发表机构
UiT the Arctic University of Norway; Politecnico di Milano; NORCE, The Norwegian Research Centre AS（挪威北极大学; 米兰理工大学; NORCE挪威研究中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对Sentinel-1 SAR影像的极地低压像素级分割难题，提出弱监督框架CREST，结合CORE模块与DB损失，在SAR及乳腺超声、人体数据集上均优于标准AER。
AI 中文摘要
极地低压是在高纬度地区快速形成的强烈海洋气旋。深度学习可在合成孔径雷达（SAR）影像中检测极地低压，但像素级分割仍是未解决的挑战：训练时无可用的像素级掩码，且极地低压的范围本质上具有主观性，其边界模糊，即便是专家绘制也不一致。我们提出了带软目标的约束区域擦除（CREST），这是一种仅用图像级标签生成掩码的弱监督语义分割（WSSS）框架。该方法基于对抗擦除（AER）构建，AER会迭代挖掘判别性区域、擦除这些区域并重新训练分类器，以揭示互补线索，这些线索会成为分割的伪标签。但标准AER也会收集无关的背景特征，降低伪标签质量。CREST通过两点解决该问题：（i）约束有序区域扩展（CORE）模块，其编码极地低压的空间连通性先验，约束从高置信度种子开始的区域扩展；（ii）动态引导（DB）损失，将挖掘顺序作为标签可靠性的代理，衰减来自噪声更大的后期挖掘区域的监督。在Sentinel-1 SAR数据上，CREST比标准AER更贴合气旋结构，且返回多类而非二元掩码，其类别表示为每个区域分配的可靠性。我们还在BUS-UCLM乳腺超声和PASCAL VOC人体数据上进行评估，这些数据的目标满足相同的连通性先验，但具备SAR数据缺乏的密集掩码。在相同设置下，CREST在这两个数据集上的表现优于等效的AER流程。
英文摘要
Polar lows are intense maritime cyclones that form rapidly at high latitudes. Deep learning can detect them in Synthetic Aperture Radar (SAR) imagery, but pixel-level segmentation remains an open challenge. No pixel-level masks are available for training, and a polar low's extent is inherently subjective, with diffuse boundaries that even experts delineate inconsistently. We propose Constrained Region Erasing with Soft Targets (CREST), a Weakly Supervised Semantic Segmentation (WSSS) framework that generates masks solely from image-level labels. Our approach builds on Adversarial Erasing (AER), which iteratively mines discriminative regions, erases them, and retrains a classifier to reveal complementary cues that become pseudo-labels for segmentation. However, standard AER also collects irrelevant background features, degrading pseudo-label quality. CREST addresses this with (i) a Constrained Ordinal Region Expansion (CORE) module that encodes the spatial-connectedness prior of polar lows, constraining region expansion from a high-confidence seed, and (ii) a Dynamic Bootstrapping (DB) loss that treats the mining order as a proxy for label reliability, attenuating supervision from noisier, later-mined regions. On Sentinel-1 SAR data, CREST follows the cyclone structure more closely than standard AER, and returns a multi-class rather than binary mask whose classes indicate the reliability assigned to each region. We further evaluate on BUS-UCLM breast ultrasound and PASCAL VOC person data, whose targets satisfy the same connectedness prior but come with the dense masks the SAR data lacks. On both datasets, CREST performs better than the equivalent AER pipeline under identical settings.
