Skip to main navigation Skip to search Skip to main content

A novel detection algorithm of microspheres based on Hough transform and ellipse fitting

  • Harbin Institute of Technology

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

Abstract

Microspheres are used in many applications but their detection is often impeded due to susceptibility to disturbance and low accuracy. An autonomous algorithm was proposed based on Hough transform and least squares ellipse fitting to achieve coarse to fine detection of microspheres. The algorithm is robust and accurate in detection of microspheres' center positons, radii, roundness, deficiency and adhesion, which facilitates the monitoring and manipulation of microspheres. The experiments are conducted to verify the performance of the algorithm. The results demonstrate a nearly 100% successful detection rate and less than 10 μm absolute deviation of radius. The average detection speed is about 100 microspheres per minute.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Mechatronics and Automation, ICMA 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages999-1004
Number of pages6
ISBN (Electronic)9781479970964
DOIs
StatePublished - 2 Sep 2015
Event12th IEEE International Conference on Mechatronics and Automation, ICMA 2015 - Beijing, China
Duration: 2 Aug 20155 Aug 2015

Publication series

Name2015 IEEE International Conference on Mechatronics and Automation, ICMA 2015

Conference

Conference12th IEEE International Conference on Mechatronics and Automation, ICMA 2015
Country/TerritoryChina
CityBeijing
Period2/08/155/08/15

Keywords

  • Hough transform
  • ellipse fitting
  • feature extraction
  • microspheres

Fingerprint

Dive into the research topics of 'A novel detection algorithm of microspheres based on Hough transform and ellipse fitting'. Together they form a unique fingerprint.

Cite this