Harbor anchored ship monitoring method and system and electronic equipment

Abstract

The invention relates to the technical field of harbor transportation management, in particular to a harbor anchored ship monitoring method and system and electronic equipment. The harbor anchored ship monitoring method comprises the steps of a, acquiring harbor anchored ship data through a camera shooting device; b, constructing a neural network model based on 'a full convolutional network + a regional target network branch'; c, inputting the harbor parked data into the neural network model based on 'the full convolutional network + the regional target network branch', wherein the neural network model based on 'the full convolutional network + the regional target network branch' conducts ship object regional segmenting and ship type classification according to the harbor anchored ship data, and the harbor anchored ship state is calculated according to the ship object regional segmenting result and the ship type classification result. By means of the harbor anchored ship monitoring method and system and the electronic equipment, the congesting state of a current harbor dock can be automatically and accurately evaluated, limiting conditions of an application scene are smaller in quantity and are more intelligent and more convenient, and the harbor anchored ship monitoring method and system are conveniently and widely applied.

Description

technical field [0001] The present application relates to the technical field of port traffic management, in particular to a port mooring monitoring method, system and electronic equipment. Background technique [0002] The rapid development of the shipping industry has made great contributions to the development of the national economy. The berthing monitoring of the port terminal is very important to the shipping industry. In the process of coordinating the berthing of ships and controlling the ships entering the port terminal, it is necessary to efficiently and accurately divide the congestion degree of the port and the terminal, which is a reasonable dispatching of ships. Key to boat docking. However, the current maritime traffic management system can no longer keep up with the needs of the development of the shipping industry. [0003] Guo Shijun of the Naval Arms Command Academy pointed out that the tonnage of ships has increased, the density of ships has increased, ...

AI Technical Summary

Problems solved by technology

[0006] 1. The existing research on marine traffic management technology based on radar, airship and other equipment needs to maintain a huge port area information knowledge base, and compare it with previous knowledge to complete image interpretation and understanding, so the computational complexity is high

[0007] 2. The ship target recognition technology in shipping using image processing only completes the extraction, recognition and classification of the main targets, and does not segment the areas of various types of ship targets, and cannot achieve satisfactory efficiency and accuracy. Effect

[0008] 3. The technology of extracting object regions in the latest research artificially divides recognition classification and region segmentation, resulting in increased computational complexity in training

[0009] 4. Up to now, there is no technology for ship type identification and area segmentation using machine learning methods

However, machine learning-based object recognition and segmentation techniques are not currently used in port and terminal congestion monitoring techniques

Images