Wei Chen

Wei Chen

Auto-generated profile, based on publicly available information

Wei Chen

College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua, Zhejiang, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 August 2024 in IEEE Access
Verified
Journal Article
Open access
WCWei Chen
Wei Chen
LZLijuan Zheng
Lijuan Zheng
JXJiping Xiong
Jiping Xiong
Crop diseases and pests cause significant economic losses to agriculture every year, making accurate identification crucial. Traditional pest detection relies on farm experts and is often very time-consuming. Computer vision technology and artificial intelligence can provide automatic disease detection, enabling real-time and precise control of crop diseases and timely prevention. To accurately identify plant diseases under complex natural conditions, an improved crop disease and pest recognition model was established based on the original YOLOv5 network model.First, we integrated the Squeeze-and-Excitation (SE) attention module into YOLOv5, allowing our proposed model to better differentiate leaf features between different crops and accurately identify disease types. Second, to enhance the model’s feature extraction capability for diseased areas and reduce the loss of disease feature information, we replaced the original upsampling operator in YOLOv5 with the lightweight Content-Aware ReAssembly of FEatures (CARAFE) module. Next, we improved the original loss function by adopting the Efficient Intersection over Union (EIoU) loss function to increase detection accuracy. Finally, to reduce the model’s complexity and meet real-time detection requirements, we introduced the Ghost convolution module into the backbone network.During the experimental phase, to validate the model’s effectiveness, we randomly divided sample images from the constructed crop disease and pest database into training, validation, and test sets. The experimental results showed that the improved YOLOv5 model achieved a precision of 90.0%, a recall of 91.4%, an mAP@.50 of 92.1%, and an mAP@.5:.95 of 64%. The number of parameters and computational load were reduced by 23.9% and 31.2%, respectively, outperforming popular methods including YOLOv5, YOLOv7, and YOLOv8.The improved model can accurately identify crop diseases and pests under natural conditions and is conducive to deployment on edge devices, providing a technical reference for crop disease and pest management.