Precision Greenhouse Rose Phenotyping from UAV Imagery Using a Multi-Source Dataset and Lightweight BloomRoseNet
Remote Sensing · 11.08.2026 · 18 (16) · 2704
DOI: 10.3390/rs18162704 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 11.08.2026. Verfasst von Yingchao Wang und 9 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 78 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Japan Science and Technology Agency’s SPRING Program, Industry-Education Integration and New Business Development Research Center Project, Southern Xinjiang Joint Youth Fund of Xinjiang Uygur Autonomous Region Natural Science Foundation.
Abstract (Originalsprache)
Accurate detection of blooming roses and flower buds is essential for greenhouse phenotyping, cultivation scheduling, harvest planning, and yield management. However, UAV-derived greenhouse imagery presents major challenges because rose targets are often small, densely distributed, partially occluded, and visually similar to complex backgrounds. This study proposes a lightweight rose detection framework that combines multi-source dataset construction with an improved YOLOv12n-based detector, termed BloomRoseNet. A GreenHouse Rose dataset was constructed by integrating self-collected UAV overhead images, screened RoseTracker images, and supplementary multi-view rose images to increase diversity in scale, growth stage, viewpoint, and background complexity. BloomRoseNet introduces task-oriented improvements for fine-grained feature extraction, adaptive feature fusion, and attention-enhanced detection. The supplementary multi-view data improved precision, recall, and mAP@50 from 85.2%, 82.8%, and 89.2% to 86.1%, 85.3%, and 90.5%, respectively. Compared with the baseline YOLOv12n, BloomRoseNet increased precision, recall, mAP@50, and mAP@50:95 by 3.2, 3.3, 3.6, and 1.6 percentage points, respectively, while reducing parameters from 2.55 M to 2.08 M and model size from 5.5 MB to 4.5 MB. The model also maintained real-time inference capability and stronger robustness under blur, occlusion, and illumination disturbances. The proposed framework provides an effective and practical solution for UAV-based greenhouse rose monitoring and supports precision cultivation management.
Förderung
- The Japan Science and Technology Agency’s SPRING Program
- Industry-Education Integration and New Business Development Research Center Project
- Southern Xinjiang Joint Youth Fund of Xinjiang Uygur Autonomous Region Natural Science Foundation
- Xinjiang University of Science and Technology
Zitieren
APA
Wang, Y., Hao, J., Zhou, P., Chen, W., Sun, S., Li, N., Xue, F., Qin, Z., Wu, H., & Zhao, F. (2026). Precision Greenhouse Rose Phenotyping from UAV Imagery Using a Multi-Source Dataset and Lightweight BloomRoseNet. Remote Sensing, 18(16), 2704. https://doi.org/10.3390/rs18162704
Vancouver
Wang Y, Hao J, Zhou P, Chen W, Sun S, Li N, et al. Precision Greenhouse Rose Phenotyping from UAV Imagery Using a Multi-Source Dataset and Lightweight BloomRoseNet. Remote Sensing. 2026;18(16):2704. doi:10.3390/rs18162704
BibTeX
@article{Wang2026,
author = {Wang, Yingchao and Hao, Jun and Zhou, Peng and Chen, Wei and Sun, Shan and Li, Na and Xue, Feng and Qin, Zixiang and Wu, Hao and Zhao, Fan},
title = {Precision Greenhouse Rose Phenotyping from UAV Imagery Using a Multi-Source Dataset and Lightweight BloomRoseNet},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {16},
pages = {2704},
doi = {10.3390/rs18162704}
}
EUVIMED