A cascaded coarse-to-fine convolutional neural network method for ship type identification

A neural network and type recognition technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of time-consuming acquisition of ship type information, adverse maritime supervision efficiency, etc., to ensure maritime traffic safety, realize automation and safety. High-precision identification and the effect of improving navigation efficiency
CN109299671AInactive Publication Date: 2019-02-01SHANGHAI MARITIME UNIVERSITY
6 Cites 9 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANGHAI MARITIME UNIVERSITY
Publication Date
2019-02-01
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides a ship type identification method of cascade type deep convolution neural network from coarse to fine, This method adopts random heuristic selection mechanism to dynamically adjust the structure and parameter settings of the depth network. The depth convolution neural network which can identify the ship type is trained by two steps of rough training and fine training. The rough training process is similar to the training process of the traditional depth convolution neural network, and the input sample of the training process is the ship picture. In order to improve the overall accuracy of ship type recognition, the depth convolution neural network is trained again in the fine-level training process for the merchant ship images with the lowest ship type recognition accuracy in the rough-level training process. The method of the invention can obtain better identification accuracy for different ship types, and provides information support for automatic ship type identification and intelligent navigation of ships.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field

[0001] The invention relates to the technical field of maritime video monitoring, in particular to a cascaded coarse-to-fine convolutional neural network ship type identification method. Background technique

[0002] Currently, Vessel Traffic Service (VTS) and Automatic Identification System (AIS) are the main means to obtain ship type information. After the ship enters the VTS reporting line, the crew on board will report the ship’s basic information, such as port of destination, port of departure, and ship type, to the maritime regulatory authority via VHF telephone. In addition, the AIS system will periodically distribute its own ship’s static and dynamic information through broadcasting, including its ship’s type, position, call sign, ship name, gross tonnage, ship’s draft and speed, etc. However, AIS users need to manually enter static information such as ship type and ship call sign for the AIS system in advance. From the above analysis, it can be seen tha...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More