车辆轨迹与遥感影像多层次融合的道路交叉口识别

doi:10.11947/j.AGCS.2021.20210255

摘要/Abstract

摘要： 交叉口是构成道路网络的基础与核心要素，起到了连接道路和承载转向的重要作用。在城市路网中，交叉口不仅数量众多、形态多样，而且结构复杂、大小不一。单一数据源对于道路交叉口的描述能力有限，难以做到道路交叉口的全面、精确识别。为此，本文设计了一种从车辆轨迹与遥感影像中识别道路交叉口的多元集成方法。首先，集成形态学处理、密度峰值聚类与张量投票提取种子交叉口，将其作为小样本集；然后，据此采用协同训练机制，分别构建基于深度卷积网络，面向车辆轨迹与遥感影像的交叉口分类器；最后，综合两模型优点，形成道路交叉口的集成分类模型。本文方法在多个层次上融合车辆轨迹与遥感影像关于交叉口的互补性描述特征，提出半监督式交叉口提取技术，无须人工标注即可有效识别复杂多样的道路交叉口。基于武汉市出租车轨迹和遥感影像的试验表明，本文方法在无人工标注样本的前提下，道路交叉口提取的准确率超过93%，召回率达到87%。

关键词: 道路交叉口, 车辆轨迹, 遥感影像, 协同训练, 自动标注

Abstract: Road intersections are important components of a road network, which are not only numerous and diverse in shape, but also complex in structure and different in size. It is difficult to recognize comprehensive and accurate road junctions based on single data source, as its limited describe information. To this end, this paper designs a multiple integration method to identify road intersections from vehicle trajectories and remote sensing images. Firstly, based on the unsupervised idea, a method combining morphological processing, density peak clustering and tensor voting is proposed to extract the seed intersections, which is regarded as a small sample set. Based on it two intersection classifiers based on deep convolution network and oriented to vehicle trajectories and remote sensing images are constructed by using collaborative training mechanism, and finally, the advantages of the two models are combined to form an integrated classification model of road intersections. In this paper, a semi supervised intersection extraction technology is proposed by fusing the complementary description features of vehicle trajectories and remote sensing images on multiple levels, which can effectively identify complex and diverse road intersections without manual labeling. Experiments based on Wuhan taxi trajectories and remote sensing images show that the accuracy of this method is more than 93% and the recall rate is 87% without manually labeled samples.

Key words: road intersection, vehicles trajectory, remote sensing image, co-training, automatic activation

中图分类号:

P208

李雅丽, 向隆刚, 张彩丽, 吴华意, 龚健雅. 车辆轨迹与遥感影像多层次融合的道路交叉口识别[J]. 测绘学报, 2021, 50(11): 1546-1557.

LI Yali, XIANG Longgang, ZHANG Caili, WU Huayi, GONG Jianya. Road intersection recognition based on a multi-level fusion of vehicle trajectory and remote sensing image[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(11): 1546-1557.

参考文献

[1] 万子健, 李连营, 杨敏, 等. 车辆轨迹数据提取道路交叉口特征的决策树模型[J]. 测绘学报, 2019, 48(11): 1391-1403. DOI:10.11947/j.AGCS.2019.20190011. WAN Zijian, LI Lianying, YANG Min, et al. Decision tree model for extracting road intersection feature from vehicle trajectory data[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(11):1391-1403. DOI:10.11947/j.AGCS.2019.20190011.
[2] LI Yali, XIANG Longgang, ZHANG Caili, et al. Fusing taxi trajectories and RS images to build road map via DCNN[J]. IEEE Access, 2019, 7:161487-161498.
[3] ZHANG Caili, LI Yali, XIANG Longgang, et al. Generating road networks for old downtown areas based on crowd-sourced vehicle trajectories[J]. Sensors (Basel), 2021, 21(1): 235.
[4] 唐炉亮, 牛乐, 杨雪, 等. 利用轨迹大数据进行城市道路交叉口识别及结构提取[J]. 测绘学报, 2017, 46(6): 770-779. DOI:10.11947/j.AGCS.2017.20160614. TANG Luliang, NIU Le, YANG Xue, et al. Urban intersection recognition and construction based on big trace data[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(6): 770-779. DOI:10.11947/j.AGCS.2017.20160614.
[5] YANG Xue, TANG Luliang, NIU Le, et al. Generating lane-based intersection maps from crowdsourcing big trace data[J]. Transportation Research Part C:Emerging Technologies, 2018, 89:168-187.
[6] WANG Jing, RUI Xiaoping, SONG Xianfeng, et al. A novel approach for generating routable road maps from vehicle GPS traces[J]. International Journal of Geographical Information Science, 2015, 29(1): 69-91.
[7] FATHI A, KRUMM J. Detecting road intersections from GPS traces[M]//Geographic Information Science. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010: 56-69.
[8] DAVIES J J, BERESFORD A R, HOPPER A. Scalable, distributed, real-time map generation[J]. IEEE Pervasive Computing, 2006, 5(4): 47-54.
[9] BIAGIONI J, ERIKSSON J. Inferring road maps from global positioning system traces[J]. Transportation Research Record: Journal of the Transportation Research Board, 2012, 2291(1): 61-71.
[10] LIU Xuemei, BIAGIONI J, ERIKSSON J, et al. Mining large-scale, sparse GPS traces for map inference: comparison of approaches[C]//Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. Beijing, China: ACM Press, 2012.
[11] 胡瀚, 向隆刚, 王德浩. 出租车轨迹数据的道路提取[J]. 测绘通报, 2018(7): 53-57. HU Han, XIANG Longgang, WANG Dehao. Road extraction based on taxi trajectory data[J]. Bulletin of Surveying and Mapping, 2018(7): 53-57.
[12] 李思宇, 向隆刚, 张彩丽, 等. 基于低频出租车轨迹的城市路网交叉口提取研究[J]. 地球信息科学学报, 2019, 21(12): 1845-1854. LI Siyu, XIANG Longgang, ZHANG Caili, et al. Extraction of urban road network intersections based on low-frequency taxi trajectory data[J]. Journal of Geo-Information Science, 2019, 21(12): 1845-1854.
[13] CHIANG Yaoyi, KNOBLOCK C A. Automatic extraction of road intersection position, connectivity, and orientations from raster maps[C]//Proceedings of the 16th ACM SIGSPATIAL international conference on Advances in geographic information systems. Irvine, California, USA: ACM Press, 2008.
[14] RAVANBAKHSH M, HEIPKE C, PAKZAD K. Road junction extraction from high-resolution aerial imagery[J]. The Photogrammetric Record, 2008, 23(124): 405-423.
[15] 曹闻, 李润生. 利用可变形部件模型检测遥感影像道路交叉口[J]. 武汉大学学报(信息科学版), 2018, 43(3): 413-419. CAO Wen, LI Runsheng. Road intersections detection using deformable part models on remote sensing image[J]. Geomatics and Information Science of Wuhan University, 2018, 43(3): 413-419.
[16] 王龙飞, 刘智, 金飞, 等. 道路交叉口自动检测算法的研究[J]. 测绘科学, 2020, 45(5): 126-131, 146. WANG Longfei, LIU Zhi, JIN Fei, et al. Research on automatic recognition algorithm of road intersection[J]. Science of Surveying and Mapping, 2020, 45(5): 126-131, 146.
[17] SHI Wenzhong, MIAO Zelang, DEBAYLE J. An integrated method for urban main-road centerline extraction from optical remotely sensed imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(6): 3359-3372.
[18] ZHANG Yingying, ZHANG Junping, LI Tong, et al. Road extraction and intersection detection based on tensor voting[C]//Proceedings of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Beijing, China: IEEE, 2016: 1587-1590.
[19] SUN Ke, ZHANG Junping, ZHANG Yingying. Roads and intersections extraction from high-resolution remote sensing imagery based on tensor voting under big data environment[J]. Wireless Communications and Mobile Computing, 2019, 2019: 1-11.
[20] NIE Dong, GAO Yaozong, WANG Li, et al. ASDNet: attention based semi-supervised deep networks for medical image segmentation[C]//Proceedings of 2018 Medical Image Computing and Computer Assisted Intervention.[S.l.]:MICCAI, 2018: 370-378.
[21] ZHOU Lichen, ZHANG Chuang, WU Ming. D-LinkNet: LinkNet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Salt Lake City, UT, USA:IEEE, 2018: 192-1924.
[22] GUY G, MEDIONI G. Inference of surfaces, 3D curves, and junctions from sparse, noisy, 3D data[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(11): 1265-1277.
[23] HONG Yi, ZHU Weiping. Spatial co-training for semi-supervised image classification[J]. Pattern Recognition Letters, 2015, 63: 59-65.
[24] CHEN Chen, LU Cewu, HUANG Qixing, et al. City-scale map creation and updating using GPS collections[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco California USA. New York, USA: ACM, 2016.
[25] WANG Chao, HAO Peng, WU Guoyuan, et al. Intersection and stop bar position extraction from vehicle positioning data[J]. IEEE Transactions on Intelligent Transportation Systems,2020(99): 1-12
[26] QI Xiaojun, HAN Yutao. Incorporating multiple SVMs for automatic image annotation[J]. Pattern Recognition, 2007, 40(2): 728-741.