Details
| Original language | English |
|---|---|
| Article number | 471 |
| Number of pages | 24 |
| Journal | Sensors |
| Volume | 21 |
| Issue number | 2 |
| Publication status | Published - 11 Jan 2021 |
Abstract
Invasive blueberry species endanger the sensitive environment of wetlands and protection laws call for management measures. Therefore, methods are needed to identify blueberry bushes, locate them, and characterise their distribution and properties with a minimum of disturbance. UAVs (Unmanned Aerial Vehicles) and image analysis have become important tools for classification and detection approaches. In this study, techniques, such as GIS (Geographical Information Systems) and deep learning, were combined in order to detect invasive blueberry species in wetland environments. Images that were collected by UAV were used to produce orthomosaics, which were analysed to produce maps of blueberry location, distribution, and spread in each study site, as well as bush height and area information. Deep learning networks were used with transfer learning and unfrozen weights in order to automatically detect blueberry bushes reaching True Positive Values (TPV) of 93.83% and an Overall Accuracy (OA) of 98.83%. A refinement of the result masks reached a Dice of 0.624. This study provides an efficient and effective methodology to study wetlands while using different techniques.
Keywords
- ArcGIS, Big data, Blueberries, Deep learning, Image analysis, Orthomosaics, Segmentation refinement, UAVs
ASJC Scopus subject areas
- Chemistry(all)
- Analytical Chemistry
- Biochemistry, Genetics and Molecular Biology(all)
- Biochemistry
- Physics and Astronomy(all)
- Atomic and Molecular Physics, and Optics
- Physics and Astronomy(all)
- Instrumentation
- Engineering(all)
- Electrical and Electronic Engineering
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Sensors, Vol. 21, No. 2, 471, 11.01.2021.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Analysis of UAV-Acquired Wetland Orthomosaics Using GIS, Computer Vision, Computational Topology and Deep Learning
AU - Kentsch, Sarah
AU - Cabezas, Mariano
AU - Tomhave, Luca
AU - Groß, Jens
AU - Burkhard, Benjamin
AU - Lopez Caceres, Maximo Larry
AU - Waki, Katsushi
AU - Diez, Yago
N1 - Funding Information: Acknowledgments: The publication of this article was funded by the Open Access Fund of the Leibniz Universität Hannover. We thank Angie Faust for language corrections and Thomas Beuster (ÖSSM - Ecological protection station Steinhuder Meer) for the opportunity to take aerial photography in the study site and for his helpful contributions to the ecology of blueberries in the Lichtenmoor.
PY - 2021/1/11
Y1 - 2021/1/11
N2 - Invasive blueberry species endanger the sensitive environment of wetlands and protection laws call for management measures. Therefore, methods are needed to identify blueberry bushes, locate them, and characterise their distribution and properties with a minimum of disturbance. UAVs (Unmanned Aerial Vehicles) and image analysis have become important tools for classification and detection approaches. In this study, techniques, such as GIS (Geographical Information Systems) and deep learning, were combined in order to detect invasive blueberry species in wetland environments. Images that were collected by UAV were used to produce orthomosaics, which were analysed to produce maps of blueberry location, distribution, and spread in each study site, as well as bush height and area information. Deep learning networks were used with transfer learning and unfrozen weights in order to automatically detect blueberry bushes reaching True Positive Values (TPV) of 93.83% and an Overall Accuracy (OA) of 98.83%. A refinement of the result masks reached a Dice of 0.624. This study provides an efficient and effective methodology to study wetlands while using different techniques.
AB - Invasive blueberry species endanger the sensitive environment of wetlands and protection laws call for management measures. Therefore, methods are needed to identify blueberry bushes, locate them, and characterise their distribution and properties with a minimum of disturbance. UAVs (Unmanned Aerial Vehicles) and image analysis have become important tools for classification and detection approaches. In this study, techniques, such as GIS (Geographical Information Systems) and deep learning, were combined in order to detect invasive blueberry species in wetland environments. Images that were collected by UAV were used to produce orthomosaics, which were analysed to produce maps of blueberry location, distribution, and spread in each study site, as well as bush height and area information. Deep learning networks were used with transfer learning and unfrozen weights in order to automatically detect blueberry bushes reaching True Positive Values (TPV) of 93.83% and an Overall Accuracy (OA) of 98.83%. A refinement of the result masks reached a Dice of 0.624. This study provides an efficient and effective methodology to study wetlands while using different techniques.
KW - ArcGIS
KW - Big data
KW - Blueberries
KW - Deep learning
KW - Image analysis
KW - Orthomosaics
KW - Segmentation refinement
KW - UAVs
UR - http://www.scopus.com/inward/record.url?scp=85099344060&partnerID=8YFLogxK
U2 - 10.3390/s21020471
DO - 10.3390/s21020471
M3 - Article
VL - 21
JO - Sensors
JF - Sensors
SN - 1424-3210
IS - 2
M1 - 471
ER -