Frontiers in Plant Science (Jan 2023)
DenseNet weed recognition model combining local variance preprocessing and attention mechanism
- Ye Mu,
- Ye Mu,
- Ye Mu,
- Ye Mu,
- Ruiwen Ni,
- Ruiwen Ni,
- Ruiwen Ni,
- Ruiwen Ni,
- Lili Fu,
- Lili Fu,
- Lili Fu,
- Lili Fu,
- Tianye Luo,
- Tianye Luo,
- Tianye Luo,
- Tianye Luo,
- Ruilong Feng,
- Ruilong Feng,
- Ruilong Feng,
- Ruilong Feng,
- Ji Li,
- Ji Li,
- Ji Li,
- Ji Li,
- Haohong Pan,
- Haohong Pan,
- Haohong Pan,
- Haohong Pan,
- Yingkai Wang,
- Yu Sun,
- Yu Sun,
- Yu Sun,
- Yu Sun,
- He Gong,
- He Gong,
- He Gong,
- He Gong,
- Ying Guo,
- Ying Guo,
- Ying Guo,
- Ying Guo,
- Tianli Hu,
- Tianli Hu,
- Tianli Hu,
- Tianli Hu,
- Yu Bao,
- Shijun Li,
- Shijun Li
Affiliations
- Ye Mu
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Ye Mu
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Ye Mu
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ye Mu
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruiwen Ni
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Ruiwen Ni
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruiwen Ni
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruiwen Ni
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Lili Fu
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Lili Fu
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Lili Fu
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Lili Fu
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianye Luo
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Tianye Luo
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianye Luo
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianye Luo
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruilong Feng
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Ruilong Feng
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruilong Feng
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ruilong Feng
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ji Li
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Ji Li
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Ji Li
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ji Li
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Haohong Pan
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Haohong Pan
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Haohong Pan
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Haohong Pan
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Yingkai Wang
- Faculty of Agronomy, Jilin Agricultural University, Changchun, China
- Yu Sun
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Yu Sun
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Yu Sun
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Yu Sun
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- He Gong
- College of Information Technology, Jilin Agricultural University, Changchun, China
- He Gong
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- He Gong
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- He Gong
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ying Guo
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Ying Guo
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Ying Guo
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Ying Guo
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianli Hu
- College of Information Technology, Jilin Agricultural University, Changchun, China
- Tianli Hu
- Jilin Province Agricultural Internet of Things Technology Collaborative Innovation Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianli Hu
- Jilin Province Intelligent Environmental Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Tianli Hu
- Jilin Province Information Technology and Intelligent Agriculture Engineering Research Center, Jilin Agricultural University, Changchun, Jilin, China
- Yu Bao
- School of Management, Changchun University, Changchun, China
- Shijun Li
- College of Information Technology, Wuzhou University, Wuzhou, China
- Shijun Li
- Guangxi Key Laboratory of Machine Vision and Intelligent Control, Wuzhou University, Wuzhou, Guangxi, China
- DOI
- https://doi.org/10.3389/fpls.2022.1041510
- Journal volume & issue
-
Vol. 13
Abstract
IntroductionThe purpose of this paper is to effectively and accurately identify weed species in crop fields in complex environments. There are many kinds of weeds in the detection area, which are densely distributed.MethodsThe paper proposes the use of local variance pre-processing method for background segmentation and data enhancement, which effectively removes the complex background and redundant information from the data, and prevents the experiment from overfitting, which can improve the accuracy rate significantly. Then, based on the optimization improvement of DenseNet network, Efficient Channel Attention (ECA) mechanism is introduced after the convolutional layer to increase the weight of important features, strengthen the weed features and suppress the background features.ResultsUsing the processed images to train the model, the accuracy rate reaches 97.98%, which is a great improvement, and the comprehensive performance is higher than that of DenseNet, VGGNet-16, VGGNet-19, ResNet-50, DANet, DNANet, and U-Net models.DiscussionThe experimental data show that the model and method we designed are well suited to solve the problem of accurate identification of crop and weed species in complex environments, laying a solid technical foundation for the development of intelligent weeding robots.
Keywords
