Abstract: The present invention proposes an ensemble deep learning method for identifying unsafe behaviors of operators in maritime working environment. Firstly, extract features of maritime images with the You Only Look Once (YOLO) V3 model, and then enhance a multi-scale detection capability by introducing a feature pyramid structure. Secondly, obtain instance-level features and time memory features of the operators and devices in the maritime working environment with the Joint Learning of Detection and Embedding (JDE) paradigm. Thirdly, transfer spatial-temporal interaction information into a feature memory pool, and update the time memory features with the asynchronous memory updating algorithm. Finally, identify the interaction between the operators, the devices, and unsafe behaviors with an asynchronous interaction aggregation network. The proposed invention can accurately determine the unsafe behaviors of the operators, and thus provide operation decisions for maritime management relevant activities.

Abstract: The present invention proposes a vessel traffic pattern identification method via data quality control and data compression. Firstly, assort a collection of Automatic Identification system (AIS) data points according to Mobile Service Identify (MMSI) code and sort each collection result by time ascending order, and then delete duplicated vessel AIS data points considering time stamp, latitude, longitude and vessel speed over ground, then segment vessel trajectories. Secondly obtain high-quality AIS data with an AIS data anomaly detection and repair and compress each vessel trajectory with the Douglas-Peucker algorithm. Thirdly, cluster vessel trajectories with the Quick Bundles algorithm, and identify maritime traffic pattern. The invention can efficiently identify vessel traffic patterns, and help maritime traffic management departments to accurately identify a traffic situation.