1-D local binary patterns for onset detection of myoelectric signals

Paul McCool; Navin Chatlani; Lykourgos Petropoulakis; John Soraghan; Radhika Menon; Heba Lakany

1-D local binary patterns for onset detection of myoelectric signals

Paul McCool, Navin Chatlani, Lykourgos Petropoulakis, John Soraghan, Radhika Menon, Heba Lakany

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed)

Abstract

This paper presents a new 1-D LBP (Local Binary Pattern) based technique for onset detection. The algorithm is tested on forearm surface myoelectric signals that occur due to lower arm gestures. Unlike other onset detection algorithms, the method does not require manual threshold setting and fine-tuning, which makes it faster and easier to
implement. The only variables are window size, histogram type and the number of histogram bins. It is also not necessary to measure the properties of the signal during a
quiescent period before the algorithm can be used. 1-D LBP Onset Detection is compared with single and double threshold methods and is shown to be more robust and accurate.

Original language	English
Title of host publication	2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO)
Publisher	IEEE
Pages	499-503
Number of pages	5
ISBN (Print)	978-1-4673-1068-0
Publication status	Published - 28 Aug 2012
Event	20th European Signal Processing Conference - Bukarest, Romania Duration: 27 Sept 2012 → 1 Oct 2012

Conference

Conference	20th European Signal Processing Conference
Country/Territory	Romania
City	Bukarest
Period	27/09/12 → 1/10/12

Keywords

signal processing
prosthesis
local binary patterns
detection
myoelectric signals
myoelectric signals
surface electromyography
manual threshold setting
histogram

Cite this

@inbook{02f720f3e6174b7dbb879f82c56af2a6,

title = "1-D local binary patterns for onset detection of myoelectric signals",

abstract = "This paper presents a new 1-D LBP (Local Binary Pattern) based technique for onset detection. The algorithm is tested on forearm surface myoelectric signals that occur due to lower arm gestures. Unlike other onset detection algorithms, the method does not require manual threshold setting and fine-tuning, which makes it faster and easier to implement. The only variables are window size, histogram type and the number of histogram bins. It is also not necessary to measure the properties of the signal during a quiescent period before the algorithm can be used. 1-D LBP Onset Detection is compared with single and double threshold methods and is shown to be more robust and accurate.",

keywords = "signal processing, prosthesis, local binary patterns , detection, myoelectric signals, myoelectric signals , surface electromyography, manual threshold setting, histogram",

author = "Paul McCool and Navin Chatlani and Lykourgos Petropoulakis and John Soraghan and Radhika Menon and Heba Lakany",

year = "2012",

month = aug,

day = "28",

language = "English",

isbn = "978-1-4673-1068-0",

pages = "499--503",

booktitle = "2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO)",

publisher = "IEEE",

note = "20th European Signal Processing Conference ; Conference date: 27-09-2012 Through 01-10-2012",

}

TY - CHAP

T1 - 1-D local binary patterns for onset detection of myoelectric signals

AU - McCool, Paul

AU - Chatlani, Navin

AU - Petropoulakis, Lykourgos

AU - Soraghan, John

AU - Menon, Radhika

AU - Lakany, Heba

PY - 2012/8/28

Y1 - 2012/8/28

N2 - This paper presents a new 1-D LBP (Local Binary Pattern) based technique for onset detection. The algorithm is tested on forearm surface myoelectric signals that occur due to lower arm gestures. Unlike other onset detection algorithms, the method does not require manual threshold setting and fine-tuning, which makes it faster and easier to implement. The only variables are window size, histogram type and the number of histogram bins. It is also not necessary to measure the properties of the signal during a quiescent period before the algorithm can be used. 1-D LBP Onset Detection is compared with single and double threshold methods and is shown to be more robust and accurate.

AB - This paper presents a new 1-D LBP (Local Binary Pattern) based technique for onset detection. The algorithm is tested on forearm surface myoelectric signals that occur due to lower arm gestures. Unlike other onset detection algorithms, the method does not require manual threshold setting and fine-tuning, which makes it faster and easier to implement. The only variables are window size, histogram type and the number of histogram bins. It is also not necessary to measure the properties of the signal during a quiescent period before the algorithm can be used. 1-D LBP Onset Detection is compared with single and double threshold methods and is shown to be more robust and accurate.

KW - signal processing

KW - prosthesis

KW - local binary patterns

KW - detection

KW - myoelectric signals

KW - surface electromyography

KW - manual threshold setting

KW - histogram

UR - http://www.eurasip.org/Proceedings/Eusipco/Eusipco2012/Conference/papers/1569581907.pdf

UR - http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6334024&contentType=Conference+Publications

M3 - Chapter (peer-reviewed)

SN - 978-1-4673-1068-0

SP - 499

EP - 503

BT - 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO)

PB - IEEE

T2 - 20th European Signal Processing Conference

Y2 - 27 September 2012 through 1 October 2012

ER -

1-D local binary patterns for onset detection of myoelectric signals

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this