CP2M: Clustered-Patch-Mixed Mosaic Augmentation for Aerial Image Segmentation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912205295845376 |
|---|---|
| author | Li, Yijie Wang, Hewei Xu, Jinfeng Ma, Zixiao Wu, Puzhen Wang, Shaofan Dev, Soumyabrata |
| author_facet | Li, Yijie Wang, Hewei Xu, Jinfeng Ma, Zixiao Wu, Puzhen Wang, Shaofan Dev, Soumyabrata |
| contents | Remote sensing image segmentation is pivotal for earth observation, underpinning applications such as environmental monitoring and urban planning. Due to the limited annotation data available in remote sensing images, numerous studies have focused on data augmentation as a means to alleviate overfitting in deep learning networks. However, some existing data augmentation strategies rely on simple transformations that may not sufficiently enhance data diversity or model generalization capabilities. This paper proposes a novel augmentation strategy, Clustered-Patch-Mixed Mosaic (CP2M), designed to address these limitations. CP2M integrates a Mosaic augmentation phase with a clustered patch mix phase. The former stage constructs a new sample from four random samples, while the latter phase uses the connected component labeling algorithm to ensure the augmented data maintains spatial coherence and avoids introducing irrelevant semantics when pasting random patches. Our experiments on the ISPRS Potsdam dataset demonstrate that CP2M substantially mitigates overfitting, setting new benchmarks for segmentation accuracy and model robustness in remote sensing tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_15389 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CP2M: Clustered-Patch-Mixed Mosaic Augmentation for Aerial Image Segmentation Li, Yijie Wang, Hewei Xu, Jinfeng Ma, Zixiao Wu, Puzhen Wang, Shaofan Dev, Soumyabrata Computer Vision and Pattern Recognition Remote sensing image segmentation is pivotal for earth observation, underpinning applications such as environmental monitoring and urban planning. Due to the limited annotation data available in remote sensing images, numerous studies have focused on data augmentation as a means to alleviate overfitting in deep learning networks. However, some existing data augmentation strategies rely on simple transformations that may not sufficiently enhance data diversity or model generalization capabilities. This paper proposes a novel augmentation strategy, Clustered-Patch-Mixed Mosaic (CP2M), designed to address these limitations. CP2M integrates a Mosaic augmentation phase with a clustered patch mix phase. The former stage constructs a new sample from four random samples, while the latter phase uses the connected component labeling algorithm to ensure the augmented data maintains spatial coherence and avoids introducing irrelevant semantics when pasting random patches. Our experiments on the ISPRS Potsdam dataset demonstrate that CP2M substantially mitigates overfitting, setting new benchmarks for segmentation accuracy and model robustness in remote sensing tasks. |
| title | CP2M: Clustered-Patch-Mixed Mosaic Augmentation for Aerial Image Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.15389 |