Tatsoi Image-Environment Dataset for Growth Monitoring and Intelligent Phytoprotection Benchmarking: Part 4
Published: 30 July 2026| Version 1 | DOI: 10.17632/286gv4tmrh.1
Contributor:
xing yangDescription
This dataset contains 3456 tatsoi plant level RGB images, collected from 128 plants that were imaged repeatedly on 27 winter cultivation dates from November 12, 2025 to February 28, 2026. Each image is associated with canopy polygon annotation, derived detection boxes and segmentation masks, plant level visible state labels, standardized random partitioning, time partitioning, and plant reserve partitioning, as well as synchronized field environment and pest signal metadata. This resource supports reproducible benchmark tasks such as canopy detection, canopy segmentation, temporal growth analysis, image environment phenotype recognition, and field background modeling.
Files
Institutions
- Anhui Science and Technology UniversityAnhui, Chuzhou
Categories
Agricultural Plant, Data Collection in Agriculture