The article introduces YOLOv8-C2f-Faster-EMA, an improved underwater trash detection model based on YOLOv8. This model addresses the challenges of high miss and false detection rates in aquatic environments. The algorithm optimizes the backbone, neck layer, and C2f module for underwater characteristics and incorporates an effective attention mechanism. It improves the accuracy of underwater litter detection while simplifying the computational model. Empirical evidence shows that the method outperforms the conventional YOLOv8n framework, achieving a 6.7% increase in precision (P), a 4.1% surge in recall (R), and a 5% enhancement in mean average precision (mAP). The model has potential for integration into remote sensing applications, enhancing detection precision in localized surveillance. The YOLOv8 network has advanced underwater trash detection, demonstrating significantly enhanced efficiency and precision. This breakthrough could revolutionize the ability to identify small objects in underwater environments and holds promise for remote sensing applications. The study outlines three key advancements: the introduction of the C2f-Faster-EMA module, the integration of FasterNet and Efficient Multiscale Attention (EMA) modules, and a comprehensive suite of benchmarking experiments. The YOLOv8 network architecture is described, highlighting its design for real-time object detection. The FasterNet architecture is introduced, focusing on computational efficiency and performance. The EMA module is presented as a novel attention mechanism that enhances feature extraction and detection accuracy. The study presents an improved model based on YOLOv8, optimized for underwater target detection. The model includes enhanced C2f modules, an efficient attention mechanism, and improved backbone and neck layers. The model was evaluated using the TRASH-ICRA19 dataset, demonstrating significant improvements in detection performance. The results show that the YOLOv8-C2f-Faster-EMA model achieves high precision, recall, and mAP, outperforming other models in terms of efficiency and accuracy. The study also examines the effects of FasterNet, EMA, and Mosaic on model performance, demonstrating their contributions to improved detection accuracy. The model is compared with other object detection systems, showing its superiority in terms of computational efficiency and detection performance. The study concludes that the YOLOv8-C2f-Faster-EMA model is a significant advancement in underwater trash detection, offering high accuracy and efficiency for real-time applications.The article introduces YOLOv8-C2f-Faster-EMA, an improved underwater trash detection model based on YOLOv8. This model addresses the challenges of high miss and false detection rates in aquatic environments. The algorithm optimizes the backbone, neck layer, and C2f module for underwater characteristics and incorporates an effective attention mechanism. It improves the accuracy of underwater litter detection while simplifying the computational model. Empirical evidence shows that the method outperforms the conventional YOLOv8n framework, achieving a 6.7% increase in precision (P), a 4.1% surge in recall (R), and a 5% enhancement in mean average precision (mAP). The model has potential for integration into remote sensing applications, enhancing detection precision in localized surveillance. The YOLOv8 network has advanced underwater trash detection, demonstrating significantly enhanced efficiency and precision. This breakthrough could revolutionize the ability to identify small objects in underwater environments and holds promise for remote sensing applications. The study outlines three key advancements: the introduction of the C2f-Faster-EMA module, the integration of FasterNet and Efficient Multiscale Attention (EMA) modules, and a comprehensive suite of benchmarking experiments. The YOLOv8 network architecture is described, highlighting its design for real-time object detection. The FasterNet architecture is introduced, focusing on computational efficiency and performance. The EMA module is presented as a novel attention mechanism that enhances feature extraction and detection accuracy. The study presents an improved model based on YOLOv8, optimized for underwater target detection. The model includes enhanced C2f modules, an efficient attention mechanism, and improved backbone and neck layers. The model was evaluated using the TRASH-ICRA19 dataset, demonstrating significant improvements in detection performance. The results show that the YOLOv8-C2f-Faster-EMA model achieves high precision, recall, and mAP, outperforming other models in terms of efficiency and accuracy. The study also examines the effects of FasterNet, EMA, and Mosaic on model performance, demonstrating their contributions to improved detection accuracy. The model is compared with other object detection systems, showing its superiority in terms of computational efficiency and detection performance. The study concludes that the YOLOv8-C2f-Faster-EMA model is a significant advancement in underwater trash detection, offering high accuracy and efficiency for real-time applications.