DynaScale: An Intelligent Image Scale Selection Framework for Visual Matching in Smart IoT

Bo Lung Tsai, Kwei Jay Lin, Yuheng Cao, Yu Meng

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

1 Scopus citations

Abstract

Localization is one of the fundamental technologies to enable location-Aware services in smart cities. Image feature matching plays an essential role in visual-based localization for moving IoT devices to navigate in various scenes. Conventional matching pipelines have issues with finding an accurate transformation model when a pair of to-be-matched images have a huge scale difference between the views of interested objects. The paper introduces DynaScale, a general framework that integrates existing image detection and matching algorithms with constructed image pyramids for extended matching, and selects the best matching result from the image pairs of different scales. We have designed an intelligent evaluation scheme of potential transformation models based on various tests including reasonable projections, the resembled size of region proposals, and similarities of bounded descriptors. The experimental result shows that, on selected datasets, DynaScale has 1.9 times better mean matching accuracy than existing methods from 24.32% to 45.91%, and produces about twice the number of useful, correctly matched frames in a moving robot's video stream.

Original languageEnglish
Title of host publicationProceedings - 2020 IEEE 22nd International Conference on High Performance Computing and Communications, IEEE 18th International Conference on Smart City and IEEE 6th International Conference on Data Science and Systems, HPCC-SmartCity-DSS 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1166-1173
Number of pages8
ISBN (Electronic)9781728176499
DOIs
StatePublished - 12 2020
Externally publishedYes
Event22nd IEEE International Conference on High Performance Computing and Communications, 18th IEEE International Conference on Smart City and 6th IEEE International Conference on Data Science and Systems, HPCC-SmartCity-DSS 2020 - Virtual, Fiji, Fiji
Duration: 14 12 202016 12 2020

Publication series

NameProceedings - 2020 IEEE 22nd International Conference on High Performance Computing and Communications, IEEE 18th International Conference on Smart City and IEEE 6th International Conference on Data Science and Systems, HPCC-SmartCity-DSS 2020

Conference

Conference22nd IEEE International Conference on High Performance Computing and Communications, 18th IEEE International Conference on Smart City and 6th IEEE International Conference on Data Science and Systems, HPCC-SmartCity-DSS 2020
Country/TerritoryFiji
CityVirtual, Fiji
Period14/12/2016/12/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Keywords

  • image matching
  • scale selection
  • transformation evluation

Fingerprint

Dive into the research topics of 'DynaScale: An Intelligent Image Scale Selection Framework for Visual Matching in Smart IoT'. Together they form a unique fingerprint.

Cite this