A LANDSAT image has been classified using OB classification by setting rules. However due to uncertainty in the reference data I would like to include fuzziness in the accuracy assessment.
Dear Chitrini,
What kind of classification do you have and what kind of validation data? Which method would you want to use for the accuarcy assessment?
best regards,
Wietske Bijker
Hi,
Check out http://www.riks.nl/mck/
Included in the software (as described on the website) are
1.Fuzzy Set Map Comparison. The Fuzzy Set Map Comparison takes proximity relations and similarity between legend categories into account in order to obtain a nuances view on map similarity. The comparison method has been developed by Hagen (2003).
2.Hierarchical Fuzzy Pattern Matching. This method compares two maps on a patch-by-patch basis. In this manner it circumvents problems associated to cell-by-cell map comparison. Rather than cells, same-category clusters that are found in both maps are compared on their characteristics. The calculation of the similarity is based upon a Fuzzy Inference System evaluation of these characteristics. This method was developed by Power et al. (2001).
Regards
Riyad
Dear all, thank you for your answers. I found the method to perform the fuzzy accuracy assessment in this paper: Gopal, S., and Woodcock, C. E., 1994, Accuracy assessment of Thematic Maps using fuzzy sets I: theory and methods. Photogrammetric Engineering and Remote Sensing, 60, 181±188.
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics
- Geographic Information System