张鸿刚,李成名,武鹏达,殷勇,郭曼

• 1:山东理工大学
• 2:中国测绘科学研究院

摘要(Abstract)：

针对传统识别方法多依赖于人工设计的低层次特征,未能有效描述复杂交叉路口的细节特征,导致识别类型有限、精度不高的问题,该文提出一种基于GoogLeNet神经网络的复杂交叉路口识别方法:首先利用交叉路口结点密集的特征,构建Delaunay三角网进行点群聚类,初步确定复杂交叉路口中心位置及空间范围;其次,在全国范围内选取39个重要城市路网作为训练样本,并充分利用矢量数据结构优势,以简化、旋转、镜像等方式丰富样本类型及容量;最后,针对其局部特征丰富的特点,选取GoogLeNet神经网络进行训练,以学习其高层次模糊性特征。以天津OSM城市路网为例的实验表明,本文方法能够有效识别复杂交叉路口,且明显提高了识别的精度和准度,具有较强地泛化性和抗干扰性。

关键词(KeyWords)： 复杂交叉路口识别;矢栅结合;Delaunay三角网聚类;GoogLeNet神经网络

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金面上项目(41871375);; 中国测绘科学研究院基本科研业务费项目(AR1909,AR1916,AR1917,AR1935)

作者(Author): 张鸿刚,李成名,武鹏达,殷勇,郭曼

DOI: 10.16251/j.cnki.1009-2307.2020.10.027

参考文献(References)：

• [1] SCHEIDER S,KUHN W.Affordance-based categorization of road network data using a grounded theory of channel networks[J].International Journal of Geographical Information Science,2010,24(8):1249-1267.
• [2] SCHEIDER S,POSSIN J.Affordance-based individuation of junctions in open street map[J].Journal of Spatial Information Science,2012(4):31-56.
• [3] THOMSON R,BROOKS R.Generalization of geographical networks[M]//Generalization of Geographic Information:Cartographic Modelling and Applications.Amsterdam:Elsevier Science & Technology Books,2007:255-267.
• [4] YANG B,LUAN X,LI Q.An adaptive method for identifying the spatial patterns in road networks[J].Computers,Environment and Urban Systems,2010,34(1):40-48.
• [5] MACKANESS W A,MACKECHNIE G A.Automating 130 the detection and simplification of junctions in road networks[J].GeoInformatica,1999,3(2):185-200.
• [6] 徐柱,蒙艳姿,李志林,等.基于有向属性关系图的典型道路交叉口结构识别方法[J].测绘学报,2011,40(1):125-131.(XU Zhu,MENG Yanzi,LI Zhilin,et al.Recognition of structures of typical road junctions based on directed attributed relational graph[J].Acta Geodaetica et Cartographica Sinica,2011,40 (1):125-131.)
• [7] 王骁,钱海忠,丁雅莉,等.采用拓扑关系与道路分类的立交桥整体识别方法[J].测绘科学技术学报,2013,30(3):324-328.(WANG Xiao,QIAN Haizhong,DING Yali,et al.The integral identification method of cloverleaf junction based on topology and road classification[J].Journal of Geomatics Science and Technology,2013,30 (3):324-328.)
• [8] 付仲良,杨元维,高贤君,等.道路网多特征匹配优化算法[J].测绘学报,2016,45(5):608-615.(FU Zhongliang,YANG Yuanwei,GAO Xianjun,et al.An optimization algorithm for multi-characteristics road network matching[J].Acta Geodaetica et Cartographica Sinica,2016,45(5):608-615.)
• [9] 马超,孙群,陈换新,等.利用路段分类识别复杂道路交叉口[J].武汉大学学报(信息科学版),2016,41(9):1232-1237.(MA Chao,SUN Qun,CHEN Huanxin,et al.Recognition of road junctions based on road classification method[J].Geomatics and Information Science of Wuhan University,2016,41(9):1232-1237.)
• [10] LECUN Y,BOSER B,DENKER J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural Computation,1989,1(4):541-551.
• [11] WEI Y,WANG Z,XU M.Road structure refined CNN for road extraction in aerial image[J].IEEE Geoscience and Remote Sensing Letters,2017,14(5):709-713.
• [12] BADRINARAYANAN V,KENDALL A,CIPOLLA R.Segnet:a deep convolutional encoder-decoder architecture for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481-2495.
• [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] MUNDUR P,RAO Y,YESHA Y.Key frame based video summarization using Delaunay clustering[J].International Journal on Digital Libraries,2006,6(2):219-232.
• [15] STADLER C,LEHMANN J,H?FFNER K,et al.Linkedgeodata:a core for a web of spatial open data[J].Semantic Web Journal,2012,3(4):333-354.
• [16] SZEGEDY C,LIU W,JIA Y,et al.Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE,2015:1-9.
• [17] SZEGEDY C,VANHOUCHE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE,2016:2818-2826.
• [18] LEE S G,SUNG Y,KIM Y G,et al.Variations of AlexNet and GoogLeNet to improve Korean character recognition performance[J].Journal of Information Processing Systems,2018,14(1):205-217.
• [19] LI C M,DAI Z X,YIN Y,et al.A method for the extraction of partition lines from long and narrow patches that account for structural features[J].Transactions in GIS,2019,23(2):349-364.