Maritime traffic situation awareness analysis via high-fidelity ship imaging trajectory

Situation awareness provides crucial yet instant information to maritime traffic participants, and significant attentions are paid to implement traffic situation awareness task via various maritime data source (e.g., automatic identification system, maritime surveillance video, radar, etc.). The study aims to analyze traffic situation with the support of ship imaging trajectory. First, we employ the dark channel prior model to remove fog in maritime videos to obtain high-resolution ship images (i.e., fog-free maritime images). Second, we track ships in each maritime image with the scale adaptive kernel correlation filter (SAMF), and thus obtain raw ship imaging trajectories. Third, we cleanse abnormal ship trajectory samples via curve-fitting and down sampling method, and thus further maritime traffic situation analysis is implemented. We analyze maritime traffic situation in three typical videos (i.e., three typical maritime traffic scenarios), and experimental results suggested that the proposed framework can extract high-resolution ship imaging trajectory for fulfilling the task of accurate maritime traffic situation awareness.