Answers are grounded in this paper and cite their sources.
Automated detection of visual indicators associated with key phenological stages can support image-based monitoring of greenhouse cucumber. However, female flowers and young fruits are often small, partially occluded, and visually similar to adjacent developmental indicators, which complicates their detection in greenhouse scenes. In this study, the You Only Look Once (YOLO) version 11 nano (YOLO11n) detector was adapted to develop YOLO11n-Cucumber for detecting seedling plants, female flowers, young fruits, and marketable fruits. The model incorporates a GhostNetV3 backbone enhanced with multi-path dynamic Ghost convolution (MDGC) for feature extraction, a C2PSA module enhanced with Spatial and Channel Synergistic Attention (SCSA) (C2PSA-SCSA) for feature enhancement, and a scale- and localization-quality-reweighted Complete Intersection over Union (CIoU) loss, termed MSF-CIoU, for bounding-box regression. Experiments were conducted on a greenhouse cucumber dataset constructed in this study, comprising 3849 images and 8791 annotated instances. In an internal within-site test evaluation using images collected in the same greenhouse, YOLO11n-Cucumber achieved mean precision, recall, mean average precision (mAP)@0.5, and mAP@0.5:0.95 values of 85.7%, 83.3%, 86.0%, and 60.3%, respectively. Compared with YOLO11n, its mean mAP@0.5 and mAP@0.5:0.95 increased by 2.4 and 4.1 percentage points, respectively. The model required fewer floating-point operations (FLOPs) but had a higher parameter count. These results indicate that YOLO11n-Cucumber provides an improved accuracy–FLOPs trade-off under the greenhouse image-acquisition conditions of this study. Its generalizability across cultivars, greenhouse types, devices, and environmental conditions requires further evaluation.
No takes yet. Share an insight, caveat, or question.
Kang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: