[1]黄带娣,杨红云,孙爱珍,等.基于SwinT与ResNet34多尺度特征融合的水稻磷素营养诊断[J].江苏农业科学,2026,54(4):115-122.
 Huang Daidi,et al.Diagnosis of phosphorus nutrition in rice based on multiscale feature fusion of SwinT and ResNet34[J].Jiangsu Agricultural Sciences,2026,54(4):115-122.
点击复制

基于SwinT与ResNet34多尺度特征融合的水稻磷素营养诊断()

《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]

卷:
第54卷
期数:
2026年第4期
页码:
115-122
栏目:
营养成分智能检测
出版日期:
2026-02-20

文章信息/Info

Title:
Diagnosis of phosphorus nutrition in rice based on multiscale feature fusion of SwinT and ResNet34
作者:
黄带娣1; 杨红云1; 孙爱珍2; 周雅雯1; 刘磊锟1
1.江西农业大学软件学院,江西南昌 330045; 2.江西农业大学计算机与信息工程学院,江西南昌 330045
Author(s):
Huang Daidi; et al
关键词:
水稻; 磷素; 营养诊断; ResNet34; Swin-T; 特征融合; SE模块
Keywords:
-
分类号:
S126;TP391.41
DOI:
-
文献标志码:
A
摘要:
为实现水稻磷素营养状况的快速、精准诊断,提升科学施肥水平与水稻高效栽培管理能力,提出一种基于Swin Transformer(Swin-T)与ResNet34多尺度特征融合的水稻磷素营养诊断方法。该方法首先采用ResNet34网络提取水稻叶片图像的边缘、纹理与颜色等局部细节特征,利用特征金字塔网络(FPN)实现多尺度特征融合,并引入通道注意力机制(SE模块)强化对关键通道信息的表达能力。最终将融合增强后的特征图输入至Swin-T模块,通过滑动窗口自注意力机制建模全局依赖信息,提升模型对磷素缺失程度的综合判别能力。以晚稻黄华占为试验材料,采集不同磷素梯度条件下的叶片图像构建数据集进行试验验证。结果表明,改进后的Swin-T网络模型的准确率、精确率、召回率以及F1分数分别为95.74%、95.75%、95.74%、95.74%,明显优于AlexNet、MobileNet、ResNet34、ConvNeXt、ShuffleNet、Vision Transformer等主流模型,且模型体积仅为28.94 MB,训练过程更平稳、收敛速度更快。综上,改进后的Swin-T网络模型在水稻磷素营养诊断中表现出较高的准确性与鲁棒性,可为作物营养状态识别与智能施肥提供技术支持,具有广阔的应用前景。
Abstract:
-

参考文献/References:

[1]林瑞余,梁义元,蔡碧琼,等. 不同品种水稻产量形成过程的养分积累与分配特征研究[J]. 中国生态农业学报(中英文),2007,15(5):139-146.
[2]Liu C,Chu H L,Wu L F,et al. Regulation mechanism of phosphate homeostasis in plants[J]. Biotechnology Bulletin,2021,37(8):1-12.
[3]李婷婷,薛璟祺,王顺利,等. 植物非结构性碳水化合物代谢及体内转运研究进展[J]. 植物生理学报,2018,54(9):2535-2542.
[4]张国,崔克辉. 水稻茎鞘非结构性碳水化合物积累与转运研究进展[J]. 植物生理学报,2020,56(5):1127-1136.
[5]张锦滨,王晓云,孟圆,等. 不同磷肥用量对水稻产量效益、磷肥利用率及土壤养分的影响[J]. 中国农学通报,2021,37(32):96-101.
[6]李雨阳,赵喜辉,郝威名,等. 施磷量对不同磷敏感型水稻产量、品质及根系生长的影响[J]. 中国土壤与肥料,2023(12):219-226.
[7]张淑香,徐明岗. 土壤磷素演变与高效利用[J]. 中国农业科学,2019,52(21):3828-3829.
[8]陈明. 基于环境可持续性的水稻磷肥精准管理策略[J]. 中国农业资源与区划,2023,44(2):123-130.
[9]Veneklaas E J,Lambers H,Bragg J,et al. Opportunities for improving phosphorus-use efficiency in crop plants[J]. New Phytologist,2012,195(2):306-320.
[10]Fageria N K. The use of nutrients in crop plants[M]. Boca Raton:CRC Press,2016.
[11]Mahajan G R,Pandey R N,Sahoo R N,et al. Monitoring nitrogen,phosphorus and sulphur in hybrid rice (Oryza sativa L.) using hyperspectral remote sensing[J]. Precision Agriculture,2017,18(5):713-739.
[12]Mahlein A K,Oerke E C,Steiner U,et al. Hyperspectral sensors and imaging technologies in phytopathology:state of the art[J]. Annual Review of Phytopathology,2018,56:535-558.
[13]熊俊涛,戴森鑫,区炯洪,等. 基于深度学习的大豆生长期叶片缺素症状检测方法[J]. 农业机械学报,2020,51(1):195-202.
[14]Xu Z,Guo X,Zhu A F,et al. Using deep convolutional neural networks for image-based diagnosis of nutrient deficiencies in rice[J]. Computational Intelligence and Neuroscience,2020(1):7307252.
[15]杨红云,郭紫微,郭高飞,等. 基于Stacking集成卷积神经网络的水稻氮素营养诊断[J]. 植物营养与肥料学报,2023,29(3):573-581.
[16]钱政,杨孙哲,张国卿,等. 基于卷积神经网络的水稻氮素营养诊断[J]. 中国农业科技导报,2023,25(9):113-121.
[17]Bishwas R,Yasmin S,Turzo N A. Relative comparison of K-means clustering segmented rice leaves image-based nitrogen,phosphorus,and potassium nutrient deficiency classification using convolutional neural network[C]//Proceedings of the 2021 International Conference on Science & Contemporary Technologies. Dhaka:IEEE,2021:1-6.
[18]张林朋,杨红云,钱政,等. 基于改进的VGG16网络和迁移学习的水稻氮素营养诊断[J]. 中国农业大学学报,2023,28(6):219-229.
[19]孔杰,杨红云,黄淑梅,等. 基于残差神经网络的水稻氮磷钾元素营养诊断[J]. 中国农业大学学报,2025,30(2):163-175.
[20]Sethy P K,Barpanda N K,Rath A K,et al. Nitrogen deficiency prediction of rice crop based on convolutional neural network[J]. Journal of Ambient Intelligence and Humanized Computing,2020,11:5703-5711.
[21]黄淑梅,杨红云,孔杰,等. 基于改进ShuffleNet v2的水稻磷素营养诊断方法[J]. 中国稻米,2025,31(2):20-28,34.
[22]吴正,杨红云,孙爱珍,等. 基于PND-Net模型的水稻钾素胁迫程度诊断识别[J]. 江西农业大学学报,2024,46(6):1654-1667.
[23]Thai H T,Le K H,Nguyen N L T. FormerLeaf:an efficient vision transformer for Cassava Leaf Disease detection[J]. Computers and Electronics in Agriculture,2023,204:107518.
[24]Li G,Jiao L,Chen P,et al. Spatial convolutional self-attention-based transformer module for strawberry disease identification under complex background[J]. Computers and Electronics in Agriculture,2023,212:108121.
[25]Tan S Y,Lu H H,Yu J,et al. In-field rice panicles detection and growth stages recognition based on RiceRes2Net[J]. Computers and Electronics in Agriculture,2023,206:107704.
[26]Wang C L,Li M W,Chan Y K,et al. Multi-scale features fusion convolutional neural networks for rice leaf disease identification[J]. Journal of Imaging Science and Technology,2022,66(5):1-12.
[27]Vaswani A,Shazeer N,Parmar N,et al. Attention is all you need[C]//Advances in Neural Information Processing Systems (NeurIPS).2017:30.
[28]Lin T Y,Dollár P,Girshick R,et al. Feature pyramid networks for object detection[EB/OL]. (2017-12-09)[2024-11-12]. https://arxiv.org/abs/1612.03144.
[29]Hu J,Shen L,Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE International Conference on Computer Vision. Salt Lake City:IEEE,2018:713-722.
[30]Fageria N K,Baligar V C,Clark R B. Micronutrients in crop production[J]. Advances in Agronomy,2002,77:185-268.
[31]Liu Z,Lin Y,Cao Y,et al. Swin Transformer:hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Montreal:IEEE,2021:10012-10022.
[32]周琼,杨红云,杨珺,等. 基于BP神经网络和概率神经网络的水稻图像氮素营养诊断[J]. 植物营养与肥料学报,2019,25(1):134-141.
[33]Ghosal S,Zheng B,Chapman S C,et al. An explainable deep machine vision framework for plant stress phenotyping[J]. Proceedings of the National Academy of Sciences of the United States of America,2018,115(18):4613-4618.
[34]郭丽丽,李彦生,于镇华,等. 长期大气CO2浓度升高对大豆磷吸收及根际磷转化的影响[J]. 土壤与作物,2024,13(1):27-38.

相似文献/References:

[1]马旭俊,刘春娟,吕世博,等.绿色荧光蛋白基因在水稻遗传转化中的应用[J].江苏农业科学,2013,41(04):35.
[2]李岳峰,居立海,张来运,等.水分胁迫下丛枝菌根对水稻/绿豆间作系统 作物生长和氮磷吸收的影响[J].江苏农业科学,2013,41(04):58.
[3]崔月峰,孙国才,王桂艳,等.不同施氮水平和前氮后移措施对水稻产量 及氮素利用率的影响[J].江苏农业科学,2013,41(04):66.
[4]张其蓉,宋发菊,田进山,等.长江中下游稻区水稻区域试验品种抗稻瘟病鉴定与评价[J].江苏农业科学,2013,41(04):92.
[5]王麒,张小明,卞景阳,等.不同插秧密度对黑龙江省第二积温带水稻产量及产量构成的影响[J].江苏农业科学,2013,41(05):60.
 Wang Qi,et al.Effect of different transplanting density on yield and yield component of rice in second temperature zone of Heilongjiang Province[J].Jiangsu Agricultural Sciences,2013,41(4):60.
[6]张国良,张森林,丁秀文,等.基质厚度和含水量对水稻育秧的影响[J].江苏农业科学,2013,41(05):62.
 Zhang Guoliang,et al.Effects of substrate thickness and water content on growth of rice seedlings[J].Jiangsu Agricultural Sciences,2013,41(4):62.
[7]赵忠宝,朱清海.稻-蟹-鳅生态系统的能值分析[J].江苏农业科学,2013,41(05):349.
 Zhao Zhongbao,et al.Emergy analysis of paddy-crab-loach ecosystem[J].Jiangsu Agricultural Sciences,2013,41(4):349.
[8]杨红福,姚克兵,束兆林,等.甲氧基丙烯酸酯类杀菌剂对水稻恶苗病的田间药效[J].江苏农业科学,2014,42(12):166.
 Yang Hongfu,et al.Field efficacy of strobilurin fungicides against rice bakanae disease[J].Jiangsu Agricultural Sciences,2014,42(4):166.
[9]唐成,陈露,安敏敏,等.稻瘟病诱导水稻幼苗叶片氧化还原系统的特征谱变化[J].江苏农业科学,2014,42(12):141.
 Tang Cheng,et al.Characteristic spectral changes of redox homeostasis system in rice seedling leaves induced by rice blast[J].Jiangsu Agricultural Sciences,2014,42(4):141.
[10]万云龙.优质水稻—春甘蓝轮作高效栽培模式[J].江苏农业科学,2014,42(12):90.
 Wan Yunlong.Efficient cultivation mode of high quality rice-spring cabbage rotation[J].Jiangsu Agricultural Sciences,2014,42(4):90.

备注/Memo

备注/Memo:
收稿日期:2025-05-15
基金项目:国家自然科学基金(编号:61562039、62162030)。
作者简介:黄带娣(2001—),女,江西赣州人,硕士研究生,主要从事农业信息技术研究。E-mail:2871774998@qq.com。
通信作者:孙爱珍,硕士,副教授,主要从事农业信息技术研究。E-mail:nc_saz@163.com。
更新日期/Last Update: 2026-02-20