Stochastic Distances and Uncertainties Maps Applied to Multi Sources Data Classification
Main Article Content
Abstract
In this paper is proposed classifi cation improvement through a new multi source or multi -sensor integration method
whose combination is in diff erent level those known in literature. At this rate, data integration occurs through the w
image classifi cation results from w diff erent sources. The content of these reports refers to distances and test statistics
contained in the uncertainty maps (regarding the reliability of the obtained classifi cation) for each of the classifi cations. The data selected for this paper include a optical image and a microwave image. Such images were classifi ed by
the region based classifi er PolClass, that besides the classifi cation, the classifi er generates an uncertainty map. Five
Scenarios were established based on the individual classifi ed images and their uncertain maps. Two of these Scenarios
showed low uncertainties values. One of them presented the kappa coeffi cient and overall accuracy statistically equal
to the highest values obtained individually. The other Scenario, based on Fuzzy logic, had the best result among all
the Scenarios. The use of information from diff erent sources proved to be a positive factor in land use and land cover
classifi cation. The use of fuzzy logic led to classifi cation results with low uncertainty by allowing mixed classes.
Downloads
Metrics
Article Details
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
Authors who publish in this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors can enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) before and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (see "The Effect of Open Access").
