---
license: cc-by-sa-4.0
pretty_name: Binyang UAV Dataset (2017–2019)
tags:
  - agriculture
  - rice
  - uav
  - remote-sensing
  - crop-phenotyping
  - crop-yield
---

# Binyang UAV Dataset (2017–2019)

## Dataset description

The Binyang UAV dataset is, to our knowledge, the first UAV dataset specifically designed for crop growth monitoring/modeling and yield mapping. The Binyang Experiments are a series of agricultural experiments conducted during 2017-2019 at a 160-ha double-cropping paddy rice cultivation zone (23°5′52″–23°7′23″ N, 108°57′7″–108°58′34″ E), located in the Wuhua Irrigation District, Binyang County, Nanning City, Guangxi Zhuang Autonomous Region, China.

The complete repository is approximately **294.86 GB** (274.61 GiB).

## Repository structure

```text
BinyangExperiments/
├── AUXdata/
│   ├── GapFilledWeather_2017-2019.csv
│   ├── gridShp.zip
│   ├── riceMask4studyArea.zip
│   └── riceSegmentationDataset.zip
├── BinyangExperiment_2017/
│   ├── UAV_imagery_2017.zip
│   └── Yield_measurement_2017.zip
├── BinyangExperiment_2018/
│   ├── Crop_trait_measurement_2018.zip
│   ├── UAV_imagery_2018.zip
│   └── Yield_measurement_2018.zip
├── BinyangExperiment_2019/
│   ├── Crop_trait_measurement_2019.zip
│   ├── UAV_imagery_2019.zip
│   └── Yield_measurement_2019.zip
└── README.md
```

## File inventory

| Path | Size | Description |
| --- | ---: | --- |
| `AUXdata/GapFilledWeather_2017-2019.csv` | 43.3 KB | Gap-filled daily weather data for the study period. |
| `AUXdata/gridShp.zip` | 73.9 KB | Spatial grid layer for the study area. |
| `AUXdata/riceMask4studyArea.zip` | 43.3 MB | Rice mask for the study area. |
| `AUXdata/riceSegmentationDataset.zip` | 700.4 MB | Dataset prepared for rice segmentation. |
| `BinyangExperiment_2017/UAV_imagery_2017.zip` | 32.45 GB | UAV imagery collected during the 2017 experiment. |
| `BinyangExperiment_2017/Yield_measurement_2017.zip` | 117.4 KB | 2017 yield measurements in ESRI Shapefile format. |
| `BinyangExperiment_2018/Crop_trait_measurement_2018.zip` | 25.4 KB | 2018 biomass, phenology, plant area index, and plot-boundary data. |
| `BinyangExperiment_2018/UAV_imagery_2018.zip` | 205.58 GB | UAV imagery collected during the 2018 experiment. |
| `BinyangExperiment_2018/Yield_measurement_2018.zip` | 170.8 KB | 2018 yield measurements in ESRI Shapefile format. |
| `BinyangExperiment_2019/Crop_trait_measurement_2019.zip` | 23.2 KB | 2019 biomass, phenology, plant area index, and plot-boundary data. |
| `BinyangExperiment_2019/UAV_imagery_2019.zip` | 56.09 GB | UAV imagery collected during the 2019 experiment. |
| `BinyangExperiment_2019/Yield_measurement_2019.zip` | 11.8 KB | 2019 yield measurements in ESRI Shapefile format. |


## Auxiliary data

### Weather data

`AUXdata/GapFilledWeather_2017-2019.csv` contains the following columns:

| Column | Description |
| --- | --- |
| `Year`, `Month`, `Day` | Observation date. |
| `Tmax`, `Tmin` | Daily maximum and minimum air temperature in °C. |
| `radiation MJ/m2.day` | Daily solar radiation in MJ m⁻² day⁻¹. |
| `wind m/s` | Wind speed in m s⁻¹. |
| `Rain Fall` | Daily rainfall in mm. |
| `MeanRH` | Mean relative humidity in %. |

### Spatial and segmentation data

- `gridShp.zip` contains the study-area grid layer.
- `riceMask4studyArea.zip` contains the rice mask used for the study area.
- `riceSegmentationDataset.zip` contains data prepared for rice-segmentation applications.

## Crop-trait measurements

The listed data-collection dates are aligned with the corresponding UAV observation dates. UAV campaigns and crop-trait measurements were coordinated within the same observation window; however, because manual trait measurements were labor-intensive, completing one field measurement campaign could take 2–3 days.

The 2018 and 2019 crop-trait archives use the following internal structure:

```text
Crop_trait_measurement_<year>/
├── AboveGroundBiomass/
│   ├── Biomass_<year>_Controlled_plots.csv
│   └── Biomass_<year>_Farmers_plots.csv
├── Phenology/
│   ├── BBCH_<year>_Controlled_plots.csv
│   └── BBCH_<year>_Farmers_plots.csv
├── PlantAreaIndex/
│   ├── PAI_<year>_Controlled_plots.csv
│   └── PAI_<year>_Farmers_plots.csv
└── plotShp/
    ├── controlledPlots_<year>.*
    └── farmersPlots_<year>.*
```

Each measurement CSV is organized by `Year` and `Plot_ID`, followed by date-specific measurement columns:

- `AboveGroundBiomass` contains repeated above-ground biomass measurements in g m⁻².
- `Phenology` contains crop-development observations expressed as unitless BBCH stages.
- `PlantAreaIndex` contains plant area index (PAI) measurements in m² m⁻² (dimensionless).
- `plotShp` contains the corresponding controlled-plot and farmer-plot boundaries as ESRI Shapefiles.

The measurement units, missing-value conventions, and field definitions should be checked in the source files before analysis.

## Yield measurements

Each `Yield_measurement_<year>.zip` archive contains yield measurements in t ha⁻¹ as an ESRI Shapefile dataset, including the `.shp`, `.shx`, `.dbf`, `.prj`, and related sidecar files. Keep all components together when opening the data in GIS software.

## License

This dataset is released under the [Creative Commons Attribution-ShareAlike 4.0 International License](https://creativecommons.org/licenses/by-sa/4.0/) (`CC BY-SA 4.0`).

Users must provide appropriate attribution and distribute adapted material under the same license.

## Citation

Preprint:
> Yang, Q., Han, J., Chen, Z., Yu, J., Zha, Y., & Shi, L. Integrating computer vision, process-based model and UAV data for rice yield mapping: A hybrid model-data fusion framework with a benchmark dataset. Available at SSRN 6336282.