High-Throughput Phenotypic Analysis of Cauliflower Growth Dynamics

99.4% mask mAP50 was achieved for segmenting cauliflower plants and curds using the YOLO12s-seg model on RGB images from seedling to harvest. This dataset likely contains automatically extracted traits like plant canopy width, leaf area, and curd diameter, which showed high consistency with manual measurements (R² > 0.90). The data was published by Saichuan Cheng on figshare under a CC-BY-4.0 license.

Use Cases

• Training instance segmentation models for plant and curd detection based on the described RGB image dataset.
• Analyzing growth dynamics of leaf area and curd area using the Richards and Sine models mentioned in the description.
• Classifying cauliflower germplasm into growth types based on high-throughput phenotyping data from 47 varieties.
• Developing automated platforms for crop growth monitoring and phenotypic analysis as referenced in the work.

Strengths

• The model achieved a segmentation mask mAP50 of 99.4% for plants and curds in sparsely planted images.
• Automatically extracted traits like plant canopy width and curd diameter were highly consistent with manual measurements (R² > 0.90).
• The study systematically covers the entire growth period of cauliflower from seedling to harvest.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small in scale at 100.6 KB, indicating limited data volume.

Provenance

Source

figshare

Collection Method

RGB images of cauliflower were collected and processed using instance segmentation models (YOLO12s-seg) for automated trait extraction.

Freshness

Last updated 2026-05-13 05:56:26; freshness should be verified.