Targets recognition based on deep learning

Huan Liu, Lei Kuang, Qing Huo Liu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A deep learning based recognition algorithm to identify various targets was proposed in this paper. Compared with traditional machine learning, deep learning can extract the features of recognized targets better and obtain higher accuracy. We first simulate electromagnetic scattering of targets and acquired the scattering electric field of targets at different frequencies and scattering angles. Then we use the scattering electric field to get the ISAR image. Then we input ISAR images to the deep convolutional neural networks for training, and extract the deeper features of the targets. In order to improve the accuracy of recognition, we combine different polarization ISAR images as one sample. Numerical results show that the average recognition accuracy of our proposed method is 99.72%, which verifies the effectiveness of the method.

Original languageEnglish
Title of host publication2019 Photonics and Electromagnetics Research Symposium - Fall, PIERS - Fall 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1393-1400
Number of pages8
ISBN (Electronic)9781728153049
DOIs
StatePublished - Dec 2019
Event2019 Photonics and Electromagnetics Research Symposium - Fall, PIERS - Fall 2019 - Xiamen, China
Duration: 17 Dec 201920 Dec 2019

Publication series

Name2019 Photonics and Electromagnetics Research Symposium - Fall, PIERS - Fall 2019 - Proceedings

Conference

Conference2019 Photonics and Electromagnetics Research Symposium - Fall, PIERS - Fall 2019
Country/TerritoryChina
CityXiamen
Period17/12/1920/12/19

Fingerprint

Dive into the research topics of 'Targets recognition based on deep learning'. Together they form a unique fingerprint.

Cite this