MultiVeg: A Very High-Resolution Benchmark for Deep Learning-Based Multi-Class Vegetation Segmentation

Highlights: What are the main findings? MultiVeg is the first satellite-based very high-resolution (0.5 m) multi-class vegetation segmentation benchmark constructed from KOMPSAT-3 and 3A imagery, providing multispectral (blue, green, red, and near-infrared) image patches with reliable annotations generated by remote sensing image analysis experts. Comprehensive experiments using CNN- and Transformer-based segmentation models demonstrate that the proposed dataset enables accurate segmentation of diverse vegetation types across complex Earth surfaces. What are the implications of the main findings? MultiVeg enables consistent vegetation monitoring and urban ecological management by providing detailed mapping of heterogeneous vegetation distributions identifiable in very high-resolution satellite imagery. The dataset provides a foundation for developing advanced deep learning models for vegetation analysis and environmental monitoring using very high-resolution satellite imagery. Vegetation segmentation in Very High-Resolution (VHR) satellite imagery has become an essential task for ecological monitoring, supporting diverse applications such as large-scale vegetation conservation and detailed segmentation of small local areas. In particular, multi-class vegetation segmentation, which distinguishes various vegetation types beyond simple binary segmentation of vegetation and non-vegetation, enables detailed analysis of subtle ecosystem changes and has gained increasing importance. However, the annotation of VHR satellite imagery requires extensive time and effort, resulting in a lack of datasets for vegetation segmentation, especially those including multi-class annotations. To address this limitation, this study proposes MultiVeg, a deep learning dataset based on VHR satellite imagery for detailed multi-class vegetation segmentation. MultiVeg includes preprocessed 0.5 m resolution images collected by the KOMPSAT-3 and 3A satellites from 2014 to 2023, covering diverse environments such as urban, agricultural, and forest regions. Each image was carefully annotated by experts into three semantic classes, which are Background, Tree, and Low Vegetation, and validated through a structured quality check process. To verify the effectiveness of MultiVeg, seven representative semantic segmentation models, including convolutional neural network and Transformer-based architectures, were trained and comparatively analyzed. The results demonstrated consistent segmentation performance across all classes, confirming that MultiVeg is a high-quality and reliable dataset for deep learning-based multi-class vegetation segmentation research using VHR satellite imagery. The MultiVeg will be publicly available through GitHub (release v1.0), serving as a valuable resource for advancing deep leaning-based vegetation segmentation research in the remote sensing field.