Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Diversifying Demining: An Experimental Crowdsourcing Method for Optical Mine Detection
Linköping University, Department of Electrical Engineering.
2008 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesisAlternative title
Diversifiering av minröjning : En experimentell crowdsourcingmetod för optisk mindetektering (Swedish)
Abstract [en]

This thesis explores the concepts of crowdsourcing and the ability of diversity, applied to optical mine detection. The idea is to use the human eye and wide and diverse workforce available on the Internet to detect mines, in addition to computer algorithms.

The theory of diversity in problem solving is discussed, especially the Diversity Trumps Ability Theorem and the Diversity Prediction Theorem, and how they should be carried out for possible applications such as contrast interpretation and area reduction respectively.

A simple contrast interpretation experiment is carried out comparing the results of a laymen crowd and one of experts, having the crowds examine extracts from hyperspectral images, classifying the amount of objects or mines and the type of terrain. Due to poor participation rate of the expert group, and an erroneous experiment introduction, the experiment does not yield any statistically significant results. Therefore, no conclusion is made.

Experiment improvements are proposed as well as possible future applications.

Abstract [sv]

Denna rapport går igenom tanken bakom crowdsourcing och mångfaldens styrka tillämpad på optisk mindetektering. Tanken är att använda det mänskliga ögat och Internets skiftande och varierande arbetsstyrka som ett tillägg för att upptäcka minor tillsammans med dataalgoritmer.

Mångfaldsteorin i problemlösande diskuteras och speciellt ''Diversity Trumps Ability''-satsen och ''Diversity Prediction''-satsen och hur de ska genomföras för tillämpningar som kontrastigenkänning respektive ytreduktion.

Ett enkelt kontrastigenkänningsexperiment har genomförts för att jämföra resultaten mellan en lekmannagrupp och en expertgrupp. Grupperna tittar på delar av data från hyperspektrala bilder och klassifierar andel objekt eller minor och terrängtyp. På grund av lågt deltagande från expertgruppen och en felaktig experimentintroduktion ger inte experimentet några statistiskt signifikanta resultat, varför ingen slutsats dras.

Experimentförbättringar och framtida tillämpningar föreslås.

Place, publisher, year, edition, pages
2008. , 57 p.
Keyword [en]
Crowdsourcing, diversity, mine detection, image analysis
Keyword [sv]
Crowdsourcing, mindetektion, minspaning, mångfald, bildanalys
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:liu:diva-15813ISRN: LiTH-ISY-EX--08/4069--SEOAI: oai:DiVA.org:liu-15813DiVA: diva2:127438
Subject / course
Information Coding
Presentation
Systemet, B-huset, LiU Campus Valla, Linköping (Swedish)
Uppsok
Technology
Supervisors
Examiners
Projects
Multi Optical Mine Detection System
Available from: 2008-12-10 Created: 2008-12-06 Last updated: 2012-04-24Bibliographically approved

Open Access in DiVA

fulltext(0 kB)369 downloads
File information
File name FULLTEXT01.pdfFile size 0 kBChecksum SHA-512
ea43a9c0cde7fc6728b2fe95ebd2af1400c837512430214229b27b536a4c63dfa66bf54383ef71244147ddb039bbc497edb3be59ba0f6df6e5fb76d8d03a59ac
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Andersson, David
By organisation
Department of Electrical Engineering
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 369 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 486 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf