Litter Detection in Coastal Areas: A Segmentation Application with YOLO Family R-CNNs

• Melinne Diniz de Oliveira UEA
• Elloá B. Guedes UEA

DOI: https://doi.org/10.5753/wcama.2024.1902

Abstract

This work aims to train and evaluate Deep Learning models from the YOLO Family to segment instances of artificial waste in images of coastal regions, aiming to contribute to the monitoring and depollution of these environments. A data preparation pipeline was developed and then YOLOv7 and YOLOv8 models were tested, with YOLOv7 standing out for its best performance, possibly due to its lesser specialization for benchmarks. The results aim to assist in the creation of automatic systems to detect and remove plastic waste in coastal areas, in order to reduce the environmental impacts of this form of pollution.

References

Andrades, R., Pegado, T., Godoy, B. S., Reis-Filho, J. A., Nunes, J. L. S., Grillo, A. C., Machado, R. C., Santos, R. G., Dalcin, R. H., Freitas, M. O., Kuhnen, V. V., Barbosa, N. D., Adelir-Alves, J., Albuquerque, T., Bentes, B., and Giarrizzo, T. (2020). Anthropogenic litter on brazilian beaches: Baseline, trends and recommendations for future approaches. Mar. Pollut. Bull., 151(110842):110842.

Bao, Z., Sha, J., Li, X., Hanchiso, T., and Shifaw, E. (2018). Monitoring of beach litter by automatic interpretation of unmanned aerial vehicle images using the segmentation threshold method. Marine Pollution Bulletin, 137:388–398.

Cormen, T. H., Leiserson, C. E., Rivest, R. L., and Stein, C. (2009). Introduction to Algorithms. The MIT Press, 3 edition.

Diwan, T., Anirudh, G., and Tembhurne, J. V. (2023). Object Detection Using YOLO: Challenges, Architectural Successors, Datasets and Applications. Multimed. Tools Appl., 82(6):9243–9275.

Hidaka, M., Matsuoka, D., Sugiyama, D., Murakami, K., and Kako, S. (2022). Pixel-level image classification for detecting beach litter using a deep learning approach. Mar. Pollut. Bull., 175(113371):113371.

Jocher, G., Chaurasia, A., and Qiu, J. (2023). YOLOv8 – real-time object detection. Disponível em [link]. Acesso em 16 de maio de 2024.

Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C. L. (2014). Microsoft COCO: Common Objects in Context. In Proceedings of the 13th European Conference on Computer Vision (ECVV 2014), pages 740–755, Suíça. Springer.

Martin, C., Parkes, S., Zhang, Q., Zhang, X., McCabe, M. F., and Duarte, C. M. (2018). Use of unmanned aerial vehicles for efficient beach litter monitoring. Marine Pollution Bulletin, 131:662–673.

Michelucci, U. (2019). Advanced Applied Deep Learning: Convolutional Neural Networks and Object Detection. Apress, Suíça.

ONU (2015). Objetivos de desenvolvimento sustentável – sdgs transform our world. Assembléia Geral da ONU.

ONU (2023). Oceans. Disponível em [link]. Acesso em 16 de maio de 2024.

ONU (2024). Plastic Pollution and Marine Litter. Disponível em [link]. Acesso em 16 de maio de 2024.

Padilla, R., Netto, S. L., and da Silva, E. A. B. (2020). A Survey on Performance Metrics for Object-Detection Algorithms. In 2020 International Conference on Systems, Signals and Image Processing (IWSSIP), pages 237–242, Niterói, Brasil.

Romera-Castillo, C., Lucas, A., Mallenco-Fornies, R., Briones-Rizo, M., Calvo, E., and Pelejero, C. (2023). Abiotic plastic leaching contributes to ocean acidification. Science of The Total Environment, 854:158683.

Simul Bhuyan, M., Venkatramanan, Selvam, Szabo, S., Maruf Hossain, M., Rashed-Un-Nabi, M., Paramasivam, Jonathan, and Shafiqul Islam, M. (2021). Plastics in marine ecosystem: A review of their sources and pollution conduits. Reg. Stud. Mar. Sci., 41(101539):101539.

Sugiyama, D., Hidaka, M., Matsuoka, D., Murakami, K., and Kako, S. (2022). The BeachLitter dataset for image segmentation of beach litter. Data Brief, 42(108072):108072.

Tamin, O., Moung, E. G., Dargham, J. A., Yahya, F., Farzamnia, A., Sia, F., Naim, N. F. M., and Angeline, L. (2023). On-shore plastic waste detection with yolov5 and rgb-near-infrared fusion: A state-of-the-art solution for accurate and efficient environmental monitoring. Big Data and Cognitive Computing, 7(2).

Terven, J., Córdova-Esparza, D.-M., and Romero-González, J.-A. (2023). A comprehensive review of YOLO architectures in computer vision: From YOLOv1 to YOLOv8 and YOLONAS. Mach. Learn. Knowl. Extr., 5(4):1680–1716.

Veettil, B. K., Hong Quan, N., Hauser, L. T., Doan Van, D., and Quang, N. X. (2022). Coastal and marine plastic litter monitoring using remote sensing: A review. Estuarine, Coastal and Shelf Science, 279:108160.

Wang, C.-Y., Bochkovskiy, A., and Liao, H.-Y. M. (2022). YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Disponível em [link]. Acesso em 16 de maio de 2024.

Wang, C.-Y., Liao, H.-Y. M., Wu, Y.-H., Chen, P.-Y., Hsieh, J.-W., and Yeh, I.-H. (2020). Cspnet: A new backbone that can enhance learning capability of cnn. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshop (CVPR Workshop).