基于人工神经网络的多特征因子路网匹配算法

doi:10.3724/SP.J.1047.2016.01153

摘要/Abstract

摘要：

在综合考虑多个特征因子的线要素匹配时,根据经验知识确定各特征因子的权值会造成人为误差。针对该问题,本文提出了基于人工神经网络的多特征因子路网匹配算法,根据线要素的几何和拓扑特性选取长度、方向、形状、距离及拓扑5个特征因子的相似度作为路网匹配参考因子。首先,分别在参考图层和待匹配图层中选取样本数据组成样本对,计算样本数据的5个特征因子相似度,用样本数据的5个特征因子相似度和样本的匹配度组成学习模式对;然后,利用BP神经网络的误差反向传播机制自动学习调整各神经层之间的连接权值;最后,输入全部数据,计算参考图层的弧段和待匹配图层的弧段间的匹配度,实现综合多特征因子的路网匹配。实验结果表明,利用人工神经网络进行综合多特征因子的路网匹配可以提高匹配效率和匹配准确度。

关键词: BP神经网络, 路网, 匹配, 多特征因子

Abstract:

The matching of line features is the premise and key technology of map conflation, which also plays an important role in change detection, digital map updating and map registration. To improve the accuracy of line feature matching, multiple geometrical characteristics and topological characteristics should be considered, such as distance, length, shape, orientation, node-degree and so on. It may produce errors when the weights of the characteristic factors are determined merely by experience. Therefore, to avoid this problem, a road network matching method considering multiple geometric characteristics based on the artificial neural network is proposed. Length similarity, orientation similarity, shape similarity, distance similarity and topological similarity (which is the node-degree similarity) are the five feature similarities discussed in this paper. They are the neurons of the input layer in BP (Back Propagation) neural network. To implement this method, first of all, samples of the reference layer and the adjustment layer are selected and the values of the five similarities for these samples are calculated. Secondly, the five feature similarities of these samples incorporating the matching rate are serving as the learning modes and are used to train the BP neural network. And the connection weights between the neural network and the threshold values of neurons are adjusted automatically. Next, buffers of the reference layer's arcs are generated. Arcs within the adjustment layer's buffers are defined as the candidate arcs. At last, the five feature similarities between each arc within the reference layer and its candidate arcs are calculated. Then, these similarities are put into the BP neural network to compute the matching rates of these arcs. If the matching rate is greater than 0.8, the relevant two arcs are regarded as an accurate matching; if the matching rate is less than 0.5, the relevant two arcs are considered to be bad matching; otherwise, they are considered to be less accurate matching which needs the interference of manual assistance to determine whether the two arcs match or not. Result shows that the adoption of BP neural network makes the road networks matching more efficient and accurate, and it avoids the assignment of proper weights to different geometrical characteristics at the same time.

Key words: BP neural network, road network, matching, multiple geometric characteristics

郭宁宁, 盛业华, 黄宝群, 吕海洋, 张思阳. 基于人工神经网络的多特征因子路网匹配算法[J]. 地球信息科学学报, 2016, 18(9): 1153-1159.DOI:10.3724/SP.J.1047.2016.01153

GUO Ningning,SHENG Yehua,HUANG Baoqun,LYU Haiyang,ZHANG Siyang. Road Network Matching Considering Multiple Geometric Characteristics Based on the Artificial Neural Network[J]. Journal of Geo-information Science, 2016, 18(9): 1153-1159.DOI:10.3724/SP.J.1047.2016.01153

图/表 7

图1

图2

图3

图4

表1

表2

图5

参考文献 27

[1]	张桥平,李德仁,龚健雅.地图合并技术[J].测绘通报,2001(7):6-8. doi: 10.3969/j.issn.0494-0911.2001.07.003
	[ Zhang Q P, Li D R, Gong J Y.Map conflation[J]. Bulletin of Surveying and Mapping, 2001,7:6-8. ] doi: 10.3969/j.issn.0494-0911.2001.07.003
[2]	李德仁,龚健雅,张乔平.论地图数据库合并技术[J].测绘科学,2004,29(1):1-4. doi: 10.3771/j.issn.1009-2307.2004.01.001
	[ Li D R, Gong J Y, Zhang Q P.On the conflation of geographic databases[J]. Science of Surveying and Mapping, 2004,29(1):1-4. ] doi: 10.3771/j.issn.1009-2307.2004.01.001
[3]	陈军,蒋捷,李志林,等.基础地理数据库的持续更新问题[J].地理信息世界,2004,2(5):1-5. doi: 10.3969/j.issn.1672-1586.2004.05.001
	[ Chen J, Jiang J, Li Z L, et al.Key issues of continuous updating of geo-spatial databases[J]. Geomatics World, 2004,2(5):1-5. ] doi: 10.3969/j.issn.1672-1586.2004.05.001
[4]	Goodchild M F. Conflation: combining geographical information[EB/OL]. .
[5]	Kang H. Spatial data integration: a case study of map conflation with census bureau and local government data[EB/OL]. .
[6]	尹川,王艳慧.路网增量更新中基于OSTU的目标几何匹配阈值计算[J].武汉大学学报·信息科学版,2014,39(9):1061-1067. doi: 10.13203/j.whugis20130575
	[ Yin C, Wang Y H.Target geometry matching threshold in incremental updating of road networks based on OSTU[J]. Geomatics and Information Science of Wuhan University, 2014,39(9):1061-1067. ] doi: 10.13203/j.whugis20130575
[7]	徐敬海,李清泉.基于时态GIS的导航电子地图增量更新研究[J].武汉大学学报·信息科学版,2008,33(11):1182-1185.
	[ Xu J H, Li Q Q.Navigable digital map data increment update based on temporal GIS[J]. Geomatics and Information Science of Wuhan University, 2008,33(11):1182-1185. ]
[8]	Velaga N R, Quddus M A, Bristow A L.Developing an enhanced weight-based topological map-matching algorithm for intelligent transport systems[J]. Transportation Research Part C, 2009,17:672-683. doi: 10.1016/j.trc.2009.05.008
[9]	Zhang M.Methods and implementations of road-network matching[D]. Doktorarbeit, 2009.
[10]	童小华,邓愫愫,史文中.基于概率的地图实体匹配方法[J].测绘学报,2007,36(2):210-217. doi: 10.3321/j.issn:1001-1595.2007.02.017
	[ Tong X H, Deng S S, Shi W Z.A probabilistic theory-based matching method[J]. Acta Geodaetica et Cartographica Sinica, 2007,36(2):210-217. ] doi: 10.3321/j.issn:1001-1595.2007.02.017
[11]	鲁伟,谢顺平,邓敏,等.多源空间数据间不一致性研究现状及其进展[J].测绘科学,2009,34(4):57-60. [ Lu W, Xie S P, Deng M, et al. Research status and advance on multi-source spatial data inconsistency[J]. Science of Surveying and Mapping, 2009,34(4):57-60. ]
[12]	安晓亚,孙群,尉伯虎.利用相似性度量的不同比例尺地图数据网状要素匹配算法[J].武汉大学学报·信息科学版, 2012,37(2):224-228.
	[ An X Y, Sun Q, Yu B H.Feature matching from network data at different scales based on similarity measure[J]. Geomatics and Information Science of Wuhan University, 2012,37(2):224-228. ]
[13]	郭黎,李宏伟,张泽建,等.道路网信息投影匹配方法研究[J].武汉大学学报·信息科学版,2013,38(9):1113-1117.
	[ Guo L, Li H W, Zhang Z J, et al.Geometry matching method for transportation road network data based on projection[J]. Geomatics and Information Science of Wuhan University, 2013,38(09):1113-1117. ]
[14]	安晓亚,刘平芝,杨云,等.一种线状要素几何相似性度量方法及其应用[J].武汉大学学报·信息科学版,2015,40(9):1225-1229. doi: 10.13203/j.whugis20130495
	[ An X Y, Liu P Z, Yang Y, et al.A geometric similarity measurement method and applications to linear feature[J]. Geomatics and Information Science of Wuhan University, 2015,40(9):1225-1229. ] doi: 10.13203/j.whugis20130495
[15]	赵彬彬,邓敏,刘慧敏,等.多尺度地图的水系面目标与线目标匹配方法与实验[J].地球信息科学学报,2011,13(3):361-366. doi: 10.3724/SP.J.1047.2011.00361
	[ Zhao B B, Deng M, Liu H M, et al.An approach to matching area objects and line objects of river system in multi-scale maps[J]. Journal of Geo-Information Science, 2011,13(3):361-366. ] doi: 10.3724/SP.J.1047.2011.00361
[16]	郭黎,崔铁军,张斌.道路网数据变化检测与融合处理技术[J].地理信息世界,2011(5):29-32. doi: 10.3969/j.issn.1672-1586.2011.05.008
	[ Guo L, Cui T J, Zhang B.An recognition and fusion technology of transportation road network data[J]. Geomatics World, 2011,5:29-32. ] doi: 10.3969/j.issn.1672-1586.2011.05.008
[17]	Yang H, Cheng S, Jiang H, et al.An enhanced weight-based topological map matching algorithm for intricate urban road network[J]. Procedia-Social and Behavioral Sciences, 2013,96:1670-1678. doi: 10.1016/j.sbspro.2013.08.189
[18]	Beeri C, Kanza Y, Safra E, et al.Object fusion in geographic information systems[C]. Proceedings of the 30th Conference on Very Large Data Bases, 2004:816-827.
[19]	赵东保,盛业华.全局寻优的矢量道路网自动匹配方法研究[J].测绘学报,2010,39(4):416-421.
	[ Zhao D B, Sheng Y H.Research on automatic matching of vector road networks based on global optimization[J]. Acta Geodaetica et Cartographica Sinica, 2010,39(4):416-421 ]
[20]	张云菲,杨必胜,栾学晨.利用概率松弛法的城市路网自动匹配[J].测绘学报,2012,41(6):933-939.
	[ Zhang Y F, Yang B S, Luan X C.Automated matching urban road networks using probabilistic relaxation[J]. Acta Geodaetica et Cartographica Sinica, 2012,41(6):933-939. ]
[21]	Yang B, Zhang Y, Luan X.A probabilistic relaxation approach for matching road networks[J]. International Journal of Geographical Information Science, 2013,27(2):319-338. doi: 10.1080/13658816.2012.683486
[22]	苏海滨,王光政,王继东.基于模糊神经网络的地图匹配算法[J].北京科技大学学报,2012,34(1):43-47.
	[ Su H B, Wang G Z, Wang J D.Map matching algorithm based on fuzzy neural networks[J]. Journal of University of Science and Technology Beijing, 2012,34(1):43-47. ]
[23]	郭泰圣,张新长,梁志宇.神经网络决策树的矢量数据变化信息快速识别方法[J].测绘学报,2013,42(6):937-944.
	[ Guo T S, Zhang X C, Liang Z Y.Research on change information recognition method of vector data based on neural network decision tree[J]. Acta Geodaetica et Cartographica Sinica, 2013,42(6):937-944. ]
[24]	龚敏霞,袁赛,储征伟,等.顾及多空间相似性的地下管线数据匹配[J].测绘学报,2015,44(12):1392-1400. doi: 10.11947/j.AGCS.2015.20150207
	[ Gong M X, Yuan S, Chu Z W, et al.Underground pipeline data matching considering multiple spatial similarities[J]. Acta Geodaetica et Cartographica Sinica, 2015,44(12):1392-1400. ] doi: 10.11947/j.AGCS.2015.20150207
[25]	许俊奎,武芳,魏慧峰.人工神经网络在居民地面状匹配中的应用[J].测绘科学技术学报,2014,30(3):293-298.
	[ Xu J K, Wu F, Wei H F.Areal settlements matching algorithm based on artificial neural network technique[J]. Journal of Geomatics Science and Technology, 2014,30(3):293-298. ]
[26]	海金. 神经网络原理[M].叶世伟,史忠植,译.北京:机械工业出版社, 2004:229-247. [ Haykin S. Neural networks[M]. Beijing: China Machine Press,. ]
[27]	Saalfeld A.Automated map conflation[D]. Washington D C, US: The Unversity of Maryland, 1993.

待匹配弧段	参考弧段	长度相似度	方向相似度	形状相似度	距离相似度	拓扑相似度	匹配度
A129	B109	0.33	0.63	0.29	0.78	1.00	0.59
A207	B139	0.70	0.91	0.61	0.35	0.75	0.73
A123	B80	0.87	0.90	0.56	0.26	1.00	0.76
A79	B49	0.42	0.98	0.41	0.26	1.00	0.71

实验组	数据总量	精确匹配	比例/（%）	非精确匹配	比例/（%）	不匹配	比例/（%）	漏匹配	误匹配	比例/（%）	准确率/（%）
实验1	207	168	80.77	6	2.88	34	16.15	0	2	0.96	99.04
实验2	332	315	94.88	16	4.82	1	0.30	0	6	1.81	98.19