I'm using ERDAS for my UAV image classification (consists 6 land cover types) and I performed the first step of Object-oriented classification, Single Feature Probability (SFP). It outputted a probability layer with value ranged from 0 to 1 indicating the similarity of every pixel to the training set.

However, it did not tell me the formula of how they calculated the SFP.

Could anyone give me a clue of it?

1 Answer 1


As far as I know, SFP is the class probability derived from a Bayesian Network, i.e. a statistical supervised classifier that can incorporate both continuous and discrete information. There is a hint on that in this paper.

  • Thanks radouxju. I have also came to the same article as well.
    – Gloria
    Jan 14, 2017 at 15:02

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.