WLIB-SIFT: A Distinctive Local Image Feature Descriptor

Mao Wei, Peng Xiwei

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

22 Citations (Scopus)

Abstract

SIFT descriptor plays a great role in image mosaic, image retrieval and target recognition for good invariance of translation, rotation and zoom. However, the disadvantages of SIFT descriptor are its high dimensionality and complex computation. Besides, SIFT descriptor has poor performance when massive similar local features and complex background exist in the matching image. In this paper, a distinctive and robust weighted local intensity binary SIFT descriptor(WLIB-SIFT) is proposed. A WLIB-SIFT descriptor consists of a weighted binary SIFT descriptor(B-SIFT) and a weighted local intensity binary descriptor. The experimental results show that the WLIB-SIFT descriptor has better accuracy and faster speed than the SIFT descriptor. Compared with the B-SIFT descriptor, the WLIB-SIFT descriptor has better robustness and distinctiveness at almost the same speed.

Original languageEnglish
Title of host publication2019 2nd IEEE International Conference on Information Communication and Signal Processing, ICICSP 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages379-383
Number of pages5
ISBN (Electronic)9781728151021
DOIs
Publication statusPublished - Sept 2019
Event2nd IEEE International Conference on Information Communication and Signal Processing, ICICSP 2019 - Weihai, China
Duration: 28 Sept 201930 Sept 2019

Publication series

Name2019 2nd IEEE International Conference on Information Communication and Signal Processing, ICICSP 2019

Conference

Conference2nd IEEE International Conference on Information Communication and Signal Processing, ICICSP 2019
Country/TerritoryChina
CityWeihai
Period28/09/1930/09/19

Keywords

  • B-SIFT
  • Feature descriptor
  • Image matching
  • SIFT
  • WLIB-SIFT

Fingerprint

Dive into the research topics of 'WLIB-SIFT: A Distinctive Local Image Feature Descriptor'. Together they form a unique fingerprint.

Cite this