Vision based event classification in robotic micromanipulation

Zoran Cenev, Janne Venäläinen, Quan Zhou

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

155 Downloads (Pure)

Abstract

Autonomous positioning of small objects to create heterogeneous structures has great potential to advance the current micromanipulation procedures. To achieve autonomous micromanipulation, it is required to recognize the manipulation events. In this work, different classification algorithms including five common supervised learning methods are assessed for identifying states of manipulation. The classifiers are trained with data that consists of 3056 video frames and validated on 2545 videos frames. The best machine learning classifiers classified the events with 92.9 % accuracy, higher than the result of logic-based classification (88.9 %).
Original languageEnglish
Title of host publicationProceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-5386-0346-8
DOIs
Publication statusPublished - 2017
MoE publication typeA4 Article in a conference publication
EventInternational Conference on Manipulation, Automation and Robotics at Small Scales - University of Montreal, Montreal, Canada
Duration: 17 Jul 201721 Jul 2017
http://marss-conference.org/

Conference

ConferenceInternational Conference on Manipulation, Automation and Robotics at Small Scales
Abbreviated titleMARSS
CountryCanada
CityMontreal
Period17/07/201721/07/2017
Internet address

Keywords

  • event classification
  • micromanipulation
  • machine vision
  • supervised learning

Fingerprint

Dive into the research topics of 'Vision based event classification in robotic micromanipulation'. Together they form a unique fingerprint.

Cite this