Identification method of vegetable diseases based on transfer learning and attention mechanism
文献类型: 外文期刊
第一作者: Zhao, Xue
作者: Zhao, Xue;Li, Kaiyu;Li, Yunxia;Zhang, Lingxian;Ma, Juncheng;Zhang, Lingxian
作者机构:
关键词: Vegetable disease; Attention mechanism; Transfer learning; Convolutional neural network
期刊名称:COMPUTERS AND ELECTRONICS IN AGRICULTURE ( 影响因子:6.757; 五年影响因子:6.817 )
ISSN: 0168-1699
年卷期: 2022 年 193 卷
页码:
收录情况: SCI
摘要: Artificial Intelligence for disease identification is currently the focus of great research interest. Nonetheless, the approach has some problems, for example, identification takes a long time, has low accuracy, and is often limited to a single disease type. Here, we aimed to identify tomato powdery mildew, leaf mold and cucumber downy mildew against simple and complex backgrounds. We developed a vegetable disease identification model, DTLSE-ResNet50, optimized by SENet and pre-trained by ImageNet to form a new model, SE-ResNet50. The SEResNet50 model was trained with the AI Challenger 2018 public database to obtain a new weight. The SEResNet50 model with the new weight was then trained by dual transfer learning with a self-built database to create the DTL-SE-ResNet50 model for the identification of vegetable diseases. The model was compared with convolutional neural networks EfficientNet, AlexNet, VGG19, and Inception V3. The experimental results showed that with the same experimental conditions, the identification precision of the new model reached 97.24%, and processing of a single image required 0.13 s. Compared with DTL-CBAM-ResNet50 and DTL-SA-ResNet50, three models has almost the same precision, but time consumption of DTL-SE-ResNet50 was 0.02 s higher than that of DTL-CBAM-ResNet50. Although the time consumption of DTL-SA-ResNet50 was 0.02 s higher than the proposed model, the precision was lower. At the same time, compared with the dual transfer learning model, the model's precision was 4.1% higher, and the processing of a single image was 0.06 s shorter. Compared with convolutional neural networks, the precision of DTL-SE-ResNet50 was 3.19% higher than the best result, the time consumption of a single image was 0.58 s shorter; Recall and F1 also increased. The method proposed in this paper has high identification precision and short identification time, and it meets the requirements for accurate and rapid identification of vegetable diseases.
分类号:
- 相关文献
作者其他论文 更多>>
-
Estimating daily minimum grass temperature to quantify frost damage to winter wheat during stem elongation in the central area of Huang-Huai plain in China
作者:Wu, Yongfeng;Ji, Lin;Ma, Juncheng;Gong, Zhihong
关键词:Minimum grass temperature; Environmental variables; Winter wheat; Stem elongation; Frost damage
-
Lightweight fungal spore detection based on improved YOLOv5 in natural scenes
作者:Li, Kaiyu;Qiao, Chen;Zhu, Xinyi;Song, Yuzhaobi;Zhang, Lingxian;Zhang, Lingxian;Gao, Wei;Wang, Yong
关键词:Fungal spores; Object detection; YOLOv5; Attention mechanism
-
Wheat Yield Prediction Using Unmanned Aerial Vehicle RGB-Imagery-Based Convolutional Neural Network and Limited Training Samples
作者:Ma, Juncheng;Wu, Yongfeng;Liu, Binhui;Zhang, Wenying;Wang, Bianyin;Chen, Zhaoyang;Wang, Guangcai;Guo, Anqiang;Liu, Binhui;Zhang, Wenying;Wang, Bianyin;Chen, Zhaoyang;Wang, Guangcai;Guo, Anqiang
关键词:yield prediction; winter wheat; split-merge; convolutional neural network; UAV RGB imagery
-
The Gray Mold Spore Detection of Cucumber Based on Microscopic Image and Deep Learning
作者:Li, Kaiyu;Zhu, Xinyi;Qiao, Chen;Zhang, Lingxian;Zhang, Lingxian;Gao, Wei;Wang, Yong
关键词:
-
Irrigation depth and nitrogen rate effects on seed cotton yield, fiber quality and water-nitrogen utilization efficiency in southern Xinjiang, China
作者:Bai, Zhentao;Bai, Wenqiang;Xie, Cong;Yu, Jiang;Dai, Yulong;Pei, Shengzhao;Zhang, Fucang;Fan, Junliang;Yin, Feihu;Li, Yunxia;Fan, Junliang;Yin, Feihu;Li, Yunxia;Yin, Feihu
关键词:Crop water productivity; Partial factor productivity of nitrogen; Seed cotton yield; Nitrogen distribution; Fiber quality
-
Genome-wide identification and expression profiling of 4-coumarate:coenzyme A ligase genes influencing soybean isoflavones at the seedling stage
作者:Yang, Zhenhong;Wu, Xu;Ma, Jinglin;Zhan, Yuhang;Li, Yonguang;Li, Haiyan;Teng, Weili;Zhao, Xue;Han, Yingpeng;Yuan, Ming
关键词:4-coumarate:coenzyme A ligase; bioinformatics analysis; genome-wide expression analysis; isoflavone; qPCR; secondary metabolites; seedling stage; soybean.
-
Identification of QTL, QTL-by-environment interactions, and their candidate genes for resistance HG Type 0 and HG Type 1.2.3.5.7 in soybean using 3VmrMLM
作者:Jiang, Haipeng;Lv, Suchen;Qu, Shuo;Liu, Fang;Sun, Haowen;Zhao, Xue;Han, Yingpeng;Zhou, Changjun
关键词:soybean cyst nematode; GWAS; 3VmrMLM; QTL; candidate genes