Online Anomalous Trajectory Detection with Deep Generative Sequence Modeling

Liu, Yiding; Zhao, Kaiqi; Cong, Gao; Bao, Zhifeng

doi:10.1109/icde48307.2020.00087

Cited by 62 publications

(43 citation statements)

References 32 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lv et al [13] projected trajectories into two-dimensional images, which can better characterize spatial features of the trajectories from different special scales. Liu et al [14] mapped trajectories into grid sequences, and its chronological order and location characterize the temporal and spatial features of the trajectories, respectively.…”

Section: A Trajectory Embeddingmentioning

confidence: 99%

“…The anomaly trajectories were defined on the basis of the reconstruction error. Liu et al [14] proposed a generative model, which can identify normal routes and define trajectories with small regenerate probability as anomalies. Wu et al [3] proposed a method to model the driving behavior and measured the decision cost of the driving behavior.…”

Section: Anomalous Trajectory Detectionmentioning

confidence: 99%

“…(1) Dataset A: This dataset is an open dataset generated by 442 taxis in Porto, Portugal, from January 07 2013 to June 30 2014 1 . Following the previous study [1], [14], we group the trajectories by the source and destination pairs and select the popular sources and destinations with sufficient historical trajectories. The filtered dataset contains 10978 trajectories from 17 source and destination pairs, and 9.83% trajectories are anomalous.…”

Section: A Datasets and Settingsmentioning

confidence: 99%

“…Ground Truth: Some works [3], [6], [14] tend to use several schemes or rules to generate anomalous trajectories given that no labeled dataset is available for anomalous trajectory detection. We claim that the generated anomalies may follow few anomalous patterns, which cannot cover various anomalous patterns in the real-world scenarios.…”

Section: A Datasets and Settingsmentioning

confidence: 99%

“…SAE [17] is a sequence to sequence model based on conventional RNN, which identifies the anomaly by the reconstruction error. Gaussian Mixture Variational Sequence AutoEncoder (GM-VSAE) and its extension, Source Destination network Variational Aequence AutoEncoder (SD-1 http://www.kaggle.com/c/pkdd-15-predict-taxi-service-trajectory-i VSAE) [14] are deep generative mthords, which model probability distribution of normal routes using gaussian components and the routes that are close to the centers of the gaussian components are more likely to be normal routes. In addition, we select one variant of the proposed methods, namely, GCM (LST M only), which only uses LSTM layers to detect anomalous trajectories.…”

Section: A Datasets and Settingsmentioning

confidence: 99%

See 4 more Smart Citations

Abnormal Trajectory Detection Based on Geospatial Consistent Modeling

et al. 2020

View full text Add to dashboard Cite

Anomalous trajectory detection plays a significant role in fraud detection and adverse events monitoring for ride-hailing services. The spatial and temporal dynamics of road networks and the sparsity of trajectories make anomalous trajectory detection a challenging task. Most existing methods are based on density and isolation approaches, which ignore geographical information. Motivated by these challenges and shortcomings, we propose a novel method, which considers geospatial constraints of the trajectories and avoids sparsity issues. In our method, the geographical information and topological constraints of trajectories are embedded into structured vector space. Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) are used to model common trajectory features. Our method could identify anomalous trajectories and determine which parts are responsible for anomalies by using these features. Experiments on two real-world datasets have been conducted, and results demonstrate the effectiveness and feasibility of the proposed method.

show abstract

Section: A Trajectory Embeddingmentioning

confidence: 99%

Section: Anomalous Trajectory Detectionmentioning

confidence: 99%

Section: A Datasets and Settingsmentioning

confidence: 99%

Section: A Datasets and Settingsmentioning

confidence: 99%

Section: A Datasets and Settingsmentioning

confidence: 99%

See 3 more Smart Citations

Abnormal Trajectory Detection Based on Geospatial Consistent Modeling

et al. 2020

View full text Add to dashboard Cite

show abstract

Matching method based on similarity of working trajectories

Zhong

Chen

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

To identify whether the actual work trajectory of workers in the factory meets the predetermined work trajectory requirements, we proposed an efficient and accurate work trajectory similarity matching method. We comprehensively considered the similarity between the actual work track and the predetermined track from the two characteristics of track angle and track distance. Among them, the similarity of track rotation angle is calculated using the improved longest common subsequence algorithm, and the similarity of track distance is calculated using the improved dynamic time warping (DTW) algorithm. Then the results of the similarity calculation of these two features are weighted. Finally, the weighted results are used to evaluate the similarity between the actual work track and the predetermined track, so as to judge whether the actual work track meets the requirements of the predetermined track. Experimental data show that the trajectory similarity matching algorithm in this paper has higher accuracy and efficiency than traditional DTW and other algorithms, and has higher ability to resist the interference of trajectory point evacuation than traditional DTW and other algorithms.

show abstract