An Industrial-Grade Solution for Crop Disease Image Detection Tasks
文献类型: 外文期刊
第一作者: Dai, Guowei
作者: Dai, Guowei;Fan, Jingchao;Fan, Jingchao
作者机构:
关键词: crop disease detection; convolutional neural network; model compression; knowledge distillation; activate quantitative; model deployment
期刊名称:FRONTIERS IN PLANT SCIENCE ( 影响因子:6.627; 五年影响因子:7.255 )
ISSN: 1664-462X
年卷期: 2022 年 13 卷
页码:
收录情况: SCI
摘要: Crop leaf diseases can reflect the current health status of the crop, and the rapid and automatic detection of field diseases has become one of the difficulties in the process of industrialization of agriculture. In the widespread application of various machine learning techniques, recognition time consumption and accuracy remain the main challenges in moving agriculture toward industrialization. This article proposes a novel network architecture called YOLO V5-CAcT to identify crop diseases. The fast and efficient lightweight YOLO V5 is chosen as the base network. Repeated Augmentation, FocalLoss, and SmoothBCE strategies improve the model robustness and combat the positive and negative sample ratio imbalance problem. Early Stopping is used to improve the convergence of the model. We use two technical routes of model pruning, knowledge distillation and memory activation parameter compression ActNN for model training and identification under different hardware conditions. Finally, we use simplified operators with INT8 quantization for further optimization and deployment in the deep learning inference platform NCNN to form an industrial-grade solution. In addition, some samples from the Plant Village and AI Challenger datasets were applied to build our dataset. The average recognition accuracy of 94.24% was achieved in images of 59 crop disease categories for 10 crop species, with an average inference time of 1.563 ms per sample and model size of only 2 MB, reducing the model size by 88% and the inference time by 72% compared with the original model, with significant performance advantages. Therefore, this study can provide a solid theoretical basis for solving the common problems in current agricultural disease image detection. At the same time, the advantages in terms of accuracy and computational cost can meet the needs of agricultural industrialization.
分类号:
- 相关文献
作者其他论文 更多>>
-
PPLC-Net:Neural network-based plant disease identification model supported by weather data augmentation and multi-level attention mechanism
作者:Dai, Guowei;Fan, Jingchao;Fan, Jingchao;Tian, Zhimin;Wang, Chaoyu
关键词:Convolutional neural network; Dilated convolutions; Global average pooling; Attention mechanism (CBAM); Weather data augmentation; Leaf disease recognition
-
ITF-WPI: Image and text based cross-modal feature fusion model for wolfberry pest recognition
作者:Dai, Guowei;Fan, Jingchao;Fan, Jingchao;Dewi, Christine;Fan, Jingchao
关键词:Cross-modal fusion; Contextual transformer; Pyramid squeeze attention mechanism; Convolutional neural network and bi-; directional long short-term memory; Pest recognition
-
DA-ActNN-YOLOV5: Hybrid YOLO v5 Model with Data Augmentation and Activation of Compression Mechanism for Potato Disease Identification
作者:Dai, Guowei;Hu, Lin;Fan, Jingchao;Hu, Lin;Fan, Jingchao
关键词:
-
A Deep Learning-Based Object Detection Scheme by Improving YOLOv5 for Sprouted Potatoes Datasets
作者:Dai, Guowei;Hu, Lin;Fan, Jingchao;Yan, Shen;Li, Ruijing;Hu, Lin;Fan, Jingchao
关键词:Object detection; convolutional neural network; sprouting potato recognition; mosaic; hyperparametric optimization; spatial pyramid pooling
-
Cloud-Based Video Monitoring System Applied in Control of Diseases and Pests in Orchards
作者:Xia, Xue;Qiu, Yun;Hu, Lin;Fan, Jingchao;Guo, Xiuming;Zhou, Guomin
关键词:Video cloud;Video monitoring;Diseases and pests;Apple orchards
-
Development Research on the Expert System for the Binary Tree-Based Apple Disease Diagnosis
作者:Qiu, Yun;Fan, Jingchao;Zhou, Guomin
关键词:Binary Tree-Based Knowledge;Knowledge System;Knowledge Representation;Knowledge Editing;Apple Disease Diagnosis