Talks and Poster Presentations (with Proceedings-Entry):
A. Zlinszky:
"Will it blend? Visualization and accuracy evaluation of high-resolution fuzzy vegetation maps";
Poster: XXIII ISPRS Congress,
Prague, Czechia;
2016-07-12
- 2016-07-19; in: "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences",
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences,
XLI-B2
(2016),
ISSN: 1682-1750;
335
- 342.
English abstract:
Ensemble classifiers such as random forests can create fuzzy maps where the probability for every class is calculated for each pixel. These maps are rich sources of information without the simplification of hard classification. Spatially explicit certainty metrics have been proposed based on class probabilities, but the link to confusion matrix based indices was not yet explored. Here we present a suite of fuzzy map visualization and accuracy evaluation products. These represent the classification and its accuracy at different levels of reduction, are directly comparable with hard classifications and ready for operational use as a new standard.
"Official" electronic version of the publication (accessed through its Digital Object Identifier - DOI)
http://dx.doi.org/10.5194/isprs-archives-XLI-B2-335-2016
Electronic version of the publication:
http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLI-B2/335/2016/isprs-archives-XLI-B2-335-2016.pdf
Created from the Publication Database of the Vienna University of Technology.