A COMBINED PHOTOGRAMMETRIC AND EMPIRICAL ......The empirical method resulted had an overall RMSE of...
1
A COMBINED PHOTOGRAMMETRIC AND EMPIRICAL METHOD FOR SATELLITE-DERIVED BATHYMETRY Bailey Pegels – Shallow Water Earth Observation Lab Department of Geography, Environment, Geomatics University of Ottawa, Ottawa, Ontario Bathymetric mapping is extremely important for shipping, recreation and national defense purposes. Recent groundings in the Canadian Arctic raise questions about how well the Arctic is currently charted. Current charting methods are created using accurate but expensive time- consuming methods (Chénier et al 2017) and the requirement of in situ measurements results in the majority of the Arctic being inadequately charted (CHS 2018). Hodul et al (2018) developed a method that does not require in situ measurements for bathymetric mapping like the frequently used empirical method, which performs best with a homogeneous bottom (Stumpf et al 2003; Chénier et al 2018). This project proposes a method of satellite derived bathymetry (SDB) that uses the photogrammetric method to calibrate the empirical method. This project uses a WorldView-2 stereo pair acquired from DigitalGlobe. The through-water photogrammetry method was completed by creating a DEM in PCIGeomatica’s OrthoEngine. Then a using the refraction correction equation from Hodul et al (2018) to find actual depth: ℎ= tan & ' cos + ' + tan & - cos + - tan . ' cos + ' + tan . - cos + - ∗ℎ ' The empirical method was completed using the band ratio (Figure 1): 0= ln 2 ln 3 where B is the Top of the Atmosphere (ToA) reflectance of the blue band and G is the ToA reflectance of the green band. Depth was calibrated using through-water photogrammetry. Empirical depths were cut out if they had an RMSD greater than 1 meter between photogrammetry depth and empirical depth. The photogrammetric method resulted in an overall RMSE of 1.18 meters and bias of -0.52 meters. Depths greater than 8 meters had RMSE values greater than 2 meters. The empirical method resulted had an overall RMSE of 1.10 meters and a bias of -0.23 meters. The empirical method was able to map more area than the photogrammetric method (Figure 2 and 3). The photogrammetric method mapped 52, 068 m 2 and the empirical method mapped 288, 793 m 2 . Calibrating the empirical method using through- water photogrammetry yields reasonable depth measurements based on accuracy assessments using LiDAR depth measurements. Hodul (2018) was able to obtain depths of up to 12.2 meters compared to the 8 meters derived in this study. Chénier et al (2018) was able to derive depths of 10-15 meters with an RMSE of less than 2 meters when calibrating the empirical method using in situ data on the same image. The combined method is able to map more area than the photogrammetric method alone. Future research should include discovering why the photogrammetry method yielded such different results. This should result in a better calibration of the empirical method. The method should then be tested on more images. More efficient method of Arctic charting is required for safer navigation. The combined photogrammetric and empirical method is able to create charts of Arctic bathymetry. The methods should further be tested. Chénier, R., Faucher, M.-A., & Ahola, R. (2018). Satellite-Derived Bathymetry for Improving Canadian Hydrographic Service Charts. ISPRS International Journal of Geo-Information, 7(8), 306. https://doi.org/10.3390/ijgi7080306 Chénier, R., Abado, L., Sabourin, O., and Tardiff, L., (2017). Northern marine transport corridors: Creation and analysis of northern marine traffic routes in Canadian Waters. Transactions in GIS, 21, 1085-1097. CHS (2018). Arctic charting. Canadian Hydrographic Service. Hodul, Matus. (2018). Photogrammetric Bathymetry for the Canadian Arctic. Université d’Ottawa / University of Ottawa. Retrieved from http://hdl.handle.net/10393/37553 Hodul, Matus, Bird, S., Knudby, A., & Chénier, R. (2018). Satellite derived photogrammetric bathymetry. ISPRS Journal of Photogrammetry and Remote Sensing, 142(Complete), 268–277. https://doi.org/10.1016/j.isprsjprs.2018.06.015 Stumpf, R. P., Holderied, K., & Sinclair, M. (2003). Determination of water depth with high-resolution satellite imagery over variable bottom types. Limnology and Oceanography, 48(1part2), 547–556. https://doi.org/10.4319/lo.2003.48.1_part_2.0547 y = 260.28x 2 - 542.61x + 283.96 0 2 4 6 8 10 12 14 1 1.05 1.1 1.15 1.2 1.25 Photogrammetry depth (m) lnBlue/lnGreen 1 Introduction 2 Methods 4 Discussion 5 Conclusion 6 References 3 Results Figure 1: Results of photogrammetric depth against equation 2. Equation of the trend line was used to calibrate empirical method Figure 2: Photogrammetric method. Imagery ©2015, DigitalGlobe WorldView-2 Figure 3: Empirical method. Imagery ©2015, DigitalGlobe WorldView-2 (1) (2)
Transcript of A COMBINED PHOTOGRAMMETRIC AND EMPIRICAL ......The empirical method resulted had an overall RMSE of...
A COMBINED PHOTOGRAMMETRIC AND EMPIRICAL METHOD FOR SATELLITE-DERIVED BATHYMETRY
Bailey Pegels – Shallow Water Earth Observation Lab
Department of Geography, Environment, GeomaticsUniversity of Ottawa, Ottawa, Ontario
Bathymetric mapping is extremely important for shipping, recreation and national defense purposes. Recent groundings in the Canadian Arctic raise questions about how well the Arctic is currently charted. Current charting methods are created using accurate but expensive time-consuming methods (Chénier et al 2017) and the requirement of in situ measurements results in the majority of the Arctic being inadequately charted (CHS 2018). Hodul et al (2018) developed a method that does not require in situ measurements for bathymetric mapping like the frequently used empirical method, which performs best with a homogeneous bottom (Stumpf et al 2003; Chénier et al 2018). This project proposes a method of satellite derived bathymetry (SDB) that uses the photogrammetric method to calibrate the empirical method.
This project uses a WorldView-2 stereo pair acquired from DigitalGlobe. The through-water photogrammetry method was completed by creating a DEM in PCIGeomatica’s OrthoEngine. Then a using the refraction correction equation from Hodul et al (2018) to find actual depth:
ℎ = tan &' cos +' + tan &- cos +-tan .' cos +' + tan .- cos +-
∗ ℎ'The empirical method was completed using the band ratio (Figure 1):
0 = ln 2ln 3
where B is the Top of the Atmosphere (ToA) reflectance of the blue band and G is the ToAreflectance of the green band. Depth was calibrated using through-water photogrammetry. Empirical depths were cut out if they had an RMSD greater than 1 meter between photogrammetry depth and empirical depth.
The photogrammetric method resulted in an overall RMSE of 1.18 meters and bias of -0.52 meters. Depths greater than 8 meters had RMSE values greater than 2 meters. The empirical method resulted had an overall RMSE of 1.10 meters and a bias of -0.23 meters. The empirical method was able to map more area than the photogrammetric method (Figure 2 and 3). The photogrammetric method mapped 52, 068 m2 and the empirical method mapped 288, 793 m2 .
Calibrating the empirical method using through-water photogrammetry yields reasonable depth measurements based on accuracy assessments using LiDAR depth measurements. Hodul (2018) was able to obtain depths of up to 12.2 meters compared to the 8 meters derived in this study. Chénier et al (2018) was able to derive depths of 10-15 meters with an RMSE of less than 2 meters when calibrating the empirical method using in situ data on the same image. The combined method is able to map more area than the photogrammetric method alone. Future research should include discovering why the photogrammetry method yielded such different results. This should result in a better calibration of the empirical method. The method should then be tested on more images.
More efficient method of Arctic charting is required for safer navigation. The combined photogrammetric and empirical method is able to create charts of Arctic bathymetry. The methods should further be tested.
Chénier, R., Faucher, M.-A., & Ahola, R. (2018). Satellite-Derived Bathymetry for Improving Canadian Hydrographic Service Charts. ISPRS International Journal of Geo-Information, 7(8), 306. https://doi.org/10.3390/ijgi7080306
Chénier, R., Abado, L., Sabourin, O., and Tardiff, L., (2017). Northern marine transport corridors: Creation and analysis of northern marine traffic routes in Canadian Waters. Transactions in GIS, 21, 1085-1097.
CHS (2018). Arctic charting. Canadian Hydrographic Service.
Hodul, Matus. (2018). Photogrammetric Bathymetry for the Canadian Arctic. Université d’Ottawa / University of Ottawa. Retrieved from http://hdl.handle.net/10393/37553
Hodul, Matus, Bird, S., Knudby, A., & Chénier, R. (2018). Satellite derived photogrammetric bathymetry. ISPRS Journal of Photogrammetry and Remote Sensing, 142(Complete), 268–277. https://doi.org/10.1016/j.isprsjprs.2018.06.015
Stumpf, R. P., Holderied, K., & Sinclair, M. (2003). Determination of water depth with high-resolution satellite imagery over variable bottom types. Limnology and Oceanography, 48(1part2), 547–556. https://doi.org/10.4319/lo.2003.48.1_part_2.0547
y = 260.28x2 - 542.61x + 283.96
0
2
4
6
8
10
12
14
1 1.05 1.1 1.15 1.2 1.25P
hoto
gram
met
ry d
epth
(m)
lnBlue/lnGreen
1 Introduction
2 Methods
4 Discussion
5 Conclusion
6 References
3 Results
Figure 1: Results of photogrammetric depth against equation 2. Equation of the trend line was used to calibrate empirical method
Figure 2: Photogrammetric method. Imagery ©2015, DigitalGlobe WorldView-2 Figure 3: Empirical method. Imagery ©2015, DigitalGlobe WorldView-2
(1)
(2)
