Abstract
Abstract
This paper introduces the research background of water traffic problems, analyzes the Marine equipment in the process of shipping the deficiencies, on the basis of introducing the concept of image recognition, computer vision and computer vision image recognition are briefly introduced the development of the status situation, sums up and summarizes the recent years research direction such as neural networks based on machine learning, convolution algorithm, For ship image recognition method, and from static image recognition to the complex environment of image recognition, and machine learning, convolutional neural networks based algorithm was analyzed, and the advantages and disadvantages and applicable environment, finally put forward to build a multilateral participation information fusion detection framework to implement the idea of the safe navigation of the ship.
Subject
General Physics and Astronomy
Reference18 articles.
1. Application of Formal Safety Assessment(FSA) to the Shipping Traffic Safety Management of the Port and its Adjoining Waters[D];Duan,2006
2. Region-Based Convolutional Networks for Accurate Object Detection and Segmentation.[J];Girshick;IEEE transactions on pattern analysis and machine intelligence,2015
3. SSD(Single Shot MultiBox Detector) Research and improvement of target detection algorithm [J];Wang;Industrial control computer,2019
4. Multi-scale ship image target recognition method for inland navigation safety Monitoring [J/OL];Zhang,2021