回转库档案实时定位中基于鱼眼镜头图像的处理识别技术
详细信息    查看全文 | 推荐本文 |
  • 英文篇名:Processing and Recognition Technology Based on Fisheye Lens Image in Real-Time Positioning of Rotary Library Files
  • 作者:成杰 ; 叶文武 ; 徐寅林
  • 英文作者:Cheng Jie;Ye Wenwu;Xu Yinlin;School of Physics and Technology,Nanjing Normal University;
  • 关键词:鱼眼镜头 ; 图像识别 ; 卷积神经网络 ; 档案识别
  • 英文关键词:fisheye lens;;image identification;;convolutional neural network;;file identification
  • 中文刊名:NJSE
  • 英文刊名:Journal of Nanjing Normal University(Engineering and Technology Edition)
  • 机构:南京师范大学物理科学与技术学院;
  • 出版日期:2019-06-20
  • 出版单位:南京师范大学学报(工程技术版)
  • 年:2019
  • 期:v.19;No.74
  • 基金:江苏省研究生科研与实践创新计划项目(SJCX19_0186)
  • 语种:中文;
  • 页:NJSE201902011
  • 页数:7
  • CN:02
  • ISSN:32-1684/T
  • 分类号:81-87
摘要
介绍了一种基于鱼眼镜头图像处理识别的回转库档案盘点技术,采用鱼眼镜头宽视角拍摄回转库中排列成行的带有数字标签的多本档案;利用图像矫正、分割技术处理得到标签上的每一个数字图;由构造的卷积神经网络识别出图像内容;以标签数字列在镜头中的位序实现档案定位.该技术实时性较好,识别准确且成本低.
        This paper introduces a rotary library file inventory technology based on fisheye lens image processing and recognition. A wide angle of view of the fisheye lens is used to capture multiple files with digital labels arranged in rows. Image correction and segmentation technology are used to process each label. Image content is recognized by constructed Convolutional Neural Network. The file position is realized by the bit sequence of the label number listed in the lens. The technology has good real-time performance,accurate identification and low cost.
引文
[1]王璐瑶.基于RFID技术的实体档案管理研究[J].机电兵船档案,2018(2):39-41.WANG L Y.Research on entity archive management based on RFID technology[J].Electromechanical ship file,2018(2):39-41.(in Chinese)
    [2]秦荪榛.基于物联网技术的档案馆实时自动盘库控制系统[D].南京:南京师范大学,2014.QIN S Z.Archives real-time automatic library control system based on internet of things technology[D].Nanjing:Nanjing Normal University,2014.(in Chinese)
    [3]郭盛威.鱼眼图像视频校正方法研究[D].武汉:武汉工程大学,2017.GUO S W.Fisheye image video correction method research[D].Wuhan:Wuhan Institute of Technology,2017.(in Chinese)
    [4]冯为嘉,张宝峰,曹作良.基于鱼眼镜头的全方位视觉参数标定与畸变矫正[J].天津大学学报,2011,44(5):417-424.FENG W J,ZHANG B F,CAO Z L.Omni-directional vision parameter calibration and rectification based on fisheye lens[J].Journal of Tianjin university,2011,44(5):417-424.(in Chinese)
    [5]朱安娜.基于卷积神经网络的场景文本定位及多方向字符识别研究[D].武汉:华中科技大学,2016.ZHU A N.Research on convolution neural network-based scene text detection and multi-orientational character recognition[D].Wuhan:Huazhong University of Science and Technology,2016.(in Chinese)
    [6]余承波.基于支持向量机(SVM)的车牌识别[D].淮南:安徽理工大学,2018.YU C B.License plate recognition based on support vector machine(SVM)[D].Huainan:Anhui University of Science and Technology,2018.(in Chinese)
    [7]靳涛,张永爱.Tensor Flow平台下基于深度学习的数字识别[J].信息技术与网络安全,2018,37(4):74-78.JIN T,ZHANG Y A.Digital recognition based on deep learning in Tensor Flow platform[J].Information technology and network security,2018,37(4):74-78.(in Chinese)
    [8]张立.基于卷积神经网络的车牌字符识别方法研究[D].镇江:江苏大学,2015.ZHANG L.Research of license plate recognition based on convolutional neural network[D].Zhenjiang:Jiangsu University,2015.(in Chinese).

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700