HED-UNet：结合分割与边缘检测以监测南极海岸线
HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline
AI总结：
本文提出 HED-UNet，融合 UNet 分割与 HED 边缘检测，并通过深度监督和分层注意力融合多尺度预测，在南极 Sentinel-1 影像上提升海岸线检测性能。
摘要：
近年来，基于深度学习的海岸线检测算法已开始胜过传统统计方法。然而，它们通常只被训练为单一用途模型，要么对陆地与水体进行分割，要么描绘海岸线。相比之下，人工标注者在执行手动海岸线检测时，通常会在脑中同时保留分割和描绘的映射。为兼顾这种任务双重性，我们因此设计了一种新模型，在一个深度学习模型中统一这两种方法。通过从语义分割框架（UNet）和边缘检测框架（HED）的主要构建模块中汲取灵感，两项任务以自然方式结合。训练通过在多种分辨率的侧边预测上采用深度监督而变得高效。最后，引入一种分层注意力机制，将这些多尺度预测自适应地融合为最终模型输出。该方法相对于其他传统和基于深度学习的海岸线检测方法的优势，在一个覆盖南极海岸部分区域的 Sentinel-1 影像数据集上得到证明，而该地区的海岸线检测众所周知地困难。我们方法的实现可在 https://github.com/khdlr/HED-UNet 获取。
英文摘要：
Deep learning-based coastline detection algorithms have begun to outshine traditional statistical methods in recent years. However, they are usually trained only as single-purpose models to either segment land and water or delineate the coastline. In contrast to this, a human annotator will usually keep a mental map of both segmentation and delineation when performing manual coastline detection. To take into account this task duality, we therefore devise a new model to unite these two approaches in a deep learning model. By taking inspiration from the main building blocks of a semantic segmentation framework (UNet) and an edge detection framework (HED), both tasks are combined in a natural way. Training is made efficient by employing deep supervision on side predictions at multiple resolutions. Finally, a hierarchical attention mechanism is introduced to adaptively merge these multiscale predictions into the final model output. The advantages of this approach over other traditional and deep learning-based methods for coastline detection are demonstrated on a dataset of Sentinel-1 imagery covering parts of the Antarctic coast, where coastline detection is notoriously difficult. An implementation of our method is available at \url{https://github.com/khdlr/HED-UNet}.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
