对比YOLO系列在水下视觉任务中的表现，发现YOLOv10最适合AUV部署。
A Comparative Study of YOLOv8 to YOLOv11 Performance in Underwater Vision Tasks
- 在水下图像上对比YOLOv8至v11，固定训练条件评估性能。
- YOLOv9后精度饱和，但v10-v11推理速度显著提升。
- 提供开源数据集与代码，助力水下视觉研究可复现。
自主水下航行器（AUV）日益依赖板载计算机视觉系统完成栖息地制图、生态监测和基础设施检查。然而，水下影像受光衰减、浑浊度及严重类别不平衡影响，且计算资源有限。单阶段检测器如YOLO系列因其融合定位与分类的低延迟特性备受青睐，但其陆地基准（COCO、PASCAL-VOC、Open Images）未涵盖海洋场景下的性能演进。本研究构建两个公开数据集：珊瑚疾病集（4,480张图像，18类）和鱼类物种集（7,500张图像，20类），针对每个数据集设计四种训练比例（25%、50%、75%、100%），保持验证与测试集平衡固定。使用相同超参数（100轮、640像素输入、批大小16、T4 GPU）训练YOLOv8-s、YOLOv9-s、YOLOv10-s、YOLOv11-s，并评估精确率、召回率、mAP50、mAP50-95、每图像推理时间与帧率（FPS）。事后Grad-CAM可视化分析特征利用与定位可靠性。结果显示，在两类数据集上，准确率在YOLOv9后趋于饱和，表明架构创新更侧重效率而非精度；但推理速度显著提升。研究首次实现近期YOLO版本在水下图像上的可控比较，证明轻量级YOLOv10在嵌入式AUV部署中具有最优速度-精度权衡，并开放提供基准数据集与代码库，推动未来海洋视觉研究发展。
原文摘要 · Abstract (English)
Autonomous underwater vehicles (AUVs) increasingly rely on on-board computer-vision systems for tasks such as habitat mapping, ecological monitoring, and infrastructure inspection. However, underwater imagery is hindered by light attenuation, turbidity, and severe class imbalance, while the computational resources available on AUVs are limited. One-stage detectors from the YOLO family are attractive because they fuse localization and classification in a single, low-latency network; however, their terrestrial benchmarks (COCO, PASCAL-VOC, Open Images) leave open the question of how successive YOLO releases perform in the marine domain. We curate two openly available datasets that span contrasting operating conditions: a Coral Disease set (4,480 images, 18 classes) and a Fish Species set (7,500 images, 20 classes). For each dataset, we create four training regimes (25 %, 50 %, 75 %, 100 % of the images) while keeping balanced validation and test partitions fixed. We train YOLOv8-s, YOLOv9-s, YOLOv10-s, and YOLOv11-s with identical hyperparameters (100 epochs, 640 px input, batch = 16, T4 GPU) and evaluate precision, recall, mAP50, mAP50-95, per-image inference time, and frames-per-second (FPS). Post-hoc Grad-CAM visualizations probe feature utilization and localization faithfulness. Across both datasets, accuracy saturates after YOLOv9, suggesting architectural innovations primarily target efficiency rather than accuracy. Inference speed, however, improves markedly. Our results (i) provide the first controlled comparison of recent YOLO variants on underwater imagery, (ii) show that lightweight YOLOv10 offers the best speed-accuracy trade-off for embedded AUV deployment, and (iii) deliver an open, reproducible benchmark and codebase to accelerate future marine-vision research.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。