基于遷移學(xué)習(xí)的MobileViT-CBAM鮮煙葉成熟度識(shí)別模型研究
MobileViT-CBAM Model for Fresh Tobacco Leaf Maturity Recognition Base on Transfer Learning
-
摘要: 為建立更加經(jīng)濟(jì)高效的煙葉成熟度無損智能識(shí)別技術(shù)本研究構(gòu)建了一種可以部署到移動(dòng)設(shè)備的輕量級(jí)網(wǎng)絡(luò)模型MobileViT-CBAM,。首先采集云煙87中,、上部不同成熟度煙葉圖像構(gòu)建數(shù)據(jù)集,，將CBAM注意力機(jī)制模塊引入到MobileViT結(jié)構(gòu)中增強(qiáng)鮮煙葉成熟度圖像特征的表達(dá)能力,，其次將原有激活函數(shù)Swish函數(shù)替換為更平滑的SMU函數(shù)，幫助模型更快收斂,，最后使用遷移學(xué)習(xí)提高模型的訓(xùn)練效率和泛化能力,，以實(shí)現(xiàn)復(fù)雜大田環(huán)境下鮮煙葉成熟度分類。結(jié)果表明,，MobileViT-CBAM在鮮煙葉成熟度分類任務(wù)中準(zhǔn)確率達(dá)92.81%,，較VGG16、ResNet34,、Vision Transformer,、Swin Transformer、MobileNetV2 和MobileViT等模型具有顯著的性能提升,。所提出的MobileViT-CBAM模型能夠有效識(shí)別煙葉成熟程度,，為煙葉智能采集裝備視覺系統(tǒng)提供技術(shù)支撐。Abstract: To establish more economical and efficient non-destructive intelligent recognition technology for tobacco leaf maturity, a lightweight network model MobileViT-CBAM on mobile devices was constructed. Firstly, a dataset was built by collecting images of the middle and upper leaves of ‘Yunyan87’ with different maturity. The CBAM attention mechanism module was introduced into the MobileViT structure to enhance the feature expression ability of fresh tobacco leaf maturity images. Secondly, the original activation function Swish was replaced with the smoother SMU function to help the model converge faster. Finally, transfer learning was employed to improve the training efficiency and generalization ability of the model and achieve the classification of fresh tobacco leaf maturity in complex field environment. Results showed that MobileViT-CBAM exhibited an accuracy of 92.81% in maturity classification of fresh tobacco leaves, which is significantly superior to the models of VGG16, ResNet34, Vision Transformer, Swin Transformer, MobileNetV2, and MobileViT. The proposed MobileViT-CBAM model can effectively identify the maturity degree of tobacco leaves, providing technical support for the visual system of intelligent tobacco harvesting equipment.
下載: