RESUMO
This study developed an automatic monitoring system for Floating Marine Debris (FMD) aimed at reducing the labor-intensiveness of traditional visual surveys. It involved creating a comprehensive FMD database using 55.6 h of video footage and numerous annotated images, which facilitated the training of a deep learning model based on the YOLOv8 architecture. Additionally, the study implemented the BoT-SORT algorithm for FMD tracking, significantly enhancing detection accuracy by effectively filtering out disturbances such as sea waves and seabirds, based on the movement patterns observed in FMD trajectories. Tested across 16 voyages in various marine environments, the system demonstrated high accuracy in recognizing different types of FMD, achieving a mean Average Precision (mAP@0.5) of 0.97. In terms of detecting FMD from video footage, the system reached an F1 score of 83.63 %. It showed potential as a viable substitute for manual methods for FMD larger than 20 cm.
Assuntos
Monitoramento Ambiental , Navios , Monitoramento Ambiental/métodos , Resíduos , Gravação em Vídeo , AlgoritmosRESUMO
To verify data accuracy regarding microplastics (MPs), the effects of chemical preprocessing and optical identification with Fourier transform infrared spectroscopy (FTIR) were examined using survey data. We aimed to clarify the cause of overestimation of the number concentration of MPs without FTIR by focusing on shape types. The proportion of fragments (foam) decreased (increased) after FTIR analysis. Although significant difference was found in terms of the size distribution of fragments (after FTIR > before FTIR), no such difference was found with regard to foam. These findings indicate that impurities categorized as fragments via visual confirmation, especially those measuring <1000 µm, were removed using the FTIR technique. Although previous studies recommended the use of data on MPs measuring >1000 µm, this threshold should be applied only to fragments and not to foam.