A New Pest Detection Method Based on Improved YOLOv5m

被引:50
作者
Dai, Min [1 ]
Dorjoy, Md Mehedi Hassan [1 ]
Miao, Hong [1 ]
Zhang, Shanwen [1 ]
机构
[1] Yangzhou Univ, Coll Mech Engn, Yangzhou 225127, Peoples R China
关键词
deep learning; convolutional neural networks; pest detection; YOLOv5m; NETWORK; ALGORITHM;
D O I
10.3390/insects14010054
中图分类号
Q96 [昆虫学];
学科分类号
摘要
Simple Summary Insect pests can damage crops and food production, causing problems for farmers. The detection of plant pests is essential for ensuring the excellent productivity of plants and food. Traditional methods for pest detection are generally time-consuming and inefficient. There has been a lot of use of deep learning for detecting plant pests in recent years. YOLOv5 is one of the most effective deep learning algorithms used for object detection. A new pest detection method with higher accuracy based on a deep convolutional neural network (CNN) is proposed in this paper. Experimental results on the pest dataset indicate that the proposed method performs well and can achieve high precision and robustness for recognizing plant pests. The proposed method is more effective and can detect pests precisely with higher accuracy. Pest detection in plants is essential for ensuring high productivity. Convolutional neural networks (CNN)-based deep learning advancements recently have made it possible for researchers to increase object detection accuracy. In this study, pest detection in plants with higher accuracy is proposed by an improved YOLOv5m-based method. First, the SWin Transformer (SWinTR) and Transformer (C3TR) mechanisms are introduced into the YOLOv5m network so that they can capture more global features and can increase the receptive field. Then, in the backbone, ResSPP is considered to make the network extract more features. Furthermore, the global features of the feature map are extracted in the feature fusion phase and forwarded to the detection phase via a modification of the three output necks C3 into SWinTR. Finally, WConcat is added to the fusion feature, which increases the feature fusion capability of the network. Experimental results demonstrate that the improved YOLOv5m achieved 95.7% precision rate, 93.1% recall rate, 94.38% F-1 score, and 96.4% Mean Average Precision (mAP). Meanwhile, the proposed model is significantly better than the original YOLOv3, YOLOv4, and YOLOv5m models. The improved YOLOv5m model shows greater robustness and effectiveness in detecting pests, and it could more precisely detect different pests from the dataset.
引用
收藏
页数:17
相关论文
共 34 条
[1]   Automated Pest Detection With DNN on the Edge for Precision Agriculture [J].
Albanese, Andrea ;
Nardello, Matteo ;
Brunelli, Davide .
IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS, 2021, 11 (03) :458-467
[2]   Insect detection from imagery using YOLOv3-based adaptive feature fusion convolution network [J].
Amrani, Abderraouf ;
Sohel, Ferdous ;
Diepeveen, Dean ;
Murray, David ;
Jones, Michael G. K. .
CROP & PASTURE SCIENCE, 2023, 74 (06) :615-627
[3]   Automatic Detection and Monitoring of Insect Pests-A Review [J].
Cardim Ferreira Lima, Matheus ;
Damascena de Almeida Leandro, Maria Elisa ;
Valero, Constantino ;
Pereira Coronel, Luis Carlos ;
Goncalves Bazzo, Clara Oliva .
AGRICULTURE-BASEL, 2020, 10 (05)
[4]   A Smartphone-Based Application for Scale Pest Detection Using Multiple-Object Detection Methods [J].
Chen, Jian-Wen ;
Lin, Wan-Ju ;
Cheng, Hui-Jun ;
Hung, Che-Lun ;
Lin, Chun-Yuan ;
Chen, Shu-Pei .
ELECTRONICS, 2021, 10 (04) :1-14
[5]   Machine Learning for Detection and Prediction of Crop Diseases and Pests: A Comprehensive Survey [J].
Domingues, Tiago ;
Brandao, Tomas ;
Ferreira, Joao C. .
AGRICULTURE-BASEL, 2022, 12 (09)
[6]   Vision-based pest detection based on SVM classification method [J].
Ebrahimi, M. A. ;
Khoshtaghaz, M. H. ;
Minaei, S. ;
Jamshidi, B. .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2017, 137 :52-58
[7]   MS-ALN: Multiscale Attention Learning Network for Pest Recognition [J].
Feng, Fuxiang ;
Dong, Hanlin ;
Zhang, Youmei ;
Zhang, Yu ;
Li, Bin .
IEEE ACCESS, 2022, 10 :40888-40898
[8]   Cas-VSwin transformer: A variant swin transformer for surface-defect detection [J].
Gao, Linfeng ;
Zhang, Jianxun ;
Yang, Changhui ;
Zhou, Yuechuan .
COMPUTERS IN INDUSTRY, 2022, 140
[9]   PCT: Point cloud transformer [J].
Guo, Meng-Hao ;
Cai, Jun-Xiong ;
Liu, Zheng-Ning ;
Mu, Tai-Jiang ;
Martin, Ralph R. ;
Hu, Shi-Min .
COMPUTATIONAL VISUAL MEDIA, 2021, 7 (02) :187-199
[10]   Automatic monitoring of flying vegetable insect pests using an RGB camera and YOLO-SIP detector [J].
Guo, Qingwen ;
Wang, Chuntao ;
Xiao, Deqin ;
Huang, Qiong .
PRECISION AGRICULTURE, 2023, 24 (02) :436-457