A new model based on improved VGG16 for corn weed identification

被引:7
作者
Yang, Le [1 ]
Xu, Shuang [2 ]
Yu, XiaoYun [1 ]
Long, HuiBin [1 ]
Zhang, HuanHuan [1 ]
Zhu, YingWen [1 ]
机构
[1] Jiangxi Agr Univ, Sch Comp & Informat Engn, Nanchang, Peoples R China
[2] Jiangxi Agr Univ, Software Coll, Nanchang, Peoples R China
来源
FRONTIERS IN PLANT SCIENCE | 2023年 / 14卷
基金
中国国家自然科学基金;
关键词
attention mechanism; corn weed; deep convolutional neural network; global average pooling; Leaky ReLU; RECOGNITION;
D O I
10.3389/fpls.2023.1205151
中图分类号
Q94 [植物学];
学科分类号
071001 ;
摘要
Weeds remain one of the most important factors affecting the yield and quality of corn in modern agricultural production. To use deep convolutional neural networks to accurately, efficiently, and losslessly identify weeds in corn fields, a new corn weed identification model, SE-VGG16, is proposed. The SE-VGG16 model uses VGG16 as the basis and adds the SE attention mechanism to realize that the network automatically focuses on useful parts and allocates limited information processing resources to important parts. Then the 3 x 3 convolutional kernels in the first block are reduced to 1 x 1 convolutional kernels, and the ReLU activation function is replaced by Leaky ReLU to perform feature extraction while dimensionality reduction. Finally, it is replaced by a global average pooling layer for the fully connected layer of VGG16, and the output is performed by softmax. The experimental results verify that the SE-VGG16 model classifies corn weeds superiorly to other classical and advanced multiscale models with an average accuracy of 99.67%, which is more than the 97.75% of the original VGG16 model. Based on the three evaluation indices of precision rate, recall rate, and F1, it was concluded that SE-VGG16 has good robustness, high stability, and a high recognition rate, and the network model can be used to accurately identify weeds in corn fields, which can provide an effective solution for weed control in corn fields in practical applications.
引用
收藏
页数:11
相关论文
共 35 条
[1]   Plant disease identification from individual lesions and spots using deep learning [J].
Arnal Barbedo, Jayme Garcia .
BIOSYSTEMS ENGINEERING, 2019, 180 :96-107
[2]  
Bahdanau D, 2016, Arxiv, DOI [arXiv:1409.0473, 10.48550/arXiv.1409.0473, DOI 10.48550/ARXIV.1409.0473]
[3]  
Bongulwar Deepali M., 2021, IOP Conference Series: Materials Science and Engineering, V1049, DOI 10.1088/1757-899X/1049/1/012004
[4]   Automated recognition of optical image based potato leaf blight diseases using deep learning [J].
Chakraborty, Kulendu Kashyap ;
Mukherjee, Rashmi ;
Chakroborty, Chandan ;
Bora, Kangkana .
PHYSIOLOGICAL AND MOLECULAR PLANT PATHOLOGY, 2022, 117
[5]   Multi-channel feature fusion networks with hard coordinate attention mechanism for maize disease identification under complex backgrounds [J].
Fang, Shundong ;
Wang, Yanfeng ;
Zhou, Guoxiong ;
Chen, Aibin ;
Cai, Weiwei ;
Wang, Qifan ;
Hu, Yahui ;
Li, Liujun .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2022, 203
[6]   Field Weed Recognition Based on an Improved VGG With Inception Module [J].
Fu, Lifang ;
Lv, Xingchen ;
Wu, Qiufeng ;
Pei, Chengyan .
INTERNATIONAL JOURNAL OF AGRICULTURAL AND ENVIRONMENTAL INFORMATION SYSTEMS, 2020, 11 (02) :1-13
[7]  
Haijian Ye, 2019, Journal of Physics: Conference Series, V1237, DOI 10.1088/1742-6596/1237/3/032018
[8]  
Hu J, 2018, PROC CVPR IEEE, P7132, DOI [10.1109/CVPR.2018.00745, 10.1109/TPAMI.2019.2913372]
[9]   Rock Image Classification Based on EfficientNet and Triplet Attention Mechanism [J].
Huang, Zhihao ;
Su, Lumei ;
Wu, Jiajun ;
Chen, Yuhan .
APPLIED SCIENCES-BASEL, 2023, 13 (05)
[10]  
Jaderberg M, 2015, ADV NEUR IN, V28