Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Science of Surveying and Mapping》 2019-01
Add to Favorite Get Latest Update

Intelligent identification method of network map images based on convolutional neural network

CUI Tengteng;LIU Jiping;LUO An;Chinese Academy of Surveying and Mapping;  
Aiming at the problem that it is difficult to quickly search the map images in the network and to mark them effectively at present,this paper proposed the use of deep web search engine to continuously obtain network images.In the case of fewer marked samples,network map images could be quickly and effectively drawn out using an actively learning-based sample diffusion algorithm,a network map picture automatic semantic indexing method based on the convolutional neural network(CNN).The recognition accuracy of the map image was93.64%.The difficulty to acquire sample and image identification in an artificial method was overcame,the problem of intelligent extraction of network map images under rare samples was effectively solved.It could provide effective guarantee for geographic information mining and problem map supervision,and so on.
【Fund】: 国家重点研发计划项目(2017YFB0503502 2017YFB0503601);; 中国测绘科学研究院基本科研业务费项目(7771817)
【CateGory Index】: TP391.41;TP183
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved