Knowledge Commons of Institute of Automation,CAS
RSSFormer: Foreground Saliency Enhancement for Remote Sensing Land-Cover Segmentatio | |
Rongtao Xu1,3![]() ![]() ![]() ![]() ![]() ![]() | |
发表期刊 | IEEE Transactions on Image Processing
![]() |
ISSN | 1941-0042 |
2023-01 | |
卷号 | 32期号:无页码:1052-1064 |
文章类型 | 期刊论文 |
摘要 | High spatial resolution (HSR) remote sensing images contain complex foreground-background relationships, which makes the remote sensing land cover segmentation a special semantic segmentation task. The main challenges come from the large-scale variation, complex background samples and imbalanced foreground-background distribution. These issues make recent context modeling methods sub-optimal due to the lack of foreground saliency modeling. To handle these problems, we propose a Remote Sensing Segmentation framework (RSSFormer), including Adaptive TransFormer Fusion Module, Detail-aware Attention Layer and Foreground Saliency Guided Loss. Specifically, from the perspective of relation-based fore ground saliency modeling, our Adaptive Transformer Fusion Module can adaptively suppress background noise and enhance object saliency when fusing multi-scale features. Then our Detail-aware Attention Layer extracts the detail and foreground related information via the interplay of spatial attention and channel attention, which further enhances the foreground saliency. From the perspective of optimization-based foreground saliency modeling, our Foreground Saliency Guided Loss can guide the network to focus on hard samples with low foreground saliency responses to achieve balanced optimization. Experimental results on LoveDA datasets, Vaihingen datasets, Potsdam datasets and iSAID datasets validate that our method outperforms existing general semantic segmentation methods and remote sensing segmentation methods, and achieves a good compromise between computational overhead and accuracy. |
关键词 | 遥感图像语义分割 |
学科门类 | 工学 ; 工学::控制科学与工程 ; 工学::计算机科学与技术(可授工学、理学学位) |
DOI | 10.1109/TIP.2023.3238648 |
URL | 查看原文 |
收录类别 | SCI |
语种 | 英语 |
是否为代表性论文 | 是 |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 环境多维感知 |
是否有论文关联数据集需要存交 | 否 |
中文导读 | 高空间分辨率(HSR)遥感图像包含复杂的前景-背景关系,这使得遥感土地覆盖分割成为一项特殊的语义分割任务。主要挑战来自于大尺度变化、复杂的背景样本和不平衡的前景-背景分布。这些问题使最近的上下文建模方法由于缺乏前景显著性建模而不是最优的。为了处理这些问题,我们提出了一个遥感分割框架(RSSFormer),包括自适应Transformer融合模块、细节感知注意层和前景显著性引导损失。。 |
视频解析 | 具体而言,从基于关系的前景显著性建模的角度来看,我们的自适应Transformer融合模块可以在融合多尺度特征时自适应地抑制背景噪声并增强对象显著性。然后,我们的细节感知注意层通过空间注意和通道注意的相互作用提取细节和前景相关信息,从而进一步增强前景显著性。从基于优化的前景显著性建模角度,我们的前景显著性引导损失可以引导网络关注具有低前景显著性响应的难样本,以实现均衡优化。在 LoveDA 数据集、Vaihingen 数据集、Potsdam 数据集和 iSAID 数据集上的实验结果验证了我们的方法优于现有的通用语义分割方法和遥感分割方法,并在计算开销和准确性之间取得了良好的折衷 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/56670 |
专题 | 多模态人工智能系统全国重点实验室_三维可视计算 |
通讯作者 | Shibiao Xu; Weiliang Meng |
作者单位 | 1.State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, China 2.School of Artificial Intelligence, Beijing University of Posts and Telecommunications, China 3.School of Artificial Intelligence, University of Chinese Academy of Sciences, China |
推荐引用方式 GB/T 7714 | Rongtao Xu,Changwei Wang,Jiguang Zhang,et al. RSSFormer: Foreground Saliency Enhancement for Remote Sensing Land-Cover Segmentatio[J]. IEEE Transactions on Image Processing,2023,32(无):1052-1064. |
APA | Rongtao Xu,Changwei Wang,Jiguang Zhang,Shibiao Xu,Weiliang Meng,&Xiaopeng Zhang.(2023).RSSFormer: Foreground Saliency Enhancement for Remote Sensing Land-Cover Segmentatio.IEEE Transactions on Image Processing,32(无),1052-1064. |
MLA | Rongtao Xu,et al."RSSFormer: Foreground Saliency Enhancement for Remote Sensing Land-Cover Segmentatio".IEEE Transactions on Image Processing 32.无(2023):1052-1064. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
RSSFormer_Foreground(8262KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论