引用本文
  • 张扬虎,白琳,陈峰练,张茜,李淘深.基于深度卷积神经网络技术的棉花顶芽探测技术[J].广西科学,2021,28(3):257-264.    [点击复制]
  • ZHANG Yanghu,BAI Lin,CHEN Fenglian,ZHANG Qian,LI Taoshen.Detection of Cotton Top Buds Based on Deep Convolutional Neural Networks Technology[J].Guangxi Sciences,2021,28(3):257-264.   [点击复制]
【打印本页】 【在线阅读全文】【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览 49次   下载 51 本文二维码信息
码上扫一扫!
基于深度卷积神经网络技术的棉花顶芽探测技术
张扬虎1, 白琳1,2, 陈峰练1, 张茜3, 李淘深1,2
0
(1.广西大学计算机与电子信息学院, 广西南宁 530004;2.广西高校并行与分布式计算技术重点实验室, 广西南宁 530004;3.广西医科大学第一附属医院, 广西南宁 530021)
摘要:
建立一种改进深度学习模型,用于农业自动化检测和识别棉花顶芽,以提高棉花劳作工作效率。通过把深度网络模型ResNet-101融入到基于深度学习(Deep Learning,DL)机制的感兴趣区域的目标检测算法Faster RCNN中,得到统一的多结构层次的改进深度学习模型。对比实验验证结果表明,相较于传统Faster RCNN模型,该模型在棉花顶芽探测和识别性能上有较大的提升。本研究提出的改进深度学习模型取得了比较好的平均精度,为棉花顶芽的探测和识别提出新的解决方案,为农业生产智能化提供新的思路。
关键词:  深度学习  神经网络  目标检测  Faster RCNN  农业智能化  棉花顶芽探测
DOI:10.13656/j.cnki.gxkx.20210830.011
投稿时间:2020-12-15
基金项目:国家自然科学基金项目(61966003)和广西自然科学基金项目(2020GXNSFAA159171)资助。
Detection of Cotton Top Buds Based on Deep Convolutional Neural Networks Technology
ZHANG Yanghu1, BAI Lin1,2, CHEN Fenglian1, ZHANG Qian3, LI Taoshen1,2
(1.School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, 530004, China;2.Guangxi Colleges and Universities Key Laboratory of Parallel and Distributed Computing Technology, Nanning, Guangxi, 530004, China;3.The First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, 530021, China)
Abstract:
An improved deep learning model was established to detect and identify cotton top buds in agricultural automation, so as to improve the work efficiency of cotton. By integrating the deep network model ResNet-101 into the target detection algorithm Faster RCNN based on the deep learning mechanism, a unified multi-structured and improved deep learning model is obtained. The results of comparative experiments show that compared with the traditional Faster RCNN model, this model has greatly improved the detection and recognition performance of cotton top bud. The improved deep learning model proposed in this study has achieved good average accuracy, which provides a new solution for the detection and identification of cotton top buds and a new idea for the intelligent agricultural production.
Key words:  deep learning  neural networks  object detection  Faster RCNN  intelligent agriculture  detection of cotton top buds

用微信扫一扫

用微信扫一扫