TY - JOUR
T1 - Enabling a large-scale assessment of litter along Saudi Arabian red sea shores by combining drones and machine learning.
AU - Martin, Cecilia
AU - Zhang, Qiannan
AU - Zhai, Dongjun
AU - Zhang, Xiangliang
AU - Duarte, Carlos M.
N1 - KAUST Repository Item: Exported on 2021-03-05
Acknowledgements: We thank Julia Reisser, Elena Corona, Núria Marbá and Dorte Krause-Jensen for help during fieldwork; the staff from the Coastal and Marine Resources core lab at KAUST for support during sampling cruises. We thank Matthew McCabe and Stephen Parkes for providing the DJI Phantom 3 Adv and the relative training. We thank anonymous reviewers and editors for the useful comments during the revision process.
PY - 2021/3/2
Y1 - 2021/3/2
N2 - Beach litter assessments rely on time inefficient and high human cost protocols, mining the attainment of global beach litter estimates. Here we show the application of an emerging technique, the use of drones for acquisition of high-resolution beach images coupled with machine learning for their automatic processing, aimed at achieving the first national-scale beach litter survey completed by only one operator. The aerial survey had a time efficiency of 570 ± 40 m2 min-1 and the machine learning reached a mean (±SE) detection sensitivity of 59 ± 3% with high resolution images. The resulting mean (±SE) litter density on Saudi Arabian shores of the Red Sea is of 0.12 ± 0.02 litter items m-2, distributed independently of the population density in the area around the sampling station. Instead, accumulation of litter depended on the exposure of the beach to the prevailing wind and litter composition differed between islands and the main shore, where recreational activities are the major source of anthropogenic debris.
AB - Beach litter assessments rely on time inefficient and high human cost protocols, mining the attainment of global beach litter estimates. Here we show the application of an emerging technique, the use of drones for acquisition of high-resolution beach images coupled with machine learning for their automatic processing, aimed at achieving the first national-scale beach litter survey completed by only one operator. The aerial survey had a time efficiency of 570 ± 40 m2 min-1 and the machine learning reached a mean (±SE) detection sensitivity of 59 ± 3% with high resolution images. The resulting mean (±SE) litter density on Saudi Arabian shores of the Red Sea is of 0.12 ± 0.02 litter items m-2, distributed independently of the population density in the area around the sampling station. Instead, accumulation of litter depended on the exposure of the beach to the prevailing wind and litter composition differed between islands and the main shore, where recreational activities are the major source of anthropogenic debris.
UR - http://hdl.handle.net/10754/667865
UR - https://linkinghub.elsevier.com/retrieve/pii/S0269749121003109
U2 - 10.1016/j.envpol.2021.116730
DO - 10.1016/j.envpol.2021.116730
M3 - Article
C2 - 33652184
SN - 0269-7491
VL - 277
SP - 116730
JO - Environmental pollution (Barking, Essex : 1987)
JF - Environmental pollution (Barking, Essex : 1987)
ER -