The automatic recognition of traffic flow regions can provide decision support for ships’ automatic route design and route planning. This study analyzes the characteristics of ships’ trajectory structures and builds a course and distance model. Pearson correlation coefficients are used for measuring the similarities of the models and clustering trajectories, and kernel density estimation is used for estimating the probability density of clustered trajectories. An automatic recognition algorithm for traffic flow regions is proposed. This study examines ships’ automatic identification system data in Laotieshan channel, China. The traffic separation scheme regions and traffic intersectional regions are recognized automatically, and the obtained results show good agreement with actual circumstances, thus verifying the applicability of the algorithm
Li, Wei-Feng; Mei, Bin; and Shi, Guo-You
"AUTOMATIC RECOGNITION OF MARINE TRAFFIC FLOW REGIONS BASED ON KERNEL DENSITY ESTIMATION,"
Journal of Marine Science and Technology: Vol. 26:
1, Article 8.
Available at: https://jmstt.ntou.edu.tw/journal/vol26/iss1/8