Enhanced hand shape identification using random forests

El Sayed M. El-Alfy*

*Corresponding author for this work

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

Abstract

Over the past ten years, there has been a growing interest in hand-based recognition in biometric technology systems. In this paper, we investigated the application of random decision tree forests for hand identification using geometric hand measurements. We evaluated and compared the performance of the proposed method using out-of-bag validation and 10-fold cross validation in terms of identification. We also studied the impact of the forest size on the performance. The experimental results showed significant improvement over single decision trees, rule-based and nearest-neighbor machine learning algorithms.

Original languageEnglish
Title of host publicationNeural Information Processing - 20th International Conference, ICONIP 2013, Proceedings
Pages441-447
Number of pages7
EditionPART 2
DOIs
StatePublished - 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume8227 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Keywords

  • Biometrics
  • Decision trees
  • Geometric features
  • Hand recognition
  • Machine learning
  • Random forests

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Enhanced hand shape identification using random forests'. Together they form a unique fingerprint.

Cite this