RL-ISegNet: Refining Instance Segmentation for Remote Sensing Imagery via Iterative Reward Maximization With Limited Samples
文献类型：期刊论文
| 作者 | Liu, Yang5; Zhang, Tong5; Huang, Yanru1,2; Wang, Peixiao3; Wang, Gang4 |
| 刊名 | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
 |
| 出版日期 | 2025 |
| 卷号 | 63页码:5637521 |
| 关键词 | Instance segmentation Training Optimization Remote sensing Reinforcement learning Data models Data mining Trajectory Computational modeling Natural language processing limited samples reinforcement learning (RL) remote sensing imagery (RSI) |
| ISSN号 | 0196-2892 |
| DOI | 10.1109/TGRS.2025.3598943 |
| 产权排序 | 4 |
| 文献子类 | Article |
| 英文摘要 | Instance segmentation of remote sensing imagery (RSI) is vital for applications like geographic information system (GIS) updates and urban planning. Due to RSI's diversity (e.g., scale variations and complex object shapes), instance segmentation models require extensive labeled data, which is costly and labor-intensive. To mitigate this dependence on extensive labeled data, we propose a reinforcement learning (RL)-driven iteratively-refined instance segmentation model for RSI by RL (RL-ISegNet) framework, which uses policy gradient optimization to enhance the learning of hard samples in limited labeled data, thereby improving data utilization efficiency under limited samples. Specifically, our RL-ISegNet defines a sequential optimization process for hard samples, using cumulative rewards from each subtask (region proposal, detection, and segmentation) to provide future feedback guidance for optimizing network parameters. Meanwhile, considering the conflicts between subtasks, we propose a metric to quantify the positive or negative impacts between subtasks as a loss constraint, thereby increasing the sampling probability of trajectories with positive impacts. Additionally, we developed a Fisher information-based method to freeze low uncertainty parameters, reducing the training time. The experimental results show that our RL-ISegNet improves data utilization efficiency by +12.2% and achieves improvements of +7.1% in instance segmentation metrics with 4% of the training dataset. |
| URL标识 | 查看原文 |
| WOS关键词 | OBJECT DETECTION ; CLASSIFICATION |
| WOS研究方向 | Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001560173500005 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/216079]  |
| 专题 | 资源与环境信息系统国家重点实验室_外文论文 |
| 通讯作者 | Zhang, Tong |
| 作者单位 | 1.Chinese Acad Sci, Aerosp Informat Res Inst, State Key Lab Digital Earth Sci, Beijing 100094, Peoples R China; 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China; 3.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Geog Informat Sci & Technol, Beijing 100101, Peoples R China; 4.Tianjin Railway Tech & Vocat Coll, Tianjin 300240, Peoples R China 5.Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & Re, Wuhan 430072, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Liu, Yang,Zhang, Tong,Huang, Yanru,et al. RL-ISegNet: Refining Instance Segmentation for Remote Sensing Imagery via Iterative Reward Maximization With Limited Samples[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2025,63:5637521. |
| APA | Liu, Yang,Zhang, Tong,Huang, Yanru,Wang, Peixiao,&Wang, Gang.(2025).RL-ISegNet: Refining Instance Segmentation for Remote Sensing Imagery via Iterative Reward Maximization With Limited Samples.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,63,5637521. |
| MLA | Liu, Yang,et al."RL-ISegNet: Refining Instance Segmentation for Remote Sensing Imagery via Iterative Reward Maximization With Limited Samples".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 63(2025):5637521. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
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。