Skip to main navigation Skip to search Skip to main content

Millimeter-wave image target recognition based on the combination of shape features

  • Ling Dai
  • , Hong Hu
  • , Yifan Chen
  • , Min Zhou
  • Harbin Institute of Technology Shenzhen
  • Southern University of Science and Technology

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

Abstract

Shape features are widely used in image target recognition due to their useful property of translation, rotation and scaling invariance. In this paper, low order Hu moments and other four shape features are combined together to classify different kinds of knives and guns with the support vector machine in the application of millimeter wave security imaging. Experimental results show that the combined features have better invariance and differences than other invariant moments.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Information and Automation, IEEE ICIA 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1732-1736
Number of pages5
ISBN (Electronic)9781509041022
DOIs
StatePublished - 24 Jan 2017
Externally publishedYes
Event2016 IEEE International Conference on Information and Automation, IEEE ICIA 2016 - Ningbo, China
Duration: 1 Aug 20163 Aug 2016

Publication series

Name2016 IEEE International Conference on Information and Automation, IEEE ICIA 2016

Conference

Conference2016 IEEE International Conference on Information and Automation, IEEE ICIA 2016
Country/TerritoryChina
CityNingbo
Period1/08/163/08/16

Keywords

  • Combined features
  • Feature extraction
  • Invariants
  • Target recognition

Fingerprint

Dive into the research topics of 'Millimeter-wave image target recognition based on the combination of shape features'. Together they form a unique fingerprint.

Cite this