IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples
Currently, samples are a critical driving force in the application of deep learning. However, the use of samples encounters problems, such as an inconsistent annotation quality, mismatches between images and labels, and a lack of fine-grained labels. Refining sample labels is essential for training...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/13/2125 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839631606101311488 |
|---|---|
| author | Yibing Xiong Xiangyun Hu Xin Geng Lizhen Lei Aokun Liang |
| author_facet | Yibing Xiong Xiangyun Hu Xin Geng Lizhen Lei Aokun Liang |
| author_sort | Yibing Xiong |
| collection | DOAJ |
| description | Currently, samples are a critical driving force in the application of deep learning. However, the use of samples encounters problems, such as an inconsistent annotation quality, mismatches between images and labels, and a lack of fine-grained labels. Refining sample labels is essential for training a sophisticated model. Refining sample labels through manual verification is labor-intensive, especially for training large models. Additionally, existing label refinement methods based on deep neural networks (DNNs) typically rely on image features to directly predict segmentation results, often overlooking the potential information embedded in existing noisy labels. To address these challenges and shortcomings, this study proposes a novel remote sensing sample label refinement (LR) network, named the identify–update–refine network (IUR-Net). IUR-Net leverages newly acquired remote sensing images and their corresponding noisy labels to automatically identify erroneous regions, update them with more accurate information, and refine the results to improve label quality. A multi-scale, error-aware localization module (Ms-EALM) is designed to capture label–image inconsistencies, enabling the more accurate localization of erroneous label regions. To evaluate the proposed framework, we first constructed and publicly released two benchmark datasets for the label refinement task: WHU-LR and EVLAB-LR. The experimental results on these datasets demonstrate that the labels refined by IUR-Net not only outperform the baseline model in both IoU and F1 scores, but also effectively identify errors in noisy annotations. |
| format | Article |
| id | doaj-art-3d114b14d63241bb9d5f3d29ce948d22 |
| institution | Matheson Library |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-3d114b14d63241bb9d5f3d29ce948d222025-07-11T14:42:12ZengMDPI AGRemote Sensing2072-42922025-06-011713212510.3390/rs17132125IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing SamplesYibing Xiong0Xiangyun Hu1Xin Geng2Lizhen Lei3Aokun Liang4School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaGuangdong Lands and Resource Technology Center, No. 468 Huanshi East Road, Yuexiu District, Guangzhou 510075, ChinaGuangdong Lands and Resource Technology Center, No. 468 Huanshi East Road, Yuexiu District, Guangzhou 510075, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaCurrently, samples are a critical driving force in the application of deep learning. However, the use of samples encounters problems, such as an inconsistent annotation quality, mismatches between images and labels, and a lack of fine-grained labels. Refining sample labels is essential for training a sophisticated model. Refining sample labels through manual verification is labor-intensive, especially for training large models. Additionally, existing label refinement methods based on deep neural networks (DNNs) typically rely on image features to directly predict segmentation results, often overlooking the potential information embedded in existing noisy labels. To address these challenges and shortcomings, this study proposes a novel remote sensing sample label refinement (LR) network, named the identify–update–refine network (IUR-Net). IUR-Net leverages newly acquired remote sensing images and their corresponding noisy labels to automatically identify erroneous regions, update them with more accurate information, and refine the results to improve label quality. A multi-scale, error-aware localization module (Ms-EALM) is designed to capture label–image inconsistencies, enabling the more accurate localization of erroneous label regions. To evaluate the proposed framework, we first constructed and publicly released two benchmark datasets for the label refinement task: WHU-LR and EVLAB-LR. The experimental results on these datasets demonstrate that the labels refined by IUR-Net not only outperform the baseline model in both IoU and F1 scores, but also effectively identify errors in noisy annotations.https://www.mdpi.com/2072-4292/17/13/2125identify–update–refine network (IUR-Net)remote sensingsemantic segmentationlabel refinementcoarse-to-fine |
| spellingShingle | Yibing Xiong Xiangyun Hu Xin Geng Lizhen Lei Aokun Liang IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples Remote Sensing identify–update–refine network (IUR-Net) remote sensing semantic segmentation label refinement coarse-to-fine |
| title | IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples |
| title_full | IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples |
| title_fullStr | IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples |
| title_full_unstemmed | IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples |
| title_short | IUR-Net: A Multi-Stage Framework for Label Refinement Tasks in Noisy Remote Sensing Samples |
| title_sort | iur net a multi stage framework for label refinement tasks in noisy remote sensing samples |
| topic | identify–update–refine network (IUR-Net) remote sensing semantic segmentation label refinement coarse-to-fine |
| url | https://www.mdpi.com/2072-4292/17/13/2125 |
| work_keys_str_mv | AT yibingxiong iurnetamultistageframeworkforlabelrefinementtasksinnoisyremotesensingsamples AT xiangyunhu iurnetamultistageframeworkforlabelrefinementtasksinnoisyremotesensingsamples AT xingeng iurnetamultistageframeworkforlabelrefinementtasksinnoisyremotesensingsamples AT lizhenlei iurnetamultistageframeworkforlabelrefinementtasksinnoisyremotesensingsamples AT aokunliang iurnetamultistageframeworkforlabelrefinementtasksinnoisyremotesensingsamples |