EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A task-specific architecture with multi-scale attention and shape-aware loss for strawberry phenophase recognition in complex fields

Shilin Li, Shangjian Guo, Nan Yang, Lili Sun, Shujuan Zhang, Fuzhong Li

Frontiers in Plant Science · 22.06.2026 · 17

DOI: 10.3389/fpls.2026.1888504 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Plant Science, online erschienen am 22.06.2026. Verfasst von Shilin Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Software Technologies, Shanxi Agricultural University, Jinzhong, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanxi Agricultural University, Department of Human Resources and Social Security of Shanxi Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the challenges of recognizing small strawberry targets and achieving accurate phenological perception in complex field environments, this paper proposes a novel end-to-end lightweight detection architecture named HCMS-Net. The backbone is a Residual Efficient Layer Aggregation Network (R-ELAN) enhanced with a Multi-Scale Convolutional Attention (MSCA) mechanism, which emphasizes subtle color and texture variations to differentiate key phenological phases. For feature fusion, hypergraph convolution (from HyperC2Net) and a Mixed Aggregation Network (MANet) are incorporated, modeling the clustered morphology of strawberries and strengthening the representation of sparse small fruits. The detection head incorporates a lightweight Conv2Former module to capture long-range dependencies and spatial contextual information across growth stages, thereby enhancing the model’s capacity to represent continuous phenological changes. A Shape-Normalized Wasserstein Distance (Shape-NWD) loss is introduced to stabilize optimization against minor pixel deviations. Experimental results demonstrated that HCMS-Net achieved a mean average precision (mAP) of 94.9% and an F1-score of 90.0%. Specifically, the average precision (AP) values for the flowering, young fruit, green fruit, veraison, and mature fruit stages reached 99.3%, 88.3%, 90.9%, 97.0%, and 98.2%, respectively. Heatmaps confirmed HCMS-Net’s precise attention focus across all five phenological stages, effectively suppressing irrelevant backgrounds. Compared to ten mainstream detectors, HCMS-Net surpassed alternatives such as RT-DETR and the YOLOv5n to v13n by 3.4–8.0 percentage points in mAP. It even surpassed YOLOv12s by 2.7 percentage points, while containing only 32.86% of its parameters. The model offers high accuracy and efficiency for phenological period detection, supporting selective harvesting and intelligent agricultural management.

Schlagwörter

Stichwörter der Autor:innen

agricultural roboticsobject detectionphenological period detectionprecision agricultureselective picking

Autor:innen und Institutionen

  • Faculty of Software Technologies, Shanxi Agricultural University, Jinzhong, China.
  • College of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, China.

Förderung

  • Shanxi Agricultural University
  • Department of Human Resources and Social Security of Shanxi Province

Zitieren

APA

Li, S., Guo, S., Yang, N., Sun, L., Zhang, S., & Li, F. (2026). A task-specific architecture with multi-scale attention and shape-aware loss for strawberry phenophase recognition in complex fields. Frontiers in Plant Science, 17. https://doi.org/10.3389/fpls.2026.1888504

Vancouver

Li S, Guo S, Yang N, Sun L, Zhang S, Li F. A task-specific architecture with multi-scale attention and shape-aware loss for strawberry phenophase recognition in complex fields. Frontiers in Plant Science. 2026;17. doi:10.3389/fpls.2026.1888504

BibTeX
@article{Li2026,
  author = {Li, Shilin and Guo, Shangjian and Yang, Nan and Sun, Lili and Zhang, Shujuan and Li, Fuzhong},
  title = {A task-specific architecture with multi-scale attention and shape-aware loss for strawberry phenophase recognition in complex fields},
  journal = {Frontiers in Plant Science},
  year = {2026},
  volume = {17},
  doi = {10.3389/fpls.2026.1888504}
}
Kostenlos starten →