Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries
Accurately extracting the contours of ground objects has been an important research topic in the field of semantic segmentation of remote sensing imagery. However, existing efforts have primarily focused on refining the boundaries of predictive masks, with little consideration given to pseudo bounda...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843223004399 |
| _version_ | 1827384118793469952 |
|---|---|
| author | Zhong Chen Anqi Cao He Deng Xiaofei Mi Jian Yang |
| author_facet | Zhong Chen Anqi Cao He Deng Xiaofei Mi Jian Yang |
| author_sort | Zhong Chen |
| collection | DOAJ |
| description | Accurately extracting the contours of ground objects has been an important research topic in the field of semantic segmentation of remote sensing imagery. However, existing efforts have primarily focused on refining the boundaries of predictive masks, with little consideration given to pseudo boundaries caused by abrupt changes in surface textures. Therefore, this paper addresses this challenge with the contour preservation network (CPNet), a novel semantic segmentation network that effectively mitigates pseudo-boundary effects and produces more precise contours. The key of CPNet is the boundary-guided feature alignment module (BGAM). This module employs supervised boundary guidance to adaptively transfer the model’s attention from salient areas to correct semantic boundaries. This adaptive attention transfer mechanism enables the model to suppress the impact of internal pseudo boundaries and refine contours. To further refine boundaries, a boundary point feature rectification module (ReBPM) is designed to rectify the classification of boundary points with neighbor features. Extensive experimental validations have demonstrated the effectiveness and flexibility of the proposed CPNet on ISPRS Potsdam and Vaihingen datasets. The results showed that our model outperforms other state-of-the-art methods in terms of boundary IoU, mean IoU, and mean F1-score, and it exhibits significantly superior contour preservation ability compared to other models, notably in the presence of pseudo-boundaries. The code is available at: https://github.com/angiecao/CPNet. |
| first_indexed | 2024-03-08T14:49:58Z |
| format | Article |
| id | doaj.art-553bb8a19e114ca09076e7acf9675a33 |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-03-08T14:49:58Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-553bb8a19e114ca09076e7acf9675a332024-01-11T04:30:27ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322024-02-01126103615Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundariesZhong Chen0Anqi Cao1He Deng2Xiaofei Mi3Jian Yang4School of Artificial Intelligence and Automation, State Key Laboratory for Multispectral Information Processing Technology, Huazhong University of Science and Technology, Wuhan 430074, Hubei, ChinaSchool of Artificial Intelligence and Automation, State Key Laboratory for Multispectral Information Processing Technology, Huazhong University of Science and Technology, Wuhan 430074, Hubei, China; Corresponding author.School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan 430081, Hubei, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAccurately extracting the contours of ground objects has been an important research topic in the field of semantic segmentation of remote sensing imagery. However, existing efforts have primarily focused on refining the boundaries of predictive masks, with little consideration given to pseudo boundaries caused by abrupt changes in surface textures. Therefore, this paper addresses this challenge with the contour preservation network (CPNet), a novel semantic segmentation network that effectively mitigates pseudo-boundary effects and produces more precise contours. The key of CPNet is the boundary-guided feature alignment module (BGAM). This module employs supervised boundary guidance to adaptively transfer the model’s attention from salient areas to correct semantic boundaries. This adaptive attention transfer mechanism enables the model to suppress the impact of internal pseudo boundaries and refine contours. To further refine boundaries, a boundary point feature rectification module (ReBPM) is designed to rectify the classification of boundary points with neighbor features. Extensive experimental validations have demonstrated the effectiveness and flexibility of the proposed CPNet on ISPRS Potsdam and Vaihingen datasets. The results showed that our model outperforms other state-of-the-art methods in terms of boundary IoU, mean IoU, and mean F1-score, and it exhibits significantly superior contour preservation ability compared to other models, notably in the presence of pseudo-boundaries. The code is available at: https://github.com/angiecao/CPNet.http://www.sciencedirect.com/science/article/pii/S1569843223004399Semantic segmentationContour preservationRemote sensing imagesFeature alignment |
| spellingShingle | Zhong Chen Anqi Cao He Deng Xiaofei Mi Jian Yang Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries International Journal of Applied Earth Observations and Geoinformation Semantic segmentation Contour preservation Remote sensing images Feature alignment |
| title | Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries |
| title_full | Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries |
| title_fullStr | Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries |
| title_full_unstemmed | Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries |
| title_short | Accurate contour preservation for semantic segmentation by mitigating the impact of pseudo-boundaries |
| title_sort | accurate contour preservation for semantic segmentation by mitigating the impact of pseudo boundaries |
| topic | Semantic segmentation Contour preservation Remote sensing images Feature alignment |
| url | http://www.sciencedirect.com/science/article/pii/S1569843223004399 |
| work_keys_str_mv | AT zhongchen accuratecontourpreservationforsemanticsegmentationbymitigatingtheimpactofpseudoboundaries AT anqicao accuratecontourpreservationforsemanticsegmentationbymitigatingtheimpactofpseudoboundaries AT hedeng accuratecontourpreservationforsemanticsegmentationbymitigatingtheimpactofpseudoboundaries AT xiaofeimi accuratecontourpreservationforsemanticsegmentationbymitigatingtheimpactofpseudoboundaries AT jianyang accuratecontourpreservationforsemanticsegmentationbymitigatingtheimpactofpseudoboundaries |