I am trying to mimic a vector classification process in QGIS that can be done using the ArcGIS Business Analyst Segmentation Module in ArcMap.

My current best solution is to rasterize each of the attribute table columns one by one, put them into an image stack, and use an unsupervised Orfeo Toolbox classifier to classify the image stack. Although this method is cumbersome, time consuming, and has creates issues for analysis. Any ideas?


This can be done by using the Scikits and Pandas Python libraries, admittedly this method does not exclusively use QGIS, but still uses open source methods.

Save the vector file as a table, process and classify the table along the lines of this tutorial. Using a unique identifier to rejoin the classified data to the original geometry, will then allow the data to be viewed spatially.

| improve this answer | |

You can use also the pixel based classification framework from ORFEO Toolbox and using as input vector data and perform unsupervised classification with KMeans.

Have a look at the pixel based classification recipe in the otb cookbook:


| improve this answer | |

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.