Multimodal Aquaponic Dataset for Salicornia: RGB Imaging, Morphological traits, Biomass, and Water‑Quality Telemetry
Ashraf Sharifi, Mehran Tarif, Sara Migliorini, Davide Quaglia, Roberto Pastres
This dataset contains a multimodal collection of RGB images, morphological traits, water‑quality telemetry, and biomass measurements for Salicornia spp. grown in a pilot‑scale recirculating aquaponic system at Ca’ Foscari University of Venice (BeBlue project). The imaging infrastructure consisted of five ESP32‑CAM modules capturing plant images every 15 minutes from May to October 2025, producing over 38,000 quality‑filtered frames. Plant instance segmentation was performed using PixelLib’s Mask R‑CNN (COCO pretrained model), followed by chromatic refinement and PlantCV‑based morphological feature extraction. All pixel‑based traits were converted to metric units using depth‑corrected calibration patches. Water‑quality parameters (pH, dissolved oxygen, temperature, salinity, and oxygen saturation) were recorded at 5–15 minute intervals at both the tank inlet (SxLoad) and outlet (SxWaste). Environmental data were aligned to image timestamps using a directional nearest‑neighbor strategy, and outliers were corrected using strict physical limits and time‑based interpolation. Manual biomass measurements were collected weekly and interpolated to daily resolution. The final dataset provides daily synchronized records of morphological traits, environmental conditions, and biomass, enabling research in plant phenotyping, aquaponics optimization, environmental modeling, and digital‑twin development.