Research on ship target recognition system based on Neural Network

Yan Wang

Abstract


Ship Target Recognition usually uses band image recognition technology, and multi band image fusion recognition can
expand the application scope of recognition system. In order to improve the efficiency and accuracy of fusion recognition, this paper
attempts to introduce convolutional neural network technology to design an intelligent fusion recognition method , which uses alexnet
network model, from the visible light, medium wave, medium waveThe features of the three bands are extracted from the long wave
infrared three band image, and the features of the three bands are screened by using mutual information to determine the feature vector with
fi xed length; According to the diff erent feature levels, three fusion methods, early, middle and late, are used to verify the algorithm. The
experimental results show that the accuracy of the method is the highest, reaching 84.5%.

Keywords


Target recognition;Ship identifi cation; Feature f usion; Convolutional neural network; Multi band image; Feature selection

Full Text:

PDF

Included Database


References


[1] Ru Fei, litieyingReview of artifi cial neural network system identifi cation [j]Software guide, 2011,10 (03): 134-135

[2] Wei Na, Ben Kerong, Zhang Linke, pangyunfuResearch on the analysis of underwater vehicle source contribution based on neural network system

identification [c]/ / theoretical computer science professional committee of the Chinese computer society. Proceedings of the 2005 National Academic

Conference on theoretical computer scienceJournal of computer science, 2005:207-209

[3] YanghuaijiangPerformance analysis of neural network system identification and adaptive control for photodynamic systems [j]Optical precision

engineering, 1998 (02): 37-44

[4] Ran Qiquan, Li Shilun, duzhiminOptimized reservoir geological model established by neural network system identifi cation theory [j]China off shore oil

and gasGeology, 1996 (01): 33-37

[5] Liuxinghua, liuxianying, Hu ZeResearch on system identifi cation based on RBF neural network [j]Modern electronic technology, 2004 (24): 57-59

[6] LijianxinDesign and Simulation of PID control based on neural network system identifi cation [j]Electronic technology and software engineering, 2021 (08):

122-125

[7] Huang Haocai, Huang Yijian, Yang GuanluNeural network system identification based on LM algorithm [j]Modular machine tools and automatic

processing technology, 2003 (02): 8-10+13

[8] Yuhaibo, Ma CuihongSystem identifi cation and application based on diagonal recurrent neural network [j]Microcomputer information, 2007 (31): 216-217+172

[9] LiuhaifengResearch on system identifi cation method based on neural network [d]Xi’an University of Electronic Science and technology: 2007

[10] Xuxianfeng, Zhang Li, Lang bin, Xia ZhenResearch on face recognition algorithm based on improved twin convolutional neural network with perceptual

model [j]Acta electronica Sinica, 2020,48 (04): 643-647

[11] Li Nan, caijianyong, Li Ke, Cheng Yu, zhangmingweiConvolutional neural network face recognition algorithm based on multi perception structure [j]

Computer system applications, 2020,29 (02): 157-162

[12] Wangweimin, Tang Yang, Zhang Jian, Zhang YiqiuFace recognition algorithm based on convolutional neural network feature fusion [j]Computer and

digital engineering, 2020,48 (01): 88-92+105

[13] Yu YichunFace recognition based on neural network algorithm [j]Electronic technology and software engineering, 2019 (24): 247-248

[14] ShenjiyunAnalysis of face recognition technology based on neural network deep learning algorithm [j]Information technology and informatization, 2020

(05): 228-230

[15] Zhang HanResearch on the application of BP neural network algorithm in face recognition [j]Software, 2020,41 (05): 193-197




DOI: https://doi.org/10.18686/esta.v10i1.306

Refbacks

  • There are currently no refbacks.


Copyright (c) 2023 Yan Wang