A comparative study on chrominance based methods in dorsal hand recognition: Single image case

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62690094%3A18450%2F18%3A50014705" target="_blank" >RIV/62690094:18450/18:50014705 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-319-92058-0_68" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-319-92058-0_68</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-92058-0_68" target="_blank" >10.1007/978-3-319-92058-0_68</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A comparative study on chrominance based methods in dorsal hand recognition: Single image case

Popis výsledku v původním jazyce

Dorsal hand recognition is a crucial topic in biometrics and human-machine interaction; however most of the identification systems identify and segment the hands from the images consisting of high contrast backgrounds. In other words, capturing and analyzing images of hands on a white or black or any colored background is way too easy to achieve high accuracy. On the contrary, in continuous authentication or in interactive human-machine systems, it can be not possible nor feasible to process high contrast images, like hands on computer keyboards which is not as simple as single color backgrounds even the feature to be extracted is solely the hand color. Therefore we deal with processing of the images consisting of hands on computer keyboards to compare various luminance and chrominance methods by YCbCr color space extraction and to find ways to achieve higher accuracy without any succeeding erosion, dilation or filtering. The methods focused on chromatic intervals could be summarized as: fixed intervals, covariance intervals and fuzzy 2-means. Our main contribution briefly is a necessary accuracy comparison and validation of the common methods on the single images. The highest accuracy is found as 96% by fuzzy 2-means applied to chrominance layers of the image.

Název v anglickém jazyce

A comparative study on chrominance based methods in dorsal hand recognition: Single image case

Popis výsledku anglicky

Dorsal hand recognition is a crucial topic in biometrics and human-machine interaction; however most of the identification systems identify and segment the hands from the images consisting of high contrast backgrounds. In other words, capturing and analyzing images of hands on a white or black or any colored background is way too easy to achieve high accuracy. On the contrary, in continuous authentication or in interactive human-machine systems, it can be not possible nor feasible to process high contrast images, like hands on computer keyboards which is not as simple as single color backgrounds even the feature to be extracted is solely the hand color. Therefore we deal with processing of the images consisting of hands on computer keyboards to compare various luminance and chrominance methods by YCbCr color space extraction and to find ways to achieve higher accuracy without any succeeding erosion, dilation or filtering. The methods focused on chromatic intervals could be summarized as: fixed intervals, covariance intervals and fuzzy 2-means. Our main contribution briefly is a necessary accuracy comparison and validation of the common methods on the single images. The highest accuracy is found as 96% by fuzzy 2-means applied to chrominance layers of the image.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Lecture notes in computer science

ISBN

978-3-319-92057-3

ISSN

0302-9743

e-ISSN

neuvedeno

Počet stran výsledku

11

Strana od-do

711-721

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Montreal

Datum konání akce

25. 6. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—