CVRP: Multi-Cultivar Rice Plant Images with Annotated Masks

CVRPDataset created a collection of 2,303 field images of rice plants, captured from 231 landraces and 50 modern cultivars. Each image is paired with an annotated mask. The dataset also includes 123 indoor images focused on individual panicles and 3D target reconstruction files, with the last update recorded on 2025-02-28.

Use Cases

• Train image segmentation models based on the provided annotated masks.
• Develop cultivar classification models based on the multi-cultivar field images.
• Perform 3D reconstruction of plant structures based on the provided 3D target files.
• Analyze panicle morphology and development based on the focused indoor images.

Strengths

• Contains 2,303 field images, providing a substantial base for model training.
• Covers a diverse genetic pool with images from 231 landraces and 50 modern cultivars.
• Includes paired annotated masks for all field images, enabling supervised learning tasks.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for specific subsets (e.g., indoor images vs. field images) is unknown, which may limit suitability assessment.
• Geographic and temporal coverage for the image collection is not specified, which may introduce bias.

Provenance

Source

CVRPDataset

Collection Method

Field and indoor image capture, with manual or algorithmic annotation.

Time Range

null

Freshness

Last updated 2025-02-28 06:54:43; freshness should be verified.

Geography

null