CP2M: Clustered-Patch-Mixed Mosaic Augmentation for Aerial Image Segmentation

TL;DR

This paper introduces CP2M, a novel data augmentation method for remote sensing image segmentation that combines mosaic and clustered patch mixing to improve model generalization and reduce overfitting.

Contribution

The paper proposes CP2M, a new augmentation strategy that enhances data diversity and spatial coherence in remote sensing image segmentation.

Findings

CP2M significantly reduces overfitting in segmentation models.

CP2M achieves state-of-the-art accuracy on the ISPRS Potsdam dataset.

CP2M improves model robustness in remote sensing tasks.

Abstract

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.