| 计算机视觉下的农作物病虫害图像识别研究 |
| |
| 引用本文： | 钟林忆,刘海峰,董力中,高翔,黄家怿,司志恒.计算机视觉下的农作物病虫害图像识别研究[J].现代农业装备,2021,42(1):51-55. |
| |
| 作者姓名： | 钟林忆 刘海峰 董力中 高翔 黄家怿 司志恒 |
| |
| 作者单位： | 广东省现代农业装备研究所,广东 广州 510630;广州市健坤网络科技发展有限公司,广东广州 510630;广州市健坤网络科技发展有限公司,广东广州 510630;广东省现代农业装备研究所,广东 广州 510630;广东省农业技术推广总站,广东 广州 510520 |
| |
| 基金项目： | 广东省重点领域研发计划项目(2019B020223003);广东省科技创新战略专项资金(省属科研机构稳定性支持)项目(粤科资字[2020]81号);广东省智能农机装备产业技术创新项目(粤财农[2020]39号)。 |
| |
| 摘 要： | 本文针对农作物病虫害图像识别需求,探索了基于数据增广技术的深度卷积神经网络迁移学习方法及识别模型,将原始样本量扩增至50倍,并通过抑制模型过拟合,从而提升模型的泛化能力和农作物病虫害识别的准确率。同时基于边缘计算理论方法与技术,将识别模型部署到边缘端,设计了基于计算机视觉与边缘计算的智能识别装置,通过该装置实时采集农作物图像,并进行图像推理与识别,解决了农作物病虫害图像识别的实际应用问题。
|
| 关 键 词： | 小样本 图像识别 数据增广 卷积神经网络 农作物病虫害 |
Image Recognition of Crop Diseases and Insect Pests based on Computer Vision |
| |
| Zhong Linyi,Liu Haifeng,Dong Lizhong,Gao Xiang,Liu Chaoyang,Si Zhiheng.Image Recognition of Crop Diseases and Insect Pests based on Computer Vision[J].Modern Agricultural Equipments,2021,42(1):51-55. |
| |
| Authors: | Zhong Linyi Liu Haifeng Dong Lizhong Gao Xiang Liu Chaoyang Si Zhiheng |
| |
| Institution: | (Guangdong Institute of Modern Agricultural Equipment,Guangzhou 510630,China;Guangzhou Joinken Network Technology Development Co.,Ltd.,Guangzhou 510630,China;Guangdong Province General Station for Agriculture Technology Extension,Guangzhou 510520,China) |
| |
| Abstract: | Aiming at the needs of crop diseases and insect pests image recognition,this paper explores the deep convolution neural network migration learning method and recognition model based on data augmentation technology.The original sample size is expanded to 50 times,and the model over fitting is suppressed,so as to improve the generalization ability of the model and the accuracy of crop diseases and insect pests recognition.At the same time,based on the theory and technology of edge computing,the recognition model is deployed to the edge,and an intelligent recognition device based on computer vision and edge computing is designed.Through the intelligent recognition device,the crop images are collected in real time,and the image reasoning and recognition are carried out,which solves the practical application problems of image recognition of agricultural pests and diseases. |
| |
| Keywords: | small sample image recognition data augmentation convolution neural network crop diseases and insect pests |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
