Ship behavior identification method based on deep learning

Abstract

The invention discloses a ship behavior identification method based on deep learning, and belongs to the field of mode identification. The ship behavior identification method can be applied to the fields of intelligent marine monitoring, ship intelligent supervision and the like. The ship behavior identification method specifically comprises the steps: S1, acquiring original ship track data, and building a ship behavior identification data set through a data preprocessing and track segmentation method; and S2, designing a ship behavior identification network in which the multi-scale convolution module and the long-short-term memory network are cascaded, and realizing behavior identification of the ship track data by using the ship behavior identification network trained by the self-built ship behavior identification data set. By the adoption of the technical scheme, the ship behavior identification technology is applied to the field of ship supervision, and ship behaviors are automatically analyzed from mass ship track data, and effective evidence obtaining and supervision can be conducted on ship behavior activities of offshore oceans, and a low-efficiency manual inspection mode is replaced. The whole scheme has the characteristics of low dependence on equipment, high identification accuracy, high identification speed and the like.

Description

technical field [0001] The invention relates to the technical field of ship behavior pattern recognition, in particular to a method for ship behavior recognition based on deep learning. Background technique [0002] my country has abundant marine resources and port resources. With the further development and utilization of marine resources, the number of ships of various types is increasing day by day, and maritime traffic activities are becoming more and more frequent. There are various acts of using ships to carry out illegal activities, including smuggling, Illegal immigration, illegal fishing, etc. The automatic identification system (Automatic Identification System, AIS) is a new type of navigation aid system, which can help relevant departments coordinate marine traffic and supervise marine activities. For example, during the fishing ban period of marine pastures, AIS data can be used to analyze whether there is illegal fishing. At present, most ships with a displacem...

AI Technical Summary

Problems solved by technology

[0004] 1. Need to manually design features, it is easy to ignore a lot of effective information;

[0005] 2. The calculation overhead depends on the established model, which is very time-consuming;

[0006] 3. The noise interferes a lot with the distribution of the clustering method, and the threshold setting has a strong subjective factor

Images