基于改进YOLOv8s-Seg模型的番茄成熟度检测

打开文本图片集

关键词：改进YOLOv8s-Seg模型；番茄（SolanumlycopersicumL.）;成熟度；检测

中图分类号：S126；TP391.41 文献标识码：A

文章编号：0439-8114（2025）06-0178-07

DOI：10.14088/j.cnki.issn0439-8114.2025.06.030 开放科学（资源服务）标识码（OSID）：

Maturity detection of tomato based on the improved YOLOv8s-Seg model

YANG Shuang，ZHOU Zhong-lin （Economics and Management School，Yangtze University，Jingzhou 434023，Hubei， China）

Abstract：Toachievereal-timedetectionof tomatoatdiferentmaturitystages，animprovedYOLOv8s-Seg model wasproposed to meetthe precisionmanagementrequirementsof modernagriculture.Byenhancing theneck moduleof theimprovedYOLOv8s-Seg model，a1×1SimConvlayerwasaddedbeforeeachupsamplingoperation，andtheremainingconventionalconvolutionsintheneck werereplacedwith 3×3 SimConvlayers，significantlyimprovingfeaturefusioncapability.Theresultsshowed thattheimprovedYOLOv8s-Seg model achieved segmentation precision rates of 92.7% ， 92.3% ，and 89.9% for mature，semi-mature，and immature tomatoes，respectively.Compared withtheoriginalYOLOv8s-Seg model，theimprovedYOLOv8s-Segmodeldemonstrated increasesof 1.6，0.4，1.0，and 2.4 percentage points in precision，recall，F1-score，and ，respectively. The improved YOLOv8s-Seg model outperformed YOLOv8s-Seg，YOLOv5s-Seg，YOLOv7-Seg，and Mask R-CNN models in precision，recal，F1-score，and mAP@0.5.Theinference timeoftheimprovedYOLOv8s-Segmodelwas3.5ms，showingaslightincreasecomparedtoYOLOv5s-Seg andYOLOv8s-Seg models，butremainedsignificantlylowerthanYOLOv7-Segand Mask R-CNNmodels.Theimproved YOLOv8sSegmodelexhibitedsuperiorperformanceintomatomaturitysegmentationundercomplexenvironments，achievinghigh precision across scenarios involving leaf occlusion，fruit overlap，lighting variations，and viewpoint changes.

Key words： improved YOLOv8s-Seg model; tomato（Solanum lycopersicum L.）；maturity；detection

在复杂背景下准确分割不同成熟度的番茄对于有效采摘、水果监测及评估番茄质量至关重要。（剩余9945字）