PC4C_CAPSI: image data of capsicum plant growth in protected horticulture

Namal Jayasuriya, Oula Ghannoum, Wen Hu, Norbert Klause, Weiguang Liang, Yi Guo

Research output: Contribution to journalArticlepeer-review

10 Downloads (Pure)

Abstract

Feeding the increasing global population and reducing the carbon footprint of agricultural activities are two critical challenges of our century. Growing crops under protected horticulture and precise crop monitoring have emerged to address these challenges. Crop monitoring in commercial protected facilities remains mostly manual and labour intensive. Using computer vision to solve specific problems in image-based crop monitoring in these compact and complex growth environments is currently hindered by the scarcity of available data. We collected an RGBD dataset for vertically supported, hydroponically-grown capsicum plants in a commercial-scale glasshouse facility to fill this gap. Data were collected weekly using a single top-angled stereo camera mounted on a mobile platform running between the hydroponic gutters. The RGBD streams covered 80 % of the crop growing season in three different light conditions. The metadata include camera configurations and light condition information. Manually measured plant heights of ten selected plants per gutter are provided as ground truth. The images covered the whole plants and focused on the top third. This dataset will support research on plant height estimation, plant organ identification, object segmentation, organ measurements, 3D reconstruction, 3D data processing, and depth noise reduction. The usability of the dataset has been successfully demonstrated in a previously published study on plant height estimation using machine learning and 3D point cloud.

Original languageEnglish
Article number110735
Number of pages15
JournalData in Brief
Volume55
DOIs
Publication statusPublished - Aug 2024

Bibliographical note

Publisher Copyright:
© 2024 The Author(s)

Keywords

  • Capsicum Annum
  • Glasshouse
  • Growth cycle
  • RGBD
  • Single camera
  • Time series

Fingerprint

Dive into the research topics of 'PC4C_CAPSI: image data of capsicum plant growth in protected horticulture'. Together they form a unique fingerprint.

Cite this