Light quality plays a decisive role in controlled-environment agriculture, shaping plant morphology, physiology, and productivity. This study investigated the impact of far-red (FR) light on Cannabis sativa L. by comparing two different application strategies: continuous FR supplementation throughout 12 h of the photoperiod and end-of-day (EOD) FR exposure applied only at the end of the light period. In both treatments, FR was added to a background spectrum of red and blue (RB) light, while a control group grown under RB light alone was included to assess the specific effects of FR on plant growth, physiological responses, and flowering. Continuous FR exposure induced pronounced shade-avoidance traits, increasing plant height by 9% and petiole length by 17% relative to the control, and raised leaf dry weight to 12.9 g, 9% higher than under EOD (11.7 g) and 16.3% higher than under RB alone (10.8 g). Besides plant height and petiole length, both FR and EOD treatment induced limited morphological adjustments but increased chlorophyll content by 9%, resulting in greater canopy expansion and photosynthetic potential. However, flowering time was unaffected by spectral treatment, confirming that Cannabis floral induction is tightly regulated by photoperiod rather than light quality. Energy-use analysis revealed that EOD supplementation achieved many of the benefits of continuous FR while reducing overall consumption, but energy-use efficiency analysis proved FR as the more efficient treatment. These findings highlight the potential of FR light, particularly when applied continuously, to optimize vegetative growth and canopy physiology in controlled-environment Cannabis cultivation, while EOD strategies offer a practical compromise between cost savings and physiological benefits.

Automated cucumber harvesting requires a perception system that can detect fruit, assess ripeness and localize the peduncle for cutting under real commercial greenhouse conditions. This study proposes an RGB-D perception framework combining YOLOv8n object detection with depth-based geometric estimation. The framework was evaluated on data collected in commercial greenhouses with dense, continuously varying occlusion, variable illumination and multiple co-visible fruits at different maturity stages. Detections were assigned to five operational classes: ripe, unripe, flagged (non-marketable due to overripeness or deformity), obstructed and peduncle. The detector achieved an overall mAP50 of 0.568, but appearance-based ripeness classification proved unreliable. Confusion concentrated along two boundaries: a continuous, judgment-based occlusion criterion and an absolute length threshold that RGB appearance cannot directly resolve. Depth-based length estimation, by contrast, achieved an MAE of 0.72 cm (RMSE = 1.15 cm, MAPE = 4.04%) directly in the greenhouse. Applying the ripeness length threshold to the estimated lengths misclassified 6.4% of fruits at the RIPE/UNRIPE boundary on the dimensional validation set, indicating that length-based ripeness assignment is feasible. Diameter estimation was markedly less accurate (R² = 0.49) and is not proposed as a validated measurement for grading or yield estimation. The framework also estimated candidate 3D cutting points that remained spatially consistent across frames; cutting accuracy requires future robotic validation. These results suggest that, in complex greenhouse environments, ripeness decisions of cucumber should rely on depth-based geometric measurement, with object detection used primarily for fruit and peduncle localization. Combining visual detection with quantitative measurement therefore appears to be a promising basis for robotic cucumber harvesting.