海底磁异常条带是地磁倒转和海底扩张耦合的产物，其形态和分布特征为重塑大洋中各构造单元的形成和演化提供了宝贵依据，具有重要的地球动力学意义。然而，传统磁条带解释主要依赖专家目视判读与人工处理，在大范围数据处理中普遍存在耗时耗力、主观性强、解释一致性不足等问题。为提升处理效率与结果客观性，将YOLOv8 深度学习框架引入海底磁条带自动识别，提出一种从磁异常网格数据输入到分类结果输出的端到端的自动识别方法。研究选取北大西洋两个子区及西太平洋沙茨基海隆周边区域，采用全球磁异常网格数据（EMAG2v3）与高分辨率区域磁异常网格数据，构建了从原始区域磁异常网格到深度学习输入样本的标准化处理流程，通过滑动窗口切片与统一色标渲染生成标准化磁异常图像样本，并基于经纬度映射实现与全球磁条带数据库（GSFML）的精确对齐，建立磁条带/非磁条带（stripe/nonstripe）二分类数据集。基于YOLOv8 轻量化分类模型（YOLOv8n-cls）开展监督训练，在北大西洋和西太平洋区域的验证集上分别取得97.58% 和96.84% 的准确率，测试集F1 分数达0.98 和0.97，能够稳定区分磁条带和非磁条带纹理，展现出良好的识别精度与跨区域可移植性。此外，该方法可输出类别概率，并将其作为识别结果的置信度表征，支持分级筛查与人工复核，有效提升大范围磁异常数据的处理效率，减少人工干预带来的主观误差，为海底扩张研究和构造演化重建提供了新的技术路径，并具有向其他研究区扩展应用的潜力。

Marine magnetic anomaly stripes result from the coupling of geomagnetic reversals and seafloorspreading. Their morphology and spatial distribution provide valuable evidence for reconstructing the formationand evolution of tectonic units in the oceans and therefore carry important geodynamic significance. However,conventional interpretation of magnetic stripes mainly relies on expert visual inspection and profile comparison,and in processing large-area datasets, it commonly suffers from high labor and time costs, strong subjectivity, andlimited interpretive consistency. To improve processing efficiency and result objectivity, this study introduces theYOLOv8 deep-learning framework for automatic marine magnetic stripe recognition and proposes an end-to-endmethod that spans from magnetic anomaly grid data to classification output. The study selects two subregions inthe North Atlantic and the area surrounding Shatsky Rise in the western Pacific and uses global magneticanomaly grid data (EMAG2v3) together with high-resolution regional magnetic anomaly grid data to establish astandardized workflow that transforms raw regional magnetic anomaly grids into deep-learning input samples.Standardized magnetic anomaly image samples are generated through sliding-window slicing and uniformcolormap rendering, and precise alignment with the Global Seafloor Fabric and Magnetic Lineation Data BaseProject (GSFML) is achieved through geographic coordinate mapping, thereby constructing a binaryclassification dataset of stripe and nonstripe samples. This workflow ensures standardized sample generation andpreserves the spatial correspondence between anomaly slices and geological labels. Supervised training is thenconducted using the lightweight YOLOv8 classification model (YOLOv8n-cls), and the optimal weights areselected. The resulting model achieves validation accuracies of 97.58% and 96.84% in the North Atlantic andwestern Pacific regions, respectively, while the test-set F1-scores reach 0.98 and 0.97, demonstrating stablediscrimination between stripe and nonstripe textures, high recognition accuracy, and good cross-regionaltransferability. In addition, the method outputs class probabilities, which are used as confidence measures for therecognition results, thereby supporting hierarchical screening and manual review. This effectively improves theprocessing efficiency of large-scale magnetic anomaly data, reduces subjective errors introduced by manualintervention, provides a new technical pathway for studies of seafloor spreading and tectonic evolutionreconstruction, and shows potential for extension to other study areas.