Gaussian muti-scale fast registration algorithm and its application in machine vision

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

1 Scopus citations

Abstract

To meet the real-time requirement of image processing in machine vision applications, a novel Gaussian Muti-scale Fast Registration (GMFR) algorithm is proposed. GMFR avoids the problem of losing details of images in sub-sampling process by means of Gaussian filtering based on muti-scale images. The advantage of image Cross-Correlation (CC) and Cross-Power Spectrum (CPS) are integrated by the presented method to achieve both accuracy and efficiency of image registration. The performance of GMFR is analyzed quantificationally based on the pre-defined performance function, which shows that the advantage of GMFR algorithm increases exponentially as the image gets larger. A machine vision environment is built for surface defects detection of lead frames and verifies the accuracy, robustness and high computing efficiency of GMFR in noisy image registration.

Original languageEnglish
Title of host publicationTesting and Measurement
Subtitle of host publicationTechniques and Applications - Proceedings of the 2015 International Conference on Testing and Measurement: Techniques and Applications, TMTA 2015
EditorsKennis Chan
PublisherCRC Press/Balkema
Pages325-328
Number of pages4
ISBN (Print)9781138028128
StatePublished - 2015
EventInternational Conference on Testing and Measurement: Techniques and Applications, TMTA 2015 - Phuket Island, Thailand
Duration: 16 Jan 201517 Jan 2015

Publication series

NameTesting and Measurement: Techniques and Applications - Proceedings of the 2015 International Conference on Testing and Measurement: Techniques and Applications, TMTA 2015

Conference

ConferenceInternational Conference on Testing and Measurement: Techniques and Applications, TMTA 2015
Country/TerritoryThailand
CityPhuket Island
Period16/01/1517/01/15

Fingerprint

Dive into the research topics of 'Gaussian muti-scale fast registration algorithm and its application in machine vision'. Together they form a unique fingerprint.

Cite this