Mapping marine litter with Unmanned Aerial Systems: A showcase comparison among manual image screening and machine learning techniques.

Recent works have shown the feasibility of Unmanned Aerial Systems (UAS) for monitoring marine pollution. We provide a comparison among techniques to detect and map marine litter objects on an UAS-derived orthophoto of a sandy beach-dune system. Manual image screening technique allowed a detailed description of marine litter categories. Random forest classifier returned the best-automated detection rate (F-score 70%), while convolutional neural network performed slightly worse (F-score 60%) due to a higher number of false positive detections. We show that automatic methods allow faster and more frequent surveys, while still providing a reliable density map of the marine litter load. Image manual screening should be preferred when the characterization of marine litter type and material is required. Our analysis suggests that the use of UAS-derived orthophoto is appropriate to obtain a detailed geolocation of marine litter items, requires much less human effort and allows a wider area coverage.

[1]  P. Randerson,et al.  Distribution of beach litter along the coastline of Cádiz, Spain. , 2016, Marine pollution bulletin.

[2]  D. Mcglashan,et al.  Development and management of a coastal litter campaign: The voluntary coastal partnership approach , 2006 .

[3]  M. Costa,et al.  Anthropogenic Litter on Beaches With Different Levels of Development and Use: A Snapshot of a Coast in Pernambuco (Brazil) , 2018, Front. Mar. Sci..

[4]  A. Isobe,et al.  Sequential monitoring of beach litter using webcams. , 2010, Marine pollution bulletin.

[5]  Tomoya Kataoka,et al.  A new technique for detecting colored macro plastic debris on beaches using webcam images and CIELUV. , 2012, Marine pollution bulletin.

[6]  D. Fischer,et al.  Quantification and classification of marine litter on the municipal beach of Ensenada, Baja California, Mexico. , 2003, Marine pollution bulletin.

[7]  B. Öztürk,et al.  Origin and abundance of marine litter along sandy beaches of the Turkish Western Black Sea Coast. , 2013, Marine environmental research.

[8]  Weitao Chen,et al.  Forested landslide detection using LiDAR data and the random forest algorithm: A case study of the Three Gorges, China , 2014 .

[9]  K. Edyvane,et al.  Long-term marine litter monitoring in the remote Great Australian Bight, South Australia. , 2004, Marine pollution bulletin.

[10]  Elias Fakiris,et al.  Beach litter dynamics on Mediterranean coasts: Distinguishing sources and pathways. , 2017, Marine pollution bulletin.

[11]  C. Rosevelt,et al.  Marine debris in central California: quantifying type and abundance of beach litter in Monterey Bay, CA. , 2013, Marine pollution bulletin.

[12]  P. Galli,et al.  Anthropogenic Marine Debris assessment with Unmanned Aerial Vehicle imagery and deep learning: A case study along the beaches of the Republic of Maldives. , 2019, The Science of the total environment.

[13]  C. Botero,et al.  Litter impacts on scenery and tourism on the Colombian north Caribbean coast , 2016 .

[14]  George Vosselman,et al.  Multi-Resolution Feature Fusion for Image Classification of Building Damages with Convolutional Neural Networks , 2018, Remote. Sens..

[15]  S. Cheung,et al.  Field test of beach litter assessment by commercial aerial drone. , 2020, Marine pollution bulletin.

[16]  J. Creed,et al.  Quantification and Classification of Marine Litter on Beaches along Armação dos Búzios, Rio de Janeiro, Brazil , 2007 .

[17]  F. Galgani,et al.  Global Distribution, Composition and Abundance of Marine Litter , 2015 .

[18]  W. Cai,et al.  The abundance, composition and sources of marine debris in coastal seawaters or beaches around the northern South China Sea (China). , 2011, Marine pollution bulletin.

[19]  S. Pomponi,et al.  Oceans and human health: Emerging public health risks in the marine environment. , 2006, Marine pollution bulletin.

[20]  R. L. Savio,et al.  A baseline assessment of beach macrolitter and microplastics along northeastern Atlantic shores , 2019, Marine Pollution Bulletin.

[21]  R. Carriço,et al.  Spatio-temporal variability of beached macro-litter on remote islands of the North Atlantic. , 2018, Marine pollution bulletin.

[22]  T. Kusui,et al.  International survey on the distribution of stranded and buried litter on beaches along the Sea of Japan. , 2003, Marine pollution bulletin.

[23]  M. Schulz,et al.  OSPAR standard method and software for statistical analysis of beach litter data. , 2017, Marine pollution bulletin.

[24]  Brian Neilson,et al.  Mapping coastal marine debris using aerial imagery and spatial analysis. , 2017, Marine pollution bulletin.

[25]  K. Pond,et al.  Marine litter monitoring programmes—A review of methods with special reference to national surveys , 1995 .

[26]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[27]  Daeseok Kang,et al.  Quantities, composition, and sources of beach debris in Korea from the results of nationwide monitoring. , 2014, Marine pollution bulletin.

[28]  J. Niemeyer,et al.  Contextual classification of lidar data and building object detection in urban areas , 2014 .

[29]  Colin P.D. Birch,et al.  Rectangular and hexagonal grids used for observation, experiment and simulation in ecology , 2007 .

[30]  C. Munari,et al.  Marine litter on Mediterranean shores: Analysis of composition, spatial distribution and sources in north-western Adriatic beaches. , 2016, Waste management.

[31]  G. Schernewski,et al.  Monitoring methods for large micro- and meso-litter and applications at Baltic beaches , 2018, Journal of Coastal Conservation.

[32]  U. Andriolo,et al.  Mapping marine litter using UAS on a beach-dune system: a multidisciplinary approach. , 2019, The Science of the total environment.

[33]  J. Gago,et al.  Characteristics (abundance, type and origin) of beach litter on the Galician coast (NW Spain) from 2001 to 2010 , 2014 .

[34]  M. Witt,et al.  Marine anthropogenic litter on British beaches: A 10-year nationwide assessment using citizen science data. , 2017, The Science of the total environment.

[35]  A. Deidun,et al.  Optimising beached litter monitoring protocols through aerial imagery. , 2018, Marine pollution bulletin.

[36]  Martin Schulz,et al.  Daily accumulation rates of marine debris on sub-Antarctic island beaches. , 2013, Marine pollution bulletin.

[37]  Jinming Sha,et al.  Monitoring of beach litter by automatic interpretation of unmanned aerial vehicle images using the segmentation threshold method. , 2018, Marine pollution bulletin.

[38]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[39]  Veiga Joana Mira,et al.  Identifying Sources of Marine Litter , 2016 .

[40]  François Galgani,et al.  Marine litter, future prospects for research , 2015, Front. Mar. Sci..

[41]  Z. Lucas.  Monitoring persistent litter in the marine environment on Sable Island, Nova Scotia , 1992 .

[42]  O. Ogunola,et al.  Mitigation measures to avert the impacts of plastics and microplastics in the marine environment (a review) , 2018, Environmental Science and Pollution Research.

[43]  Norman Kerle,et al.  Structural Building Damage Detection with Deep Learning: Assessment of a State-of-the-Art CNN in Operational Conditions , 2019, Remote. Sens..

[44]  Wenji Mao,et al.  Advances in Intelligence and Security Informatics , 2012 .

[45]  Xiangliang Zhang,et al.  Use of unmanned aerial vehicles for efficient beach litter monitoring. , 2018, Marine pollution bulletin.

[46]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[47]  David J. Fleet,et al.  Statistical analyses of the results of 25 years of beach litter surveys on the south-eastern North Sea coast. , 2015, Marine environmental research.

[48]  P. Sobral,et al.  Local marine litter survey: A case study in Alcobaça municipality, Portugal , 2013 .

[49]  M. Malavasi,et al.  Beach litter occurrence in sandy littorals: The potential role of urban areas, rivers and beach users in central Italy , 2016 .