首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于深度学习的蔬菜鳞翅目害虫自动识别与检测计数
引用本文:董伟,钱蓉,张洁,张立平,陈红波,张萌,朱静波,卜英乔.基于深度学习的蔬菜鳞翅目害虫自动识别与检测计数[J].中国农业科技导报,2019,21(12):76-84.
作者姓名:董伟  钱蓉  张洁  张立平  陈红波  张萌  朱静波  卜英乔
作者单位:1.安徽省农业科学院农业经济与信息研究所, 合肥 230001; 2.中国科学院合肥物质科学研究院合肥智能机械研究所, 合肥 230031; 3.国防科技大学电子对抗学院, 合肥 230037
基金项目:国家自然科学基金项目(31671586);中国科学院科技服务网络计划(STS计划)项目(KFJ-STS-ZDTP-048-02);安徽省农业科学院人才发展专项资金项目(17F1414);安徽省农业科学院院所共建团队项目(18C1424)。
摘    要:鳞翅目害虫是蔬菜作物中最重要且常见的一类害虫。由于受到复杂田间背景,光照及害虫姿态等的影响,传统的害虫自动识别与检测计数方法准确率比较低。为实现在田间快速准确地对目标害虫进行自动识别和检测计数,分别提出了基于深度卷积神经网络的识别模型和检测计数模型。针对菜粉蝶、棉铃虫、甜菜夜蛾、小菜蛾、斜纹夜蛾这5种常见且容易混淆的蔬菜鳞翅目害虫,构建了分类识别数据集和检测计数数据集,并分别进行了实验,平均识别率达到94.5%,检测均值平均精度(mAP)达到76.6%,与传统方法相比,证明了此方法的优越性。实验结果表明,该方法对于蔬菜鳞翅目害虫的识别和检测计数是可行的,且达到了实际应用水平。

关 键 词:鳞翅目害虫  自动识别  检测计数  深度学习  卷积神经网络  

Vegetable Lepidopteran Pest Auto Recognition and Detection Counting Based on Deep Learning
DONG Wei,QIAN Rong,ZHANG Jie,ZHANG Liping,CHEN Hongbo,ZHANG Meng,ZHU Jingbo,BU Yingqiao.Vegetable Lepidopteran Pest Auto Recognition and Detection Counting Based on Deep Learning[J].Journal of Agricultural Science and Technology,2019,21(12):76-84.
Authors:DONG Wei  QIAN Rong  ZHANG Jie  ZHANG Liping  CHEN Hongbo  ZHANG Meng  ZHU Jingbo  BU Yingqiao
Institution:1.Agricultural Economy and Information Research Insitute, Anhui Academy of Agricultural Sciences, Hefei 230001, China; 2.Institute of Intelligent Machines, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China; 3.College of Electronic Countermeasures, National University of Defense Technology,  Hefei 230037, China
Abstract:Lepidopteran pests are the most important and common pests in vegetable crops. Due to the complex vegetable garden background, sunlight and pest gestures, conventional methods of pests image recognition, detection and counting may not be accurate. In order to recognize, detect and count lepidopteran pests quickly and accurately in vegetable garden, recognition model, detect and count model were introduced based on deep convolutional neural network, respectively. According to five kinds of common lepidopteran pests, such as Pieris rapae, Helicoverpa armigera, Spodoptera exigua, Plutella xylostella and Spodoptera litura, this paper built recognition image dataset, detection and counting image dataset and then carried out experiments, the average recognition accuracy could reach 94.5% and the mean average precision of detection was 76.6%, the comparison to conventional methods proved that our method was preeminent. The experiment results showed that the proposed method was a feasible way to recognize, detect and count vegetable lepidopteran pests, and it reached the level of practical application.
Keywords:lepidoptera pest  auto recognition  detection and counting  deep learning  convolutional neural network  
本文献已被 CNKI 等数据库收录!
点击此处可从《中国农业科技导报》浏览原始摘要信息
点击此处可从《中国农业科技导报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号