Performance Analysis of Deep-Learning Target Classification Algorithms Using Micro-Doppler Images 


Vol. 46,  No. 3, pp. 430-439, Mar.  2021
10.7840/kics.2021.46.3.430


PDF
  Abstract

Micro-Doppler modulation is a target signature that represents micro-motion state for each individual movement, it is used in the technology of recognizing and classifying targets. The micro-Doppler frequency appears in the form of transition of the Doppler frequency by basic movement characteristics such as rotation and vibration of an object, and thus it can make it possible to track a target and classify it with high recognition accuracy. In this paper, we model micro-motion signals of a drone, a bird, and human targets, and analyze them in the time-frequency domain through micro-Doppler images to confirm the micro-Doppler images to confirm the micro-Doppler characteristics of the target. To classify targets performing micro-movement, we apply four deep neural networks, such as AlexNet, VGGNet16, GoogLeNet, and ResNet34, to micro-Doppler images input. Through simulation, we analyze the classification performance of deep learning algorithms according to the radar measurement data input set of each target. Simulation results show that all four neural networks have more than 87% classification accuracy performance, and in the case of ResNet34, target classification performance is the best with more than 90% performance on three scales of accuracy, precision and recall.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

J. Kim, D. Park, H. Kim, "Performance Analysis of Deep-Learning Target Classification Algorithms Using Micro-Doppler Images," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 3, pp. 430-439, 2021. DOI: 10.7840/kics.2021.46.3.430.

[ACM Style]

Ji-Hyeon Kim, Do-Hyun Park, and Hyoung-Nam Kim. 2021. Performance Analysis of Deep-Learning Target Classification Algorithms Using Micro-Doppler Images. The Journal of Korean Institute of Communications and Information Sciences, 46, 3, (2021), 430-439. DOI: 10.7840/kics.2021.46.3.430.

[KICS Style]

Ji-Hyeon Kim, Do-Hyun Park, Hyoung-Nam Kim, "Performance Analysis of Deep-Learning Target Classification Algorithms Using Micro-Doppler Images," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 3, pp. 430-439, 3. 2021. (https://doi.org/10.7840/kics.2021.46.3.430)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880