An Efficient and Accurate Convolution-Based Similarity Measure for Uncertain Trajectories
Abstract
:1. Introduction
- Heterogeneous sampling rates: Due to the nature of sensing devices and object activities, trajectories are sampled from continuous paths of objects with time-varying heterogeneous sampling rates, i.e., locations in a trajectory are collected randomly and sporadically. As a result, trajectories sampled from the same path may not share overlapping locations, making it challenging to accurately measure their similarity.
- Low sampling rates: Some trajectories could be very sparse due to the low data sampling rate of the sensing system (e.g., call detail records in a telecommunication system). The time interval between two consecutive locations could be large (such as tens of minutes), and object locations are not observed during that time. The infrequent observations of object locations introduce uncertainty when measuring trajectory similarity.
2. Related Work
3. Preliminary
3.1. Path and Trajectory
3.2. Overview of Contra
- Location mapping: Given two trajectories, Contra first partitions the space into non-overlapping grids. Then, it computes the grid position for each location to map the locations into grids, generating two trajectory matrices for the two trajectories.
- Trajectory feature extraction: Based on the two trajectory matrices, Contra uses trajectory convolution and pooling with a well-defined trajectory kernel to extract high-level shape features for the two trajectories. This process generates two feature matrices that capture crucial characteristics and spatial correlations presented in the trajectories.
- Similarity comparison: Contra compares the similarity between the two trajectories based on their feature matrices. It computes the average difference between the two feature matrices to represent their similarity.
- Trajectory index: To accelerate the feature extraction process, we developed innovative trajectory index strategies for Contra. These strategies involve indexing the non-zero elements within a trajectory matrix and determining the positions of affected elements for each non-zero element. By utilizing the trajectory index, only the affected elements need to be updated during trajectory convolution and pooling operations, resulting in improved efficiency.
4. Methodology
4.1. Location Mapping
4.2. Trajectory Feature Extraction
4.2.1. Trajectory Convolution
4.2.2. Trajectory Pooling
4.3. Similarity Computation
4.4. Trajectory Index
5. Illustrative Experimental Results
5.1. Datasets and Baselines
5.2. Effectiveness of Self-Similarity
5.3. Effectiveness of Cross-Similarity
5.4. Efficiency and Scalability
5.5. Effects of Different Parameters
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
Abbreviations
MR | mean rank |
CNN | convolution neural network |
Contra | convolution-based similarity measure for uncertain trajectories |
References
- Cui, J.; Zha, H.; Zhao, H.; Shibasaki, R. Multi-modal tracking of people using laser scanners and video camera. Image Vis. Comput. 2008, 26, 240–252. [Google Scholar] [CrossRef]
- He, S.; Chan, S.H.G. Wi-Fi fingerprint-based indoor positioning: Recent advances and comparisons. IEEE Commun. Surv. Tutorials 2015, 18, 466–490. [Google Scholar] [CrossRef]
- Li, M.; Liu, N.; Niu, Q.; Liu, C.; Chan, S.H.G.; Gao, C. SweepLoc: Automatic video-based indoor localization by camera sweeping. In Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, Singapore, 8–12 October 2018; Volume 2, pp. 1–25. [Google Scholar]
- Su, H.; Liu, S.; Zheng, B.; Zhou, X.; Zheng, K. A survey of trajectory distance measures and performance evaluation. VLDB J. 2020, 29, 3–32. [Google Scholar] [CrossRef]
- Zheng, Y. Trajectory data mining: An overview. ACM Trans. Intell. Syst. Technol. (TIST) 2015, 6.3, 1–41. [Google Scholar] [CrossRef]
- Li, G.; Hu, S.; Zhong, S.; Tsui, W.L.; Chan, S.H.G. vContact: Private WiFi-based IoT contact tracing with virus lifespan. IEEE Internet Things J. 2019, 9, 3465–3480. [Google Scholar] [CrossRef]
- Lu, G.; Nie, Y.M.; Liu, X.; Li, D. Trajectory-based traffic management inside an autonomous vehicle zone. Transp. Res. Part B Methodol. 2019, 120, 76–98. [Google Scholar] [CrossRef]
- Xiao, Z.; Wang, Y.; Fu, K.; Wu, F. Identifying different transportation modes from trajectory data using tree-based ensemble classifiers. ISPRS Int. J. Geo-Inf. 2017, 6, 57. [Google Scholar] [CrossRef]
- Shang, S.; Ding, R.; Yuan, B.; Xie, K.; Zheng, K.; Kalnis, P. User oriented trajectory search for trip recommendation. In Proceedings of the 15th International Conference on Extending Database Technology, Berlin, Germany, 27–30 March 2012; pp. 156–167. [Google Scholar]
- Tao, F.; Wu, J.; Lin, S.; Lv, Y.; Wang, Y.; Zhou, T. Revealing the Impact of COVID-19 on Urban Residential Travel Structure Based on Floating Car Trajectory Data: A Case Study of Nantong, China. ISPRS Int. J. Geo-Inf. 2023, 12, 55. [Google Scholar] [CrossRef]
- Purwanto, P.; Utaya, S.; Handoyo, B.; Bachri, S.; Astuti, I.S.; Utomo, K.S.B.; Ldianto, Y.E. Spatiotemporal analysis of COVID-19 spread with emerging hotspot analysis and space–time cube models in East Java, Indonesia. ISPRS Int. J. Geo-Inf. 2021, 10, 133. [Google Scholar] [CrossRef]
- Li, G.; Chen, C.J.; Peng, W.C.; Yi, C.W. Estimating crowd flow and crowd density from cellular data for mass rapid transit. In Proceedings of the 6th International Workshop on Urban Computing, Halifax, NS, Canada, 13–17 August 2017; pp. 18–30. [Google Scholar]
- Giannotti, F.; Nanni, M.; Pinelli, F.; Pedreschi, D. Trajectory pattern mining. In Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 330–339. [Google Scholar]
- Zheng, L.; Feng, Q.; Liu, W.; Zhao, X. Discovering trip hot routes using large scale taxi trajectory data. In Proceedings of the International Conference on Advanced Data Mining and Applications, Gold Coast, QLD, Australia, 12–15 December 2016; Springer: Cham, Switzerland, 2016; pp. 534–546. [Google Scholar]
- Wu, H.; Sun, W.; Zheng, B.; Yang, L.; Zhou, W. CLSTERS: A general system for reducing errors of trajectories under challenging localization situations. In Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, Maui, HI, USA, 11–12 September 2017; Volume 1, pp. 1–28. [Google Scholar]
- He, S.; Chan, S.H.G. Tilejunction: Mitigating signal noise for fingerprint-based indoor localization. IEEE Trans. Mob. Comput. 2015, 15, 1554–1568. [Google Scholar] [CrossRef]
- Chen, L.; Ng, R. On the marriage of lp-norms and edit distance. In Proceedings of the Thirtieth International Conference on Very Large Data Bases, Toronto, ON, Canada, 31 August–3 September 2004; Volume 30, pp. 792–803. [Google Scholar]
- Yi, B.K.; Jagadish, H.V.; Faloutsos, C. Efficient retrieval of similar time sequences under time warping. In Proceedings of the 14th International Conference on Data Engineering, Washington, DC, USA, 23–27 February 1998; pp. 201–208. [Google Scholar]
- Ranu, S.; Deepak, P.; Telang, A.D.; Deshpande, P.; Raghavan, S. Indexing and matching trajectories under inconsistent sampling rates. In Proceedings of the 2015 IEEE 31st International Conference on Data Engineering, Seoul, Republic of Korea, 13–15 April 2015; pp. 999–1010. [Google Scholar]
- Vlachos, M.; Kollios, G.; Gunopulos, D. Discovering similar multidimensional trajectories. In Proceedings of the 18th International Conference on Data Engineering, San Jose, CA, USA, 26 February–1 March 2002; pp. 673–684. [Google Scholar]
- Chen, L.; Özsu, M.T.; Oria, V. Robust and fast similarity search for moving object trajectories. In Proceedings of the 2005 ACM SIGMOD International Conference on Management of Data, Baltimore, MD, USA, 14–16 June 2005; pp. 491–502. [Google Scholar]
- Li, X.; Zhao, K.; Cong, G.; Jensen, C.S.; Wei, W. Deep representation learning for trajectory similarity computation. In Proceedings of the 2018 IEEE 34th International Conference on Data Engineering (ICDE), Paris, France, 16–19 April 2018; pp. 617–628. [Google Scholar]
- Han, P.; Wang, J.; Yao, D.; Shang, S.; Zhang, X. A graph-based approach for trajectory similarity computation in spatial networks. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, Singapore, 14–18 August 2021; pp. 556–564. [Google Scholar]
- Eiter, T.; Mannila, H. Computing Discrete Fréchet Distance. 1994. Available online: http://www.kr.tuwien.ac.at/staff/eiter/et-archive/cdtr9464.pdf (accessed on 17 October 2023).
- Hung, C.C.; Peng, W.C.; Lee, W.C. Clustering and aggregating clues of trajectories for mining trajectory patterns and routes. VLDB J. 2015, 24, 169–192. [Google Scholar] [CrossRef]
- Ketabi, R.; Alipour, B.; Helmy, A. Playing with matches: Vehicular mobility through analysis of trip similarity and matching. In Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Seattle, WA, USA, 2–5 November 2018; pp. 544–547. [Google Scholar]
- Tiakas, E.; Papadopoulos, A.N.; Nanopoulos, A.; Manolopoulos, Y.; Stojanovic, D.; Djordjevic-Kajan, S. Trajectory similarity search in spatial networks. In Proceedings of the 2006 10th International Database Engineering and Applications Symposium (IDEAS’06), Delhi, India, 11–14 December 2006; pp. 185–192. [Google Scholar]
- Nanni, M.; Pedreschi, D. Time-focused clustering of trajectories of moving objects. J. Intell. Inf. Syst. 2006, 27, 267–289. [Google Scholar] [CrossRef]
- Su, H.; Zheng, K.; Wang, H.; Huang, J.; Zhou, X. Calibrating trajectory data for similarity-based analysis. In Proceedings of the 2013 ACM SIGMOD International Conference on Management of Data, New York, NY, USA, 22–27 June 2013; pp. 833–844. [Google Scholar]
- Emrich, T.; Kriegel, H.P.; Mamoulis, N.; Renz, M.; Zufle, A. Querying uncertain spatio-temporal data. In Proceedings of the 2012 IEEE 28th International Conference on Data Engineering, Arlington, VA, USA, 1–5 April 2012; pp. 354–365. [Google Scholar]
- Niedermayer, J.; Züfle, A.; Emrich, T.; Renz, M.; Mamoulis, N.; Chen, L.; Kriegel, H.P. Probabilistic nearest neighbor queries on uncertain moving object trajectories. arXiv 2013, arXiv:1305.3407. [Google Scholar] [CrossRef]
- Li, G.; Hung, C.C.; Liu, M.; Pan, L.; Peng, W.C.; Chan, S.H.G. Spatial-temporal similarity for trajectories with location noise and sporadic sampling. In Proceedings of the 2021 IEEE 37th International Conference on Data Engineering (ICDE), Chania, Greece, 19–22 April 2021; pp. 1224–1235. [Google Scholar]
- Yao, D.; Hu, H.; Du, L.; Cong, G.; Han, S.; Bi, J. Trajgat: A graph-based long-term dependency modeling approach for trajectory similarity computation. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Washington, DC, USA, 14–18 August 2022; pp. 2275–2285. [Google Scholar]
- Liu, X.; Tan, X.; Guo, Y.; Chen, Y.; Zhang, Z. Cstrm: Contrastive self-supervised trajectory representation model for trajectory similarity computation. Comput. Commun. 2022, 185, 159–167. [Google Scholar] [CrossRef]
- Yu, Y.; Yao, Y.; Liu, Z.; An, Z.; Chen, B.; Chen, L.; Chen, R. A Bi-LSTM approach for modelling movement uncertainty of crowdsourced human trajectories under complex urban environments. Int. J. Appl. Earth Obs. Geoinf. 2023, 122, 103412. [Google Scholar] [CrossRef]
- Shi, W.; Yu, Y.; Liu, Z.; Chen, R.; Chen, L. A deep-learning approach for modelling pedestrian movement uncertainty in large-scale indoor areas. Int. J. Appl. Earth Obs. Geoinf. 2022, 114, 103065. [Google Scholar] [CrossRef]
- Yuan, J.; Zheng, Y.; Xie, X.; Sun, G. Driving with knowledge from the physical world. In Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Diego, CA, USA, 21–24 August 2011; pp. 316–324. [Google Scholar]
- Yuan, J.; Zheng, Y.; Zhang, C.; Xie, W.; Xie, X.; Sun, G.; Huang, Y. T-drive: Driving directions based on taxi trajectories. In Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems, San Jose, CA, USA, 2–5 November 2010; pp. 99–108. [Google Scholar]
Number of Trajectory | Average Time Interval (s) | Average Duration (s) | |
---|---|---|---|
Porto dataset | 1,233,766 | 15.00 | 783.45 |
Beijing dataset | 333,948 | 174.36 | 8473.26 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Li, G.; Deng, X.; Xu, J.; Liu, Y.; Zhang, J.; Xiong, S.; Gao, F. An Efficient and Accurate Convolution-Based Similarity Measure for Uncertain Trajectories. ISPRS Int. J. Geo-Inf. 2023, 12, 432. https://doi.org/10.3390/ijgi12100432
Li G, Deng X, Xu J, Liu Y, Zhang J, Xiong S, Gao F. An Efficient and Accurate Convolution-Based Similarity Measure for Uncertain Trajectories. ISPRS International Journal of Geo-Information. 2023; 12(10):432. https://doi.org/10.3390/ijgi12100432
Chicago/Turabian StyleLi, Guanyao, Xingdong Deng, Jianmin Xu, Yang Liu, Ji Zhang, Simin Xiong, and Feng Gao. 2023. "An Efficient and Accurate Convolution-Based Similarity Measure for Uncertain Trajectories" ISPRS International Journal of Geo-Information 12, no. 10: 432. https://doi.org/10.3390/ijgi12100432
APA StyleLi, G., Deng, X., Xu, J., Liu, Y., Zhang, J., Xiong, S., & Gao, F. (2023). An Efficient and Accurate Convolution-Based Similarity Measure for Uncertain Trajectories. ISPRS International Journal of Geo-Information, 12(10), 432. https://doi.org/10.3390/ijgi12100432