Design of parallelized training system of single class cascade classifier

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

Abstract

This paper proposes a new training method of a cascade classifier in order to implement on Hadoop MapReduce platform. Learning process of cascade classifier requires many computations whereas the serialized algorithm does not fit to a parallel platform well. The parallelization is achieved by dividing the training into two parts. Before starting learning for adaptation to required false positive rate, the unit classifiers are trained independently using positive examples and small set of negative examples. To make a chain of classifiers, the latter part performs training using only negative examples.

Original languageEnglish
Title of host publication2015 IEEE 2nd International Conference on InformationScience and Security, ICISS 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467386111
DOIs
StatePublished - 4 Jan 2016
Event2nd IEEE International Conference on Information Science and Security, ICISS 2015 - Seoul, Korea, Republic of
Duration: 14 Dec 201516 Dec 2015

Publication series

Name2015 IEEE 2nd International Conference on InformationScience and Security, ICISS 2015

Conference

Conference2nd IEEE International Conference on Information Science and Security, ICISS 2015
Country/TerritoryKorea, Republic of
CitySeoul
Period14/12/1516/12/15

Keywords

  • Cascade classifier
  • MapReduce
  • Object detection
  • Parallel training

Fingerprint

Dive into the research topics of 'Design of parallelized training system of single class cascade classifier'. Together they form a unique fingerprint.

Cite this