extract

 
 

Repository files navigation

Topo-Feature-Extract

Topological feature extraction is specific use case for machine learning algorithms for classification purposes of topological features from satellite or drone imagery ... thus, this repository is closely related to the GreenspaceGroup organization's more general curated list of AI, Machine Learning and Deep Learning Algorithms that might be remotely applicable in a modified form to greenspaces, landscape architecture and shared-space property management.

Heritage of this repository

This repository started by building upon the methodology of an Jupyter project notebook which studied classification of land types from the EOS SAT satellite imagery available from Sentinel 2A launched on June 23, 2015 as part of the European Commission’s Copernicus program. In that project, a variety of machine learning algorithms for classification purposes were used which included random forest, Support Vector Machine (SVM), Naive Bayes, Decision Tree CART as well as looking at other applicable machine learning algorithms. QGIS software was used to generate training data for the classifier.

About

This Jupyter notebook studies classification of land types from EOS SAT Sentinel 2A imagery with random forest, SVM, Naive Bayes, Decision Tree (CART) and other methods. QGIS software is used to generate training data for the classifier.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%