基于深度學(xué)習(xí)結(jié)合高光譜技術(shù)的大豆種子活力檢測(cè)方法

打開(kāi)文本圖片集

關(guān)鍵字：大豆；種子活力；檢測(cè)；高光譜；深度學(xué)習(xí)；注意力機(jī)制中圖分類(lèi)號(hào)：S127 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1000-4440（2025）05-0927-10

Abstract：To achieve eficient，accurate，and non-destructive identification of soybean seed vigor，this study used seedsofthesoybeanvariety Williams 82asexperimentalmaterials.Alibraryof soybeanseeds withdiferentlevelsof vigor wasconstructedthroughartificialagingtreatments.HyperspectralimagesandRGBimagesof theseedswerethencollected to generate three image datasets （RGB dataset，SIQ dataset，and ENVIdataset）.Four deep learning models （Vg16Net， GoogLeNet，MobileV3Net，and ResNet-34）were employed to detectseed vigor，and theoptimal modelsand datasets were selected.Furthermore，thecoordinate atention（CA）mechanismandlabel smothing loss function were incorporated into theoptimal models toenhancetheirdetectionperformanceandrobustnessTheresultsdemonstratedthatusingtheSIQdata

set and ResNet-34 model，the recognition accuracy reached 97.6% and 96.8% on the training set and validation set，respectively.The detection performance was superiortoothercombinationsofmodelsand datasets.The CA-ResNet-34 model，which incorporated the CA mechanismand label smoothinglossfunction into the ResNet-34 model，achieved a detection accuracy of 98.5% for soy

bean seed vigor based onthe SIQdataset.Thisrepresentedanimprovementof1.7percentage points inaccuracycompared totheoriginal ResNet-34model.Theresultsofthis studycanprovideanew methodfortheaccurate，non-destructive，and efficient detection of soybean seed vigor.

Key words： soybean；seed vigor；detection；hyperspectral；deep learning；atention mechanism

大豆是中國(guó)重要的油料作物和經(jīng)濟(jì)作物，2023年中國(guó)大豆種植面積達(dá) 1.047×107hm2 ，產(chǎn)量約2.084×107 t。（剩余14180字）