利用经自动分级机验证的改进型YOLOv8s模型增强对鲜切玫瑰开放度的预测。

Enhancement of the prediction of the openness of fresh-cut roses with an improved YOLOv8s model validated by an automatic Grading Machine.

作者信息

Lai Qinghui, Yang Zhanwei, Su Wei, Yan Chuang, Zhao Qinghui, Tan Yu, Que Yu, Zheng Jing

机构信息

School of Energy and Environmental Science, Yunnan Normal University, Kunming, China.

Faculty of Modern Agricultural Engineering, Kunming University of Science and Technology, Kunming, China.

出版信息

Front Plant Sci. 2025 Mar 25;16:1546503. doi: 10.3389/fpls.2025.1546503. eCollection 2025.

DOI:10.3389/fpls.2025.1546503

PMID:40201784

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11975861/

Abstract

INTRODUCTION

The openness grading of fresh-cut roses relies heavily on manual work, which can be inefficient and inconsistent.

METHODS

In this study, an improved YOLOv8s model is proposed for openness grading in conjunction with a newly developed automatic grading machine for fresh-cut roses. The model identifies unopened inner petals and classifies openness into five levels: degree 1, degree 2, degree 3, degree 4, and deformity. To enhance detection accuracy while reducing the model complexity and computation, the backbone network of YOLOv8s is replaced by MobileNetV3. Additionally, an Efficient Multi-scale Attention (EMA) module is introduced to enhance focus on critical features, and a Wise-IoU loss function is incorporated to accelerate convergence.

RESULTS

Field experiments revealed that the openness predictions made by the automatic fresh-cut roses grader had errors of 6.9%, 9.1%, 10.0%, 6.5%, and 12.6%, respectively, compared to manual predictions.

DISCUSSION

Therefore, the improved YOLOv8s-F model effectively meets the requirements of fresh-cut rose openness grading.

摘要

引言

鲜切玫瑰的开放度分级严重依赖人工操作，效率低下且标准不一。

方法

在本研究中，提出了一种改进的YOLOv8s模型用于开放度分级，并结合新开发的鲜切玫瑰自动分级机。该模型可识别未开放的内花瓣，并将开放度分为五个等级：1级、2级、3级、4级和畸形。为提高检测精度同时降低模型复杂度和计算量，将YOLOv8s的骨干网络替换为MobileNetV3。此外，引入高效多尺度注意力（EMA）模块以增强对关键特征的关注，并采用Wise-IoU损失函数加速收敛。