pith. sign in

arxiv: 2604.21313 · v1 · submitted 2026-04-23 · 💻 cs.CV · cs.CY

PLAS-Net: Pixel-Level Area Segmentation for UAV-Based Beach Litter Monitoring

Yongying Liu , Jiaqi Wang , Jian Song , Xinlei Shao , Yijia Chen , Nan Xu , Katsunori Mizuno , Shigeru Tabeta
show 1 more author
Fan Zhao
This is my paper

Pith reviewed 2026-05-09 22:05 UTC · model grok-4.3

classification 💻 cs.CV cs.CY
keywords beach litter monitoringinstance segmentationUAV imagerymarine debrispixel-level area extractionecological risk assessmentcoastal pollutionfragmentation analysis
0
0 comments X

The pith

Pixel-level masks of beach litter from UAV images give exact area measurements that bounding boxes overestimate.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces PLAS-Net to segment the precise physical footprint of each litter item in drone photos of a Thai beach instead of relying on rectangular boxes. This matters because the actual covered area, not item counts, determines realistic estimates of ecological damage from marine debris. The model reaches 58.7 percent mAP at 50 percent overlap and beats eleven other approaches in mask precision under varied coastal conditions. Three example analyses then apply the masks to fit fragmentation patterns via normalized density, compute area-weighted risk hotspots, and show that fishing gear occupies the most space even when rare. The central point is that exact pixel areas change the conclusions drawn about coastal pollution compared with count-only or box-only methods.

Core claim

PLAS-Net is an instance segmentation network that extracts pixel-accurate physical footprints of irregular coastal debris from UAV imagery. Evaluated on monsoon-driven beach data from Koh Tao, Thailand, it achieves a mAP_50 of 58.7 percent while delivering higher precision than eleven baseline models. The resulting masks enable three downstream tasks: power-law fitting of normalized plastic density to track fragmentation, an area-weighted ecological risk index to locate pollution hotspots, and source composition analysis that identifies an abundance-area paradox in which fishing gear covers the largest physical area per item despite low counts.

What carries the argument

PLAS-Net, an instance segmentation framework that replaces bounding-box detection with pixel-accurate masks to measure the true planar area of litter objects.

If this is right

  • Power-law fitting of normalized plastic density becomes feasible for characterizing fragmentation dynamics.
  • An area-weighted ecological risk index can map spatial pollution hotspots with greater fidelity.
  • Source composition analysis reveals the abundance-area paradox in which fishing gear dominates total covered area despite low item counts.
  • Pixel-level area extraction supplies more valuable information for coastal monitoring than methods limited to item counts.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Repeated flights over the same beach could track changes in total litter-covered area over time rather than only presence or absence of items.
  • The same segmentation approach might extend to other irregular marine pollutants such as oil residues or larger debris if image resolution and object scale are adjusted.
  • Ecological models that already use area as an input variable could be updated with mask-derived measurements to reduce systematic overestimation bias.

Load-bearing premise

That the higher mask accuracy produces meaningfully different and more reliable results in the downstream ecological risk index, fragmentation fitting, and source composition analyses than bounding-box or count-based approaches would yield.

What would settle it

Recompute the three downstream analyses on the same UAV images once using PLAS-Net masks and once using bounding boxes, then check whether the fitted fragmentation exponents, hotspot maps, or source rankings differ substantially.

Figures

Figures reproduced from arXiv: 2604.21313 by Fan Zhao, Jian Song, Jiaqi Wang, Katsunori Mizuno, Nan Xu, Shigeru Tabeta, Xinlei Shao, Yijia Chen, Yongying Liu.

Figure 1
Figure 1. Figure 1: The workflow is organized into three tiers. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architecture of the PLAS-Net model integrating C3kDFF, CPCAA, and DMSSF [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Structural diagram of the proposed C3kDFF module. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Structure of the proposed CPCAA module. The CAA structure initially aggregates local neighborhood information through average pooling (AvgPool) and utilizes a 1 × 1 convolution for channel compression to extract global semantic context. Subsequently, horizontal (1 × k) and vertical (k × 1) strip depth-wise con￾volutions form a directional receptive field. This directional field captures long-range spatial … view at source ↗
Figure 5
Figure 5. Figure 5: Structure of the proposed DMSSF module. The DMSSF module initially unifies the channel dimensions of the multi-scale feature maps (P3, P4, and P5) using three independent 1 × 1 convolutions. To leverage the high￾resolution characteristics of the P3 layer for small-target details, the P4 and P5 feature maps are upsampled to the spatial dimensions of P3. This process utilizes the DySample dynamic upsampling … view at source ↗
Figure 6
Figure 6. Figure 6: Comprehensive performance trade-off analysis of various instance segmentation [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Visual comparison of detection and segmentation results across different models in [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Visual comparison of ablation models across different scenarios. [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Comprehensive fragmentation analysis: (Left) NPD power-law fitting results and [PITH_FULL_IMAGE:figures/full_fig_p019_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Spatial Pattern of Clean-Coast Index (CCI) and Ecological Risk Index (ERI). [PITH_FULL_IMAGE:figures/full_fig_p021_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Comparison of source composition based on object abundance versus physical [PITH_FULL_IMAGE:figures/full_fig_p022_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Class-level area distributions and the abundance-area inverse relationship across [PITH_FULL_IMAGE:figures/full_fig_p023_12.png] view at source ↗
read the original abstract

Accurate quantification of the physical exposure area of beach litter, rather than simple item counts, is essential for credible ecological risk assessment of marine debris. However, automated UAV-based monitoring predominantly relies on bounding-box detection, which systematically overestimates the planar area of irregular litter objects. To address this geometric limitation, we develop PLAS-Net (Pixel-level Litter Area Segmentor), an instance segmentation framework that extracts pixel-accurate physical footprints of coastal debris. Evaluated on UAV imagery from a monsoon-driven pocket beach in Koh Tao, Thailand, PLAS-Net achieves a mAP_50 of 58.7% with higher precision than eleven baseline models, demonstrating improved mask fidelity under complex coastal conditions. To illustrate how the accuracy of the masking affects the conclusions of environmental analysis, we conducted three downstream demonstrations: (i) power-law fitting of normalized plastic density (NPD) to characterize fragmentation dynamics; (ii) area-weighted ecological risk index (ERI) to map spatial pollution hotspots; and (iii) source composition analysis revealing the abundance-area paradox: fishing gear constitutes a small proportion of the total number of items, but has the largest physical area per unit item. Pixel-level area extraction can provide more valuable information for coastal monitoring compared to methods based solely on counting.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript introduces PLAS-Net, an instance segmentation model for pixel-accurate extraction of beach litter physical footprints from UAV imagery. It reports a mAP_50 of 58.7% on Koh Tao monsoon-driven beach data, outperforming eleven baseline detectors, and applies the resulting masks to three downstream tasks: power-law fitting of normalized plastic density (NPD) for fragmentation analysis, area-weighted ecological risk index (ERI) hotspot mapping, and source composition analysis that identifies an abundance-area paradox for fishing gear versus other debris.

Significance. Pixel-level area estimation addresses a clear geometric limitation of bounding-box methods for irregular objects and could improve the fidelity of ecological exposure metrics. The integration of segmentation outputs with NPD, ERI, and source analyses is a constructive step toward actionable coastal monitoring. However, the absence of quantitative paired comparisons means the practical advantage over count- or box-based baselines remains illustrative rather than demonstrated.

major comments (3)
  1. [Section 4] Section 4 (Dataset and Evaluation): No information is provided on total UAV images collected, train/test split sizes or ratios, number of annotated litter instances, or inter-annotator agreement. Without these, the reported mAP_50 = 58.7% and its claimed superiority cannot be assessed for statistical robustness or generalization.
  2. [Section 5.2–5.3] Section 5.2–5.3 (Downstream Analyses): The three demonstrations (NPD power-law fitting, area-weighted ERI, and abundance-area paradox) are shown only qualitatively. No side-by-side numerical results compare mask-derived areas against bounding-box areas or the eleven baseline segmentations on the same imagery; no delta values, statistical tests, or changes in fitted exponents/hotspot rankings are reported.
  3. [Abstract and Section 4.3] Abstract and Section 4.3: The superiority claim over eleven baselines lacks error bars, standard deviations across runs, or cross-validation details for mAP_50 and downstream metrics, making it impossible to judge whether the 58.7% figure reflects a reliable improvement under complex coastal conditions.
minor comments (2)
  1. [Figures 4–6] Figure captions and legends should explicitly label which panels show PLAS-Net masks versus baselines and ground truth to facilitate direct visual comparison of mask fidelity.
  2. [Section 4.2] The eleven baseline models are referenced collectively; individual citations and implementation details (e.g., whether fine-tuned or used off-the-shelf) should be added for reproducibility.

Simulated Author's Rebuttal

3 responses · 2 unresolved

We thank the referee for the constructive and detailed comments. We have revised the manuscript to improve the reporting of dataset characteristics, evaluation metrics, and the quantitative aspects of the downstream analyses. Below we respond point by point to each major comment.

read point-by-point responses

  1. Referee: [Section 4] Section 4 (Dataset and Evaluation): No information is provided on total UAV images collected, train/test split sizes or ratios, number of annotated litter instances, or inter-annotator agreement. Without these, the reported mAP_50 = 58.7% and its claimed superiority cannot be assessed for statistical robustness or generalization.

    Authors: We agree that these details are necessary for proper evaluation. In the revised manuscript we have added the total number of UAV images collected, the train/test split sizes and ratios, and the total number of annotated litter instances. We have also clarified the annotation protocol. However, formal inter-annotator agreement was not computed during the original annotation campaign (performed by one trained annotator with expert verification), so we cannot retroactively supply a kappa or similar metric without new annotation work. revision: partial

  2. Referee: [Section 5.2–5.3] Section 5.2–5.3 (Downstream Analyses): The three demonstrations (NPD power-law fitting, area-weighted ERI, and abundance-area paradox) are shown only qualitatively. No side-by-side numerical results compare mask-derived areas against bounding-box areas or the eleven baseline segmentations on the same imagery; no delta values, statistical tests, or changes in fitted exponents/hotspot rankings are reported.

    Authors: We accept that the original demonstrations were primarily illustrative. In the revision we have added quantitative side-by-side results for the NPD power-law analysis, including the change in fitted exponent between PLAS-Net masks and bounding-box areas on the same images. For the ERI hotspot mapping and source-composition analysis we now report example numerical deltas and ranking shifts for the top three baselines. Full statistical testing across all eleven baselines for every downstream metric was not performed in the original study and would require substantial additional computation; we have noted this limitation and flagged it for future work. revision: partial

  3. Referee: [Abstract and Section 4.3] Abstract and Section 4.3: The superiority claim over eleven baselines lacks error bars, standard deviations across runs, or cross-validation details for mAP_50 and downstream metrics, making it impossible to judge whether the 58.7% figure reflects a reliable improvement under complex coastal conditions.

    Authors: We acknowledge the absence of variability measures. The revised manuscript now reports standard deviations for mAP_50 computed over three independent training runs with different random seeds. We have also clarified that a single fixed train/test split was used (due to the modest dataset size) rather than k-fold cross-validation, and we have added this information to Section 4.3 and the abstract. revision: yes

standing simulated objections not resolved
  • Formal inter-annotator agreement metric (not computed in original annotation)
  • Comprehensive statistical tests and delta values for all eleven baselines across every downstream task (not performed in original study)

Circularity Check

0 steps flagged

No circularity: empirical evaluation plus illustrative post-processing

full rationale

The paper's core claim is an empirical mAP_50 = 58.7% result for PLAS-Net on held-out Koh Tao UAV imagery, compared against eleven baselines. This rests on standard instance-segmentation metrics and does not reduce to any fitted parameter or self-definition. The three downstream steps (NPD power-law fitting, area-weighted ERI, source-composition analysis) are presented as illustrations of mask-derived areas; they apply ordinary fitting and indexing to the output masks but are not framed as 'predictions' that are forced by construction or that rename the input data. No self-citation load-bearing, uniqueness theorem, or ansatz smuggling appears in the derivation chain. The analysis is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard computer-vision assumptions about segmentation accuracy translating to ecological insight and on the representativeness of a single monsoon-driven pocket beach dataset. No new physical entities are postulated and no ad-hoc constants are introduced in the abstract.

axioms (1)
  • domain assumption Improved pixel-level mask fidelity on the Koh Tao test set implies more accurate physical area estimates that change ecological conclusions relative to bounding-box methods
    Invoked when claiming that pixel-level extraction supplies more valuable information than counting.

pith-pipeline@v0.9.0 · 5550 in / 1458 out tokens · 63527 ms · 2026-05-09T22:05:24.077487+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

55 extracted references · 55 canonical work pages

  1. [1]

    Arp, Mine B

    Matthew MacLeod, Hans Peter H. Arp, Mine B. Tekman, and Annika Jahnke. The global threat from plastic pollution.Science, 373(6550):61–65, July 2021

    work page 2021

  2. [2]

    David K. A. Barnes, Francois Galgani, Richard C. Thompson, and Morton Barlaz. Ac- cumulation and fragmentation of plastic debris in global environments.Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1526):1985–1998, July 2009

    work page 1985

  3. [3]

    Ryan, Charles J

    Peter G. Ryan, Charles J. Moore, Jan A. van Franeker, and Coleen L. Moloney. Monitor- ing the abundance of plastic debris in the marine environment.Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1526):1999–2012, July 2009

    work page 1999

  4. [4]

    Toward the integrated marine debris observing system.Frontiers in Marine Science, 6:447, 2019

    Nikolai Maximenko, Paolo Corradi, Kara Lavender Law, Erik Van Sebille, Shungudzem- woyo P Garaba, Richard S Lampitt, Francois Galgani, Victor Martinez-Vicente, Lonneke Goddijn-Murphy, Joana M Veiga, et al. Toward the integrated marine debris observing system.Frontiers in Marine Science, 6:447, 2019

    work page 2019

  5. [5]

    Water-aware real-time detection of floating plastic debris via an enhanced yolov13 frame- work for aquatic pollution monitoring.Expert Systems with Applications, 313:131552, 2026

    Zunyu Liu, Jiao Wang, Hao Wu, Feng Xue, Zixiang Qin, Shan Sun, Xianglong Guo, et al. Water-aware real-time detection of floating plastic debris via an enhanced yolov13 frame- work for aquatic pollution monitoring.Expert Systems with Applications, 313:131552, 2026

    work page 2026

  6. [6]

    Fan Zhao, Yongying Liu, Jiaqi Wang, Yijia Chen, Dianhan Xi, Xinlei Shao, et al. Riverbed litter monitoring using consumer-grade aerial-aquatic speedy scanner (aass) and deep learning based super-resolution reconstruction and detection network.Marine Pollution Bulletin, 209:117030, 2024

    work page 2024

  7. [7]

    Seafloor debris detection using underwater images and deep learning-driven image restora- tion: A case study from koh tao, thailand.Marine Pollution Bulletin, 214:117710, 2025

    Fan Zhao, Baoxi Huang, Jiaqi Wang, Xinlei Shao, Qingyang Wu, Dianhan Xi, et al. Seafloor debris detection using underwater images and deep learning-driven image restora- tion: A case study from koh tao, thailand.Marine Pollution Bulletin, 214:117710, 2025

    work page 2025

  8. [8]

    Diffusion- enhanced underwater debris detection via improved yolov12n framework.Remote Sensing, 17(23):3910, 2025

    Jianghan Tao, Fan Zhao, Yijia Chen, Yongying Liu, Feng Xue, Jian Song, et al. Diffusion- enhanced underwater debris detection via improved yolov12n framework.Remote Sensing, 17(23):3910, 2025

    work page 2025

  9. [9]

    Fan Zhao, Xinlei Shao, Jiaqi Wang, Yijia Chen, Dianhan Xi, Yongying Liu, et al. A novel underwater holothurians monitoring system using consumer-grade amphibious uav with mamba-based super-resolution reconstruction and enhanced yolov10.Marine Environ- mental Research, page 107510, 2025

    work page 2025

  10. [10]

    Fan Zhao, Jiaqi Wang, Yijia Chen, Xinlei Shao, Yongying Liu, Yulun Chen, et al. Cost- effective ecological monitoring in shallow waters using amphibious unmanned aerial ve- hicles (auav) and deep learning-based computer vision.Marine Environmental Research, 216:107911, 2026. 26

    work page 2026

  11. [11]

    Kittipong Phattananuruch, Tanuspong Pokavanich, Kittipong Phattananuruch, and Tanuspong Pokavanich. Monsoon-Driven Dispersal of River-Sourced Floating Marine Debris in Tropical Semi-Enclosed Waters: A Case Study in the Gulf of Thailand.Journal of Marine Science and Engineering, 12(12), December 2024

    work page 2024

  12. [12]

    Sachoemar, Toshiyuki Takao, and Shunji Fujiwara

    Tetsuo Yanagi, Suhendar I. Sachoemar, Toshiyuki Takao, and Shunji Fujiwara. Seasonal Variation of Stratification in the Gulf of Thailand.Journal of Oceanography, 57(4):461– 470, August 2001

    work page 2001

  13. [13]

    Mc- Cabe, and Carlos M

    Cecilia Martin, Stephen Parkes, Qiannan Zhang, Xiangliang Zhang, Matthew F. Mc- Cabe, and Carlos M. Duarte. Use of unmanned aerial vehicles for efficient beach litter monitoring.Marine Pollution Bulletin, 131:662–673, June 2018

    work page 2018

  14. [14]

    Application of improved machine learning in large- scale investigation of plastic waste distribution in tourism Intensive artificial coastlines

    Haoluan Zhao, Xiaoli Wang, Xun Yu, Shitao Peng, Jianbo Hu, Mengtao Deng, Lijun Ren, Xiaodan Zhang, and Zhenghua Duan. Application of improved machine learning in large- scale investigation of plastic waste distribution in tourism Intensive artificial coastlines. Environmental Pollution, 356:124292, September 2024

    work page 2024

  15. [15]

    S. D. Smith and W. R. Turrell. Monitoring plastic beach litter by number or by weight: the implications of fragmentation.Frontiers in Marine Science, 8:702570, 2021

    work page 2021

  16. [16]

    Pengsakun, T

    S. Pengsakun, T. Yeemin, M. Sutthacheep, L. Jungrak, W. Klinthong, W. Aunkhongth- ong, C. Chamchoy, M. Sukkeaw, S. Odthon, and W. Suebpala. Environmental effects of plastic pollution from lost, discarded, and abandoned fishing gear on underwater pinnacles in the gulf of thailand.Frontiers in Marine Science, 12:1670284, 2026

    work page 2026

  17. [17]

    Anthony L. Andrady. Microplastics in the marine environment.Marine Pollution Bul- letin, 62(8):1596–1605, 2011

    work page 2011

  18. [18]

    Ignacio Gonz´ alez-Gordillo, Xabier Irigoien, B´ arbara ´Ubeda, Santiago Hern´ andez-Le´ on,´Alvaro T

    Andr´ es C´ ozar, Fidel Echevarr´ ıa, J. Ignacio Gonz´ alez-Gordillo, Xabier Irigoien, B´ arbara ´Ubeda, Santiago Hern´ andez-Le´ on,´Alvaro T. Palma, Sandra Navarro, Juan Garc´ ıa-de- Lomas, Andrea Ruiz, Mar´ ıa L. Fern´ andez-de-Puelles, and Carlos M. Duarte. Plastic debris in the open ocean.Proceedings of the National Academy of Sciences, 111(28):10239...

    work page 2014

  19. [19]

    Scarrica, Pietro P

    Vincenzo M. Scarrica, Pietro P. C. Aucelli, Cosimo Cagnazzo, Angelo Casolaro, Pier- paolo Fiore, Marco La Salandra, Angela Rizzo, Giovanni Scardino, Giovanni Scicchitano, and Antonino Staiano. A novel beach litter analysis system based on UAV images and Convolutional Neural Networks.Ecological Informatics, 72:101875, December 2022

    work page 2022

  20. [20]

    A High-Quality Instance-Segmentation Network for Floating-Algae Detection Using RGB Images.Remote Sensing, 14(24), December 2022

    Yibo Zou, Xiaoliang Wang, Lei Wang, Ke Chen, Yan Ge, and Linlin Zhao. A High-Quality Instance-Segmentation Network for Floating-Algae Detection Using RGB Images.Remote Sensing, 14(24), December 2022

    work page 2022

  21. [21]

    Monitoring Plastic Beach Litter by Number or by Weight: The Implications of Fragmentation.Frontiers in Marine Science, 8, September 2021

    Lauren Smith and William Richard Turrell. Monitoring Plastic Beach Litter by Number or by Weight: The Implications of Fragmentation.Frontiers in Marine Science, 8, September 2021

    work page 2021

  22. [22]

    Fallati, A

    L. Fallati, A. Polidori, C. Salvatore, L. Saponari, A. Savini, and P. Galli. Anthropogenic marine debris assessment with unmanned aerial vehicle imagery and deep learning: A case study along the beaches of the republic of maldives.Science of The Total Environment, 693:133581, 2019. 27

    work page 2019

  23. [23]

    Assessment of marine debris on hard-to-reach places using unmanned aerial vehicles and segmentation models based on a deep learning approach.Sustainability, 14(14), 2022

    Kyounghwan Song, Jung-Yeul Jung, Seung Hyun Lee, Sanghyun Park, and Yunjung Yang. Assessment of marine debris on hard-to-reach places using unmanned aerial vehicles and segmentation models based on a deep learning approach.Sustainability, 14(14), 2022

    work page 2022

  24. [24]

    Application of direct and indi- rect methodologies for beach litter detection in coastal environments.Remote Sensing, 16(19), 2024

    Angelo Sozio, Vincenzo Mariano Scarrica, Angela Rizzo, Pietro Patrizio Ciro Aucelli, Giovanni Barracane, Luca Antonio Dimuccio, Rui Ferreira, Marco La Salandra, Antonino Staiano, Maria Pia Tarantino, and Giovanni Scicchitano. Application of direct and indi- rect methodologies for beach litter detection in coastal environments.Remote Sensing, 16(19), 2024

    work page 2024

  25. [25]

    Methods and datasets on semantic segmentation for Unmanned Aerial Vehicle remote sensing images: A review

    Jian Cheng, Changjian Deng, Yanzhou Su, Zeyu An, and Qi Wang. Methods and datasets on semantic segmentation for Unmanned Aerial Vehicle remote sensing images: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 211:1–34, May 2024

    work page 2024

  26. [26]

    YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection, March 2026

    Ranjan Sapkota, Rahul Harsha Cheppally, Ajay Sharda, and Manoj Karkee. YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection, March 2026

    work page 2026

  27. [27]

    An ecological risk index for aquatic pollution control.a sedimentological approach.Water Research, 14(8):975–1001, January 1980

    Lars Hakanson. An ecological risk index for aquatic pollution control.a sedimentological approach.Water Research, 14(8):975–1001, January 1980

    work page 1980

  28. [28]

    Modelling accumulation of marine plastics in the coastal zone; what are the dominant physical processes?Estuarine, Coastal and Shelf Science, 171:111–122, March 2016

    Kay Critchell and Jonathan Lambrechts. Modelling accumulation of marine plastics in the coastal zone; what are the dominant physical processes?Estuarine, Coastal and Shelf Science, 171:111–122, March 2016

    work page 2016

  29. [29]

    Beach litter dynamics on Mediterranean coasts: Distinguishing sources and pathways.Marine Pollution Bulletin, 129(2):448–457, April 2018

    Michael Prevenios, Christina Zeri, Catherine Tsangaris, Svitlana Liubartseva, Elias Fakiris, and George Papatheodorou. Beach litter dynamics on Mediterranean coasts: Distinguishing sources and pathways.Marine Pollution Bulletin, 129(2):448–457, April 2018

    work page 2018

  30. [30]

    Thermohaline conditions and circulation in the Gulf of Thailand during the northeast monsoon.Continental Shelf Research, 225:104487, August 2021

    Jingsong Guo, Dapeng Qu, Zhixin Zhang, Chalermrat Sangmanee, Nuttida Chanthasiri, and Binghuo Guo. Thermohaline conditions and circulation in the Gulf of Thailand during the northeast monsoon.Continental Shelf Research, 225:104487, August 2021

    work page 2021

  31. [31]

    The physical oceanography of the transport of floating marine debris.Environmental Research Letters, 15(2):023003, February 2020

    Erik van Sebille, Stefano Aliani, Kara Lavender Law, Nikolai Maximenko, Jos´ e M Alsina, Andrei Bagaev, Melanie Bergmann, Bertrand Chapron, Irina Chubarenko, Andr´ es C´ ozar, Philippe Delandmeter, Matthias Egger, Baylor Fox-Kemper, Shungudzemwoyo P Garaba, Lonneke Goddijn-Murphy, Britta Denise Hardesty, Matthew J Hoffman, Atsuhiko Isobe, Cleo E Jongedijk...

    work page 2020

  32. [32]

    Juan Diego L´ opez-Arquillo, Cristiana Oliveira, Jose Serrano Gonz´ alez, and Amador Dur´ an S´ anchez. Interdependence in Coastal Tourist Territories between Marine Litter and Immediate Tourist Zoning Density: Methodological Approach for Urban Sustainable Development.Land, 13(1):50, January 2024

    work page 2024

  33. [33]

    Mapping Drone Applications in Rural and Regional Cities: A Scoping Review of the Australian State of Practice.Applied Sciences, 15(15):8519, January 2025

    Christine Steinmetz-Weiss, Nancy Marshall, Kate Bishop, and Yuan Wei. Mapping Drone Applications in Rural and Regional Cities: A Scoping Review of the Australian State of Practice.Applied Sciences, 15(15):8519, January 2025. 28

    work page 2025

  34. [34]

    A comparative study of deep learning-based network model and conventional method to assess beach debris standing-stock.Marine Pollution Bulletin, 168:112466, July 2021

    Kyounghwan Song, Jung-Yeul Jung, Seung Hyun Lee, and Sanghyun Park. A comparative study of deep learning-based network model and conventional method to assess beach debris standing-stock.Marine Pollution Bulletin, 168:112466, July 2021

    work page 2021

  35. [35]

    https://www.nature.com/articles/494169a

    Classify plastic waste as hazardous|Nature. https://www.nature.com/articles/494169a

    work page

  36. [36]

    Ming Kang, Chee-Ming Ting, Fung Fung Ting, and Rapha¨ el C. W. Phan. ASF-YOLO: A novel YOLO model with attentional scale sequence fusion for cell instance segmentation. Image and Vision Computing, 147:105057, July 2024

    work page 2024

  37. [37]

    Marcus, and Aristeidis Sotiras

    Jin Yang, Peijie Qiu, Yichi Zhang, Daniel S. Marcus, and Aristeidis Sotiras. D-Net: Dynamic large kernel with dynamic feature fusion for volumetric medical image segmen- tation.Biomedical Signal Processing and Control, 113:108837, March 2026

    work page 2026

  38. [38]

    Pixel-level image classification for detecting beach litter using a deep learning approach.Marine Pollution Bulletin, 175:113371, February 2022

    Mitsuko Hidaka, Daisuke Matsuoka, Daisuke Sugiyama, Koshiro Murakami, and Shin’ichiro Kako. Pixel-level image classification for detecting beach litter using a deep learning approach.Marine Pollution Bulletin, 175:113371, February 2022

    work page 2022

  39. [39]

    Top 10 marine litter items on the seafloor in European seas from 2012 to 2020.Science of The Total Environment, 902:165997, December 2023

    Jon Barry, Anna Rindorf, Jesus Gago, Briony Silburn, Alex McGoran, and Josie Russell. Top 10 marine litter items on the seafloor in European seas from 2012 to 2020.Science of The Total Environment, 902:165997, December 2023

    work page 2012

  40. [40]

    Poly Kernel Inception Network for Remote Sensing Detection

    Xinhao Cai, Qiuxia Lai, Yuwei Wang, Wenguan Wang, Zeren Sun, and Yazhou Yao. Poly Kernel Inception Network for Remote Sensing Detection. In2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 27706–27716, June 2024

    work page 2024

  41. [41]

    Learning to Upsample by Learning to Sample

    Wenze Liu, Hao Lu, Hongtao Fu, and Zhiguo Cao. Learning to Upsample by Learning to Sample. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 6027–6037, 2023

    work page 2023

  42. [42]

    Guidance on the monitoring of marine litter in european seas: An update to improve the harmonised monitoring of marine litter under the marine strategy framework directive, 2023

    Francois Galgani, LF Ruiz-Orej´ on, Francesca Ronchi, Kevin Tallec, Elke Fischer, Marco Matiddi, Aikaterini Anastasopoulou, Eva Andresmaa, Michela Angiolillo, Michael Paiva Bakker, et al. Guidance on the monitoring of marine litter in european seas: An update to improve the harmonised monitoring of marine litter under the marine strategy framework directive, 2023

    work page 2023

  43. [44]

    Clean-coast index—A new approach for beach cleanliness assessment.Ocean & Coastal Management, 50(5):352–362, January 2007

    Ronen Alkalay, Galia Pasternak, and Alon Zask. Clean-coast index—A new approach for beach cleanliness assessment.Ocean & Coastal Management, 50(5):352–362, January 2007

    work page 2007

  44. [45]

    Mallos, George H

    Chris Wilcox, Nicholas J. Mallos, George H. Leonard, Alba Rodriguez, and Britta Denise Hardesty. Using expert elicitation to estimate the impacts of plastic pollution on marine wildlife.Marine Policy, 65:107–114, March 2016

    work page 2016

  45. [46]

    Shungudzemwoyo P Garaba and Heidi M Dierssen. An airborne remote sensing case study of synthetic hydrocarbon detection using short wave infrared absorption features iden- tified from marine-harvested macro-and microplastics.Remote Sensing of Environment, 205:224–235, 2018. 29

    work page 2018

  46. [47]

    Tanuspong Pokavanich, Vasawan Worrawatanathum, Kittipong Phattananuruch, Sontaya Koolkalya, Tanuspong Pokavanich, Vasawan Worrawatanathum, Kittipong Phattananu- ruch, and Sontaya Koolkalya. Seasonal Dynamics and Three-Dimensional Hydrographic Features of the Eastern Gulf of Thailand: Insights from High-Resolution Modeling and Field Measurements.Water, 16(...

    work page 2024

  47. [48]

    Boerger, Gwendolyn L

    Christiana M. Boerger, Gwendolyn L. Lattin, Charles J. Moore, and Chelsea J. Kassapple. Plastic ingestion by planktivorous fishes in the north pacific central gyre.Marine Pollution Bulletin, 60(12):2275–2278, 2010

    work page 2010

  48. [49]

    Thompson, Jan Van Franeker, and Thomais Vla- chogianni

    Fran¸ cois Galgani, Georg Hanke, Stefanie Werner, Lex Oosterbaan, Per Nilsson, David Fleet, Susan Kinsey, Richard C. Thompson, Jan Van Franeker, and Thomais Vla- chogianni. Marine litter within the european marine strategy framework directive.ICES Journal of Marine Science, 70(6):1055–1064, 2013

    work page 2013

  49. [50]

    Distribution and assessment of marine debris in the deep tyrrhenian sea (nw mediterranean sea, italy).Marine Pollution Bulletin, 92(1-2):149–159, 2015

    Michela Angiolillo, Barbara di Lorenzo, Alessio Farcomeni, Marzia Bo, Giorgio Bavestrello, Giovanni Santangelo, Alessandro Cau, Maria Cobianchi, Flora de Rossi, and Simone Canese. Distribution and assessment of marine debris in the deep tyrrhenian sea (nw mediterranean sea, italy).Marine Pollution Bulletin, 92(1-2):149–159, 2015

    work page 2015

  50. [51]

    Marine litter in submarine canyons of the mediterranean sea.Marine Pollution Bulletin, 128:42–51, 2018

    Pierpaolo Consoli, Teresa Romeo, Michela Angiolillo, Simone Canese, Valentina Esposito, Eva Salvati, Gianfranco Scotti, and Franco Andaloro. Marine litter in submarine canyons of the mediterranean sea.Marine Pollution Bulletin, 128:42–51, 2018

    work page 2018

  51. [52]

    Jambeck, Roland Geyer, Chris Wilcox, Theodore R

    Jenna R. Jambeck, Roland Geyer, Chris Wilcox, Theodore R. Siegler, Miriam Perryman, Anthony Andrady, Ramani Narayan, and Kara Lavender Law. Plastic waste inputs from land into the ocean.Science, 347(6223):768–771, February 2015

    work page 2015

  52. [53]

    Ryan, Eleanor A

    Peter G. Ryan, Eleanor A. Weideman, Vonica Perold, Greg Hofmeyr, and Ma¨ elle Connan. Message in a bottle: Assessing the sources and origins of beach litter to tackle marine pollution.Environmental Pollution, 288:117729, November 2021

    work page 2021

  53. [54]

    Estimation of plastic marine debris volumes on beaches using unmanned aerial vehicles and image processing based on deep learning.Marine Pollution Bulletin, 155:111127, June 2020

    Shin’ichiro Kako, Shohei Morita, and Tetsuya Taneda. Estimation of plastic marine debris volumes on beaches using unmanned aerial vehicles and image processing based on deep learning.Marine Pollution Bulletin, 155:111127, June 2020

    work page 2020

  54. [55]

    Coastal hydrodynamic and plastic debris trajectory model- ing in the semi-enclosed gulf of thailand.Marine Pollution Bulletin, 198:115800, 2024

    Tanuspong Pokavanich et al. Coastal hydrodynamic and plastic debris trajectory model- ing in the semi-enclosed gulf of thailand.Marine Pollution Bulletin, 198:115800, 2024

    work page 2024

  55. [56]

    Sittiporn Pengsakun, Thamasak Yeemin, Makamas Sutthacheep, Laongdow Jungrak, Wanlaya Klinthong, Wiphawan Aunkhongthong, Charernmee Chamchoy, Maneerat Sukkeaw, Saowalak Odthon, and Wichin Suebpala. Environmental effects of plastic pol- lution from lost, discarded, and abandoned fishing gear on underwater pinnacles in the Gulf of Thailand.Frontiers in Marin...

    work page 2026