Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding

Cauliflower Center Dataset The dataset is used for autonomous intrarow weeding as described in the paper of Willekens et al. (2025). Willekens, A., Callens, B., Pieters, J., Wyffels, F., & Cool, S. (2025). Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding. Precision Agriculture. Springer. Center The cauliflower centers in the images (PNG files) are provided in JSON files with the same name as the corresponding image. Each JSON file follows this structure: { "centers": [ [ , ], [ , ], ... ] } The .pkl files contain the same information as the JSON files but in a binary format. Segmentation Many of the images are also segmented. The segmentation masks are stored in files with the .sam.png extension. Dataset Overview The dataset file db_dump.sql provides the content of a PostgreSQL database that was used in the study of Willekens et al. (2025). When recovering this dataset, the PostGIS extension must be installed. Create a database in PostgreSQL CREATE DATABASE cauliflower_db; Connect to the new database and install the PostGIS plugin CREATE EXTENSION postgis;. See the PostGIS installation instructions if PostGIS is not yet installed. Recover the database psql -U your_username -d cauliflower_db -f db_dump.sql The tables contain the following information. dataset The datasets correspond to the different folders (e.g., 20230809.zip, 20230810.zip, 20230811.zip), with the folder name representing the date of data collection (YYYYMMDD). calibration Refers to the calibration files stored in calib.zip. image Holds metadata for individual images, such as geometry, time, and the associated dataset. label Stores the references of the labels for each image. label_set Bundles different labels into one label_set used for training, validation, or testing. label_set_entry Serves as a junction table associating labels with label sets. sensor Lists the sensors used during data collection. training Describes the training sessions that were initiated. It refers to a validation and training label_set and specifies the training hyperparameters.