Difference between revisions of "ALGORITHM2"
| Line 3: | Line 3: | ||
---- | ---- | ||
| − | [https://www. | + | [https://www.dlr.de/ DLR]'s flood detection algorithm (provisionally, '''Algorithm2''') is a fuzzy-logic based water class membership assignment, is a part of a flood classification refinement step described in Martinis et al. (2015)<sup>[1]</sup> and Twele et al. (2016)<sup>[2]</sup>. It aims to exclude water-lookalikes and to reduce underestimations from initial classification by constructing a fuzzy set that consists of (a) the backscatter level, (b) the elevation of an image element in comparison to the mean elevation of the initially derived water areas, (c) topographic slope information, and (d) the size of an individual flood object; degree of an element’s membership to the class water is determined by standard S and Z membership functions. The average of the individual membership degrees is computed for each pixel in order to combine all fuzzy elements into a single composite fuzzy set Figure 20. |
| − | + | The membership degree strongly depends on the position of the crossover point xc (i.e. the half of the width of the fuzzy curve), which is defined by the fuzzy thresholds x1 and x2.The fuzzy threshold values for each element are either determined according to statistical computations or are set empirically. The membership degree of the subsequently, the flood mask is derived through a threshold defuzzification step, which transforms each image element with a membership degree > 0.6 into a discrete thematic class. Furthermore, the region-growing tolerance criterion is defined by a relaxed fuzzy threshold of > 0.45 based on the composite fuzzy set. | |
| + | Uncertainty information is contained in the fuzzy mask generated over classified water pixels, which takes on values ranging from 0 and 1. | ||
'''For further details the Reader is referred to the dedicated section of the Product Description Document: https://www.gfm_pdd.org/Algorithm2''' | '''For further details the Reader is referred to the dedicated section of the Product Description Document: https://www.gfm_pdd.org/Algorithm2''' | ||
Revision as of 09:14, 10 March 2021
DLR's flood detection algorithm (provisionally, Algorithm2) is a fuzzy-logic based water class membership assignment, is a part of a flood classification refinement step described in Martinis et al. (2015)[1] and Twele et al. (2016)[2]. It aims to exclude water-lookalikes and to reduce underestimations from initial classification by constructing a fuzzy set that consists of (a) the backscatter level, (b) the elevation of an image element in comparison to the mean elevation of the initially derived water areas, (c) topographic slope information, and (d) the size of an individual flood object; degree of an element’s membership to the class water is determined by standard S and Z membership functions. The average of the individual membership degrees is computed for each pixel in order to combine all fuzzy elements into a single composite fuzzy set Figure 20. The membership degree strongly depends on the position of the crossover point xc (i.e. the half of the width of the fuzzy curve), which is defined by the fuzzy thresholds x1 and x2.The fuzzy threshold values for each element are either determined according to statistical computations or are set empirically. The membership degree of the subsequently, the flood mask is derived through a threshold defuzzification step, which transforms each image element with a membership degree > 0.6 into a discrete thematic class. Furthermore, the region-growing tolerance criterion is defined by a relaxed fuzzy threshold of > 0.45 based on the composite fuzzy set. Uncertainty information is contained in the fuzzy mask generated over classified water pixels, which takes on values ranging from 0 and 1.
For further details the Reader is referred to the dedicated section of the Product Description Document: https://www.gfm_pdd.org/Algorithm2
References
[1]Martinis, S., Twele, A., Kersten, J., (2015). A fully automated TerraSAR-X based flood service. ISPRS Journal of Photogrammetry and Remote Sensing, 104, 203-212.
download Martinis et al. (2015)
[2] Twele, A., Cao, W., Plank, S., Martinis, S., 2016. Sentinel-1 based flood mapping: a fully automated processing chain. International Journal of Remote Sensing, 37 (13), 2990-3004.
download Twele et al. (2016)
