A Method of Spatial Unmixing Based on Possibilistic Similarity in Soft Pattern Classification
Open Access
- 1 September 2016
- journal article
- research article
- Published by Tsinghua University Press in Fuzzy Information and Engineering
- Vol. 8 (3) , 295-314
Abstract
International audienceThis paper proposes an approach for pixel unmixing based on possibilistic similarity. The approach exploits possibilistic concepts to provide flexibility in the integration of both contextual information and a priori knowledge. Possibility distributions are first obtained using a priori knowledge given in the form of learning areas delimitated by an expert. These areas serve for the estimation of the probability density functions of different thematic classes also called endmembers. The resulting probability density functions are then transformed into possibility distributions using Dubois-Prade's probability-possibility transformation. The pixel unmixing is then performed based on the possibilistic similarity between a local possibility distribution estimated around the considered pixel and the obtained possibility distributions representing the predefined endmembers in the analyzed image. Several possibilistic similarity measures have been tested to improve the discrimination between endmembers. Results show that the proposed approach represents an efficient estimator of the proportion of each endmember present in the pixel (abundances) and achieves higher classification accuracy. Performance analysis has been conducted using synthetic and real imagesKeywords
This publication has 20 references indexed in Scilit:
- Spectral unmixingInternational Journal of Remote Sensing, 2012
- Endmember variability in Spectral Mixture Analysis: A reviewRemote Sensing of Environment, 2011
- Semi-supervised classification method for dynamic applicationsFuzzy Sets and Systems, 2010
- Seeded region growing: an extensive and comparative studyPattern Recognition Letters, 2005
- Probability-Possibility Transformations, Triangular Fuzzy Sets, and Probabilistic InequalitiesReliable Computing, 2004
- Spectral unmixingIEEE Signal Processing Magazine, 2002
- When upper probabilities are possibility measuresFuzzy Sets and Systems, 1992
- Fuzzy classification of remote sensing imagesIEEE Transactions on Geoscience and Remote Sensing, 1990
- ON THE NOTION OF DISTANCE REPRESENTING INFORMATION CLOSENESS: Possibility and Probability DistributionsInternational Journal of General Systems, 1983
- Fuzzy sets as a basis for a theory of possibilityFuzzy Sets and Systems, 1978