基于改进MobileNetV2的轻量化茶叶病害检测方法

打开文本图片集

中图分类号：S435.711 文献标志码：A 文章编号：1001-411X（2025）06-0801-09

Abstract： 【Objective】 To address the challenge in tea leaf disease detection， where existing deep learning models struggle to balance accuracy and eficiency，particularly unsuitable for deployment on resourceconstrained embedded devices. 【Method】 A lightweight and high-accuracy recognition model ofMobileNetV2- GCA-LS was proposed based on the MobileNetV2 architecture， incorporating two key innovations. First， a novel ghost coordinate atention （GCA） module was designed， which integrated the positional sensitivity of coordinate atention with thecomputational eficiency of GhostNet，thereby enhanced the feature representation of critical disease areas. Second， the label smoothing （LS）regularization strategy was employed to optimize the training process and improve the model generalization capability.The model was trained and validated on a publicly available tea leaf disease dataset. 【Result】 The proposed MobileNetV2-GCA-LS model achieved a recognition accuracy of 94.54% and F1 of 94.29% on the test set， significantly outperforming comparison models including MobileNetV2， MobileNetV3-Small， EficientNet-B0，ResNet50 and GhostNet. Meanwhile， the model maintained low complexity with 2.608 9×106 parameters and 0.3347×1010 floating point operations （FLOPs）of computational cost， demonstrating its feasibility for deployment on resource-constrained devices. 【Conclusion】 The proposed model effectively enhances the performance of tea leaf disease recognition， achieving a well-balanced trade-offbetween accuracyand efciency.This provides a practical technical solution for intelligent monitoring and precise control of plant diseases in smart agriculture.

Key Words： Tea leaf; Disease and pest recognition; MobileNetV2; Ghost coordinate attention （GCA）; Label smoothing; Smart agriculture

茶树作为全球范围内重要的经济作物，其叶片的健康状况直接决定了茶叶的最终产量与品质，对全球茶叶产业的经济效益与可持续发展起着决定性作用。（剩余13108字）