杜凯旋;王亮;王勇;车向红;

• 1:中国测绘科学研究院

摘要(Abstract)：

针对目前互联网中存在有大量地图图片识别精度低、分类模型训练困难等问题,该文提出了一种基于主动学习和卷积神经网络的地图图片识别方法,通过主动学习算法优化地图/非地图图片样本数量及分类结构,可使用少量的训练样本以及人工投入,来获取基于卷积神经网络的高精度地图图片识别模型。结果表明:基于主动学习方法将地图/非地图图片样本类型划分为17类时,地图图片识别精度最高,约为95.01%。本文地图识别方法可有效地推动图像地理信息挖掘、地图审查及地理信息监管等领域的技术进步。

关键词(KeyWords)： 卷积神经网络;主动学习;地图图片;图片识别

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(41901379);; 国家重点研发计划项目(2017YFB0503502,2017YFB0503601)

作者(Author): 杜凯旋;王亮;王勇;车向红;

Email:

DOI: 10.16251/j.cnki.1009-2307.2020.07.021

参考文献(References)：

• [1] 杨云.基于地图及遥感影像的地理信息提取研究[D].郑州:信息工程大学,2008.(YANG Yun.Geographic information extraction based on maps and remote sensing images[D].Zhengzhou:Information Engineering University,2008.)
• [2] 黄龙.一种快速挖掘互联网问题地图图片的方法[J].测绘与空间地理信息,2017,40(11):92-93,96.(HUANG Long.Research on the method of fast mining internet problem map picture[J].Geomatrics & Spatial Information Technology,2017,40(11):92-93,96.)
• [3] 陈会仙,程滔,白敬辉,等.国家互联网地理信息监管现状及趋势研究[J].测绘科学,2016,41(10):153-158.(CHEN Huixian,CHENG Tao,BAI Jinghui,et al.Study of status of national Internet geographic information supervision and its development trend[J].Science of Surveying and Mapping,2016,41(10):153-158.)
• [4] 樊凡渊.对地图的监管不能仅限于纸质版[N/OL].中国新闻出版广电报.(2019-02-28)[2019-03-26].http://kns.cnki.net/kns/detail/detail.aspx?FileName=XWCB201902280034&DbName=CCNDPTEM.(FAN Fanyuan.The supervision of maps should not be limited to paper editions[N/OL].China News and Publication Radio and TV News.(2019-02-28)[2019-03-26].http://kns.cnki.net/kns/detail/detail.aspx?FileName=XWCB201902280034&DbName=CCNDPTEM.)
• [5] 师亚丽.互联网地图管理的难点与发展空间浅析[J].西北人文科学评论,2014,7:216-221.(SHI Yali.Analysis on the difficulties and development space of internet map management[J].Comments on the Humanities in Northwest China,2014,7:216-221.)
• [6] 田艳玲,张维桐,张锲石,等.图像场景分类技术综述[J].电子学报,2019,47(4):915-926.(TIAN Yanling,ZHANG Weitong,ZHANG Qieshi,et al.Review on image scene classification technology[J].Acta Electronica Sinica,2019,47(4):915-926.)
• [7] 尼加提·卡斯木,师庆东,刘素红.基于卷积网络的沙漠腹地绿洲植物群落自动分类方法[J].农业机械学报,2019,50(1):217-225.(NIJAT Kasim,SHI Qingdong,LIU Suhong.Automatic classification method of oasis plant community in desert hinterland based on CNN[J].Transactions of the Chinese Society for Agricultural Machinery,2019,50(1):217-225.)
• [8] 何小飞,邹峥嵘,陶超,等.联合显著性和多层卷积神经网络的高分影像场景分类[J].测绘学报,2016,45(9):1073-1080.(HE Xiaofei,ZOU Zhengrong,TAO Chao,et al.Combined saliency with multi-convolutional neural network for high resolution remote sensing[J].Acta Geodaetica et Cartographica Sinica,2016,45(9):1073-1080.)
• [9] 马飞.基于Windows的扫描等高线地图识别系统[J].武汉测绘科技大学学报,1995(3):228-233.(MA Fei.Scanning contour map recognition system based on Windows[J].Geomatics and Information Science of Wuhan University,1995(3):228-233.)
• [10] 杨启和,朱文忠,黄文骞.地图模式识别研究的介绍与展望[J].测绘科技通讯,1996(3):16-20.(YANG Qihe,ZHU Wenzhong,HUANG Wenjian.The introduction and prospect of map pattern recognition[J].Geomatics Technology and Equipment,1996(3):16-20.)
• [11] 王雪冰,郭庆胜,王勇,等.地图图片的特征提取与自动识别方法[J].测绘与空间地理信息,2019,42(9):28-32.(WANG Xuebing,GUO Qingsheng,WANG Yong,et al.Feature extraction and automatic recognition method for map images[J].Geomatics & Spatial Information Technology,2019,42(9):28-32.)
• [12] 李超琪.基于深度卷积神经网络及HMAX模型的人脸年龄分类方法[D].上海:东华大学,2017.(LI Chaoqi.Age classification methods based on deep convolutional network and HAMX model[D].Shanghai:Donghua University,2017.)
• [13] 何海威,钱海忠,谢丽敏,等.立交桥识别的CNN卷积神经网络法[J].测绘学报,2018,47(3):385-395.(HE Haiwei,QIAN Haizhong,XIE Limin,et al.Interchange recognition method based on CNN[J].Acta Geodaetica et Cartographica Sinica,2018,47(3):385-395.)
• [14] 王景中,杨源,何云华.基于多分类和 ResNet 的不良图片识别框架[J].计算机系统应用,2018,27(9):100-106.(WANG Jingzhong,YANG Yuan,HE Yunhua.Pornographic images recognition framework based on multi-classification and ResNet[J].Computer Systems & Applications,2018,27(9):100-106.)
• [15] 田德宇,张耀南,赵国辉,等.基于卷积神经网络的遥感沙漠绿地提取方法[J].遥感技术与应用,2018,33(1):151-157.(TIAN Deyu,ZHANG Yaonan,ZHAO Guohui,et al.Convolutional neural network for remote sensing plant cover extracting[J].Remote Sensing Technology and Application,2018,33(1):151-157.)
• [16] 刘华春.卷积神经网络在车牌识别中的应用研究[J].计算机技术与发展,2019,29(4):128-132.(LIU Huachun.Research on application of convolutional neural network in License plate recognition[J].Computer Technology and Development,2019,29(4):128-132.)
• [17] 陈益民,白勇,黎传琛.基于卷积神经网络的快速人脸检测算法[J].海南大学学报(自然科学版),2018,36(4):324-329.(CHEN Yimin,BAI Yong,LL Chuanchen.A fast face detection algorithm based on convolutional neural network[J].Natural Science Journal of Hainan University,2018,36(4):324-329.)
• [18] 崔腾腾,刘纪平,罗安.一种卷积神经网络的地图图片智能识别方法[J].测绘科学,2019,44(1):118-123.(CUI Tengteng,LIU Jiping,LUO An.Intelligent identification method of network map images based on convolutional neural network[J].Science of Surveying and Mapping,2019,44(1):118-123.)
• [19] 周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.(ZHOU Feiyan,JIN Linpeng,DONG Jun.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251.)
• [20] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).[S.l.]:IEEE Computer Society,2016.
• [21] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017(99):2999-3007.
• [22] 刘康,钱旭,王自强.主动学习算法综述[J].计算机工程与应用,2012,48(34):1-4.(LIU Kang,QIAN Xu,WANG Ziqiang.Survey on active learning algorithms[J].Computer Engineering andApplications,2012,48(34):1-4.)
• [23] 李会,马红杉,吴迪,等基于主动学习的分类器融合算法[J].微计算机信息,2007,23(2):302-303.(LI Hui,MA Hongshan,WU Di,et al.The merging arithm etic of classifier based on active learning[J].Microcomputer Information,2007,23(2):302-303.)
• [24] CHAPELL E.Semi-supervised learning[J].Journal of the Royal Statistical Society,2006,172(2):530-530.
• [25] 王天珩,张轶.基于多应用场景的主动学习算法研究[J].现代计算机,2018(10):40-43.(WANG Tianheng,ZHANG Yi.Research on an active learning algorithm based on multi-application scenes[J].Modern Computer,2018(10):40-43.)