![[2602.18585] BloomNet: Exploring Single vs. Multiple Object Annotation for Flower Recognition Using YOLO Variants](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.18585] BloomNet: Exploring Single vs. Multiple Object Annotation for Flower Recognition Using YOLO Variants
Summary
The paper explores the effectiveness of single versus multiple object annotation for flower recognition using various YOLO models, presenting a new dataset and benchmarking results.
Why It Matters
This research is significant for advancements in automated agriculture, particularly in improving flower detection methods that can enhance crop monitoring and yield estimation. The findings contribute to the understanding of how annotation techniques and model architectures impact detection performance in different scenarios.
Key Takeaways
- Introduces the FloralSix dataset for flower recognition.
- Benchmarks YOLO models under single and multiple object annotation regimes.
- YOLOv8m shows superior performance in sparse scenarios, while YOLOv12n excels in dense environments.
- The choice of optimizer (SGD) consistently yields better results across models.
- Findings support applications in non-destructive crop analysis and robotic pollination.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.18585 (cs) [Submitted on 20 Feb 2026] Title:BloomNet: Exploring Single vs. Multiple Object Annotation for Flower Recognition Using YOLO Variants Authors:Safwat Nusrat, Prithwiraj Bhattacharjee View a PDF of the paper titled BloomNet: Exploring Single vs. Multiple Object Annotation for Flower Recognition Using YOLO Variants, by Safwat Nusrat and 1 other authors View PDF HTML (experimental) Abstract:Precise localization and recognition of flowers are crucial for advancing automated agriculture, particularly in plant phenotyping, crop estimation, and yield monitoring. This paper benchmarks several YOLO architectures such as YOLOv5s, YOLOv8n/s/m, and YOLOv12n for flower object detection under two annotation regimes: single-image single-bounding box (SISBB) and single-image multiple-bounding box (SIMBB). The FloralSix dataset, comprising 2,816 high-resolution photos of six different flower species, is also introduced. It is annotated for both dense (clustered) and sparse (isolated) scenarios. The models were evaluated using Precision, Recall, and Mean Average Precision (mAP) at IoU thresholds of 0.5 (mAP@0.5) and 0.5-0.95 (mAP@0.5:0.95). In SISBB, YOLOv8m (SGD) achieved the best results with Precision 0.956, Recall 0.951, mAP@0.5 0.978, and mAP@0.5:0.95 0.865, illustrating strong accuracy in detecting isolated flowers. With mAP@0.5 0.934 and mAP@0.5:0.95 0.752, YOLOv12n (SGD) outperformed the more complicated...