2020
DOI: 10.5194/egusphere-egu2020-10659
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

UAV-based training for fully fuzzy classification of Sentinel-2 fluvial scenes

Abstract: <p><span>In current fluvial remote sensing approaches, there exists a certain dichotomy between the analysis of small channels at local scales which is generally done with airborne data and the analysis of entire basins at regional and national scales with satellite data. </span><span>One possible solution to this challeng</span><span>e</span><span> is to use low-altitude imagery from low-cost UAVs t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
13
0

Year Published

2021
2021
2022
2022

Publication Types

Select...
4
2

Relationship

1
5

Authors

Journals

citations
Cited by 6 publications
(13 citation statements)
references
References 0 publications
0
13
0
Order By: Relevance
“…This map is combined with Sentinel‐2 multispectral data collected at the same site and used for model fitting. The fuzzy logic classifier of Carbonneau et al (2020) is used to isolate sediment pixels only. Afterwards, the model was applied on sediment river bars selected along 300 km of the Po River.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…This map is combined with Sentinel‐2 multispectral data collected at the same site and used for model fitting. The fuzzy logic classifier of Carbonneau et al (2020) is used to isolate sediment pixels only. Afterwards, the model was applied on sediment river bars selected along 300 km of the Po River.…”
Section: Methodsmentioning
confidence: 99%
“…In this work, bands 1, 9 and 10 were not used because these were designed to detect atmospheric quantities, thus the 10 bands available from VIS to SWIR regions constituted the reflectance dataset. The super‐resolution method has a high computational cost, but it was necessary in this work, in accordance with the requirements of the fuzzy logic classifier of Carbonneau et al (2020).…”
Section: Methodsmentioning
confidence: 99%
“…This approach is based on the idea that a pixel in remotely sensed imagery is spatially dependent and likely to be similar to those around it (Berberoglu et al, 2000). The use of a region instead of a single pixel allows for the construction of a small CNN (dubbed "compact CNN" or cCNN: Samarth et al, 2019) with fewer convolutional layers that assigns a class to the central pixel according to the properties of the region (Carbonneau et al, 2020b). It therefore combines spatial and spectral information.…”
Section: Phase 2: Model Architectures and Trainingmentioning
confidence: 99%
“…One such example is the use of drone imagery in training classification algorithms and validation of satellite imagery in estuarine wetlands in the Rachel Carson Reserve in Beaufort, NC, USA (Gray et al., 2018). Further, hyper‐spatial imagery can be used to train fuzzy classification models in fluvial studies (Carbonneau et al., 2020). UAV‐derived DSMs are valuable in peri‐urban areas to detect small‐scale hydrological features like thin walls affecting runoff and flooding (Annis et al., 2020).…”
Section: Other Applicationsmentioning
confidence: 99%