Cauliflower Plant and Curd Growth Traits from High-Throughput Image Analysis

Saichuan Cheng's dataset contains phenotypic traits automatically extracted from RGB images of cauliflower plants from seedling to harvest. The data includes plant canopy width, leaf area, and curd traits derived from instance segmentation models, with high consistency to manual measurements (R² > 0.90). The dataset was last updated on 2026-05-13.

Use Cases

• Training instance segmentation models for plant and curd detection based on the described image collection.
• Analyzing growth dynamics of leaf area and curd area using the fitted Richards and Sine models.
• Classifying curd growth types (mature and compact, peak-burst, steady-increase) based on extracted kinetic parameters.
• Clustering cauliflower germplasms into groups (comprehensively coordinated, mid-maturity compact, etc.) using high-throughput phenotyping data.

Strengths

• Segmentation model achieved a high mask mAP50 of 99.4% for plants and curds in sparse images.
• Automatically extracted traits showed high consistency with manual measurements (R² > 0.90).
• Analysis covers the entire growth period from seedling to harvest, which the description notes is often lacking.

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 file size (20.1 KB), suggesting limited scope or summary-level data.

Provenance

Source

figshare

Collection Method

Traits were automatically extracted from RGB images using instance segmentation models (YOLO12s-seg).

Freshness

Last updated 2026-05-13 05:56:35; freshness should be verified.