Vision based event classification in robotic micromanipulation

Zoran Cenev; Janne Venäläinen; Quan Zhou

doi:10.1109/MARSS.2017.8001923

Vision based event classification in robotic micromanipulation

Zoran Cenev, Janne Venäläinen, Quan Zhou

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-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 language	English
Title of host publication	Proceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-5386-0346-8
DOIs	https://doi.org/10.1109/MARSS.2017.8001923
Publication status	Published - 2017
MoE publication type	A4 Article in a conference publication
Event	International Conference on Manipulation, Automation and Robotics at Small Scales - University of Montreal, Montreal, Canada Duration: 17 Jul 2017 → 21 Jul 2017 http://marss-conference.org/

Conference

Conference	International Conference on Manipulation, Automation and Robotics at Small Scales
Abbreviated title	MARSS
Country	Canada
City	Montreal
Period	17/07/2017 → 21/07/2017
Internet address	http://marss-conference.org/

Keywords

event classification
micromanipulation
machine vision
supervised learning

Access to Document

10.1109/MARSS.2017.8001923

CenevVenäläinenZhou_MARSS-2017Accepted author manuscript, 1 MB

Cite this

@inproceedings{812b06b94a9a4d71a8e00332a93cbcf3,

title = "Vision based event classification in robotic micromanipulation",

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 %). ",

keywords = "event classification, micromanipulation, machine vision, supervised learning",

author = "Zoran Cenev and Janne Ven{\"a}l{\"a}inen and Quan Zhou",

year = "2017",

doi = "10.1109/MARSS.2017.8001923",

language = "English",

booktitle = "Proceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017",

publisher = "IEEE",

address = "United States",

note = "International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS ; Conference date: 17-07-2017 Through 21-07-2017",

url = "http://marss-conference.org/",

}

Cenev, Z, Venäläinen, J & Zhou, Q 2017, Vision based event classification in robotic micromanipulation. in Proceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017., 8001923, IEEE, International Conference on Manipulation, Automation and Robotics at Small Scales, Montreal, Canada, 17/07/2017. https://doi.org/10.1109/MARSS.2017.8001923

Vision based event classification in robotic micromanipulation. / Cenev, Zoran; Venäläinen, Janne; Zhou, Quan.

Proceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017. IEEE, 2017. 8001923.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Vision based event classification in robotic micromanipulation

AU - Cenev, Zoran

AU - Venäläinen, Janne

AU - Zhou, Quan

PY - 2017

Y1 - 2017

N2 - 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 %).

AB - 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 %).

KW - event classification

KW - micromanipulation

KW - machine vision

KW - supervised learning

U2 - 10.1109/MARSS.2017.8001923

DO - 10.1109/MARSS.2017.8001923

M3 - Conference contribution

BT - Proceedings of the 2017 International Conference on Manipulation, Automation and Robotics at Small Scales, MARSS 2017

PB - IEEE

T2 - International Conference on Manipulation, Automation and Robotics at Small Scales

Y2 - 17 July 2017 through 21 July 2017

ER -

Vision based event classification in robotic micromanipulation

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this