# Subalpine Coniferous Forest Vegetation Segmentation: 11,336 DJI Mavic 2 Drone Image Chips with Pixel Masks, Wanglang Nature Reserve, China (AgML Mirror)

## Facts

- Provider: Weibo Shi; Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://data.mendeley.com/public-api/datasets/d9f4m2735b?version=2)
- Size: 9 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/subalpine_forest_segmentation
- Access: public, readable without an account
- DOI: https://doi.org/10.17632/d9f4m2735b.2
- Category: Environmental Science
- Source: https://huggingface.co/datasets/Project-AgML/subalpine_forest_segmentation
- Page: https://scigantic.com/archive/97845091-08a8-5dc1-bbf7-d319a3f2b004
- Keywords: Environmental Science, image segmentation, Parquet, Hugging Face

## About

Drone RGB imagery of a natural subalpine coniferous forest in Wanglang National Nature Reserve, Sichuan, China, captured on 15 June 2020 with a DJI Mavic 2 UAV (1-inch CMOS camera), with pixel-level vegetation class masks for semantic segmentation of forest structure and ground cover. The original release accompanies a convolutional neural network for mapping vegetation classes in inaccessible natural forest from UAV imagery, built with the arcgis.learn module of the ArcGIS API for Python, and holds the drone image plus training and test sample sets. This mirror holds the 11,336 image chips with their masks. Useful for UAV forest mapping, land cover segmentation in mountain ecosystems, and ecological monitoring. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/subalpine_forest_segmentation") or through the AgML Python library (Project-AgML, UC Davis), which reformatted the original release. License: CC BY 4.0, as stated by the original publisher; cite Weibo Shi et al. and the DOI 10.17632/d9f4m2735b.2.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("Project-AgML/subalpine_forest_segmentation")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00007.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/97845091-08a8-5dc1-bbf7-d319a3f2b004
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/97845091-08a8-5dc1-bbf7-d319a3f2b004
- Over MCP: `get_schema_card` with id `97845091-08a8-5dc1-bbf7-d319a3f2b004` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Image segmentation datasets](https://scigantic.com/topics/image-segmentation)
- [Environmental Science datasets](https://scigantic.com/topics/environmental-science)
