An unsupervised learning method with convolutional auto-encoder for vessel trajectory similarity computation. (1st April 2021)

• Record Type: Journal Article
• Authors: Liang, Maohan; Liu, Ryan Wen; Li, Shichen; Xiao, Zhe; Liu, Xin; Lu, Feng
• Is Part Of: Ocean engineering. Volume 225(2021)
• Publication Date: 2021-04-01
• Subjects: Automatic identification system (AIS) -- Trajectory similarity -- Trajectory clustering -- Convolutional neural network (CNN) -- Convolutional auto-encoder (CAE); Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2021.108803
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: To achieve reliable mining results for massive vessel trajectories, one of the most important challenges is how to efficiently compute the similarities between different vessel trajectories. The computation of vessel trajectory similarity has recently attracted increasing attention in the maritime data mining research community. However, traditional shape- and warping-based methods often suffer from several drawbacks such as high computational cost and sensitivity to unwanted artifacts and non-uniform sampling rates, etc. To eliminate these drawbacks, we propose an unsupervised learning method which automatically extracts low-dimensional features through a convolutional auto-encoder (CAE). In particular, we first generate the informative trajectory images by remapping the raw vessel trajectories into two-dimensional matrices while maintaining the spatio-temporal properties. Based on the massive vessel trajectories collected, the CAE can learn the low-dimensional representations of informative trajectory images in an unsupervised manner. The trajectory similarity is finally equivalent to efficiently computing the similarities between the learned low-dimensional features, which strongly correlate with the raw vessel trajectories. Comprehensive experiments on realistic data sets have demonstrated that the proposed method largely outperforms traditional trajectory similarity computation methods in terms of efficiency and effectiveness. The high-quality trajectory clustering performance could also be guaranteed according to the CAE-based trajectory similarity computation results. Graphical abstract: The flowchart of our proposed CAE-based unsupervised learning method for vessel trajectory similarity computation and its application to vessel trajectory clustering. Image 1 Highlights: Similarities between vessel trajectories are equivalent to computing similarities between the informative trajectory images. An unsupervised learning method is proposed to measure similarities between different informative trajectory images. The proposed learning method shows superior performance in terms of both efficiency and effectiveness.