Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Extrahering av återkommande beteendemotiv från videoinspelningar av naturligt beteende
KTH, School of Engineering Sciences (SCI).
KTH, School of Engineering Sciences (SCI).
2019 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Extraction of recurring behavioral motifs from video recordings of natural behavior (English)
Abstract [sv]

I denna rapport visar vi hur estimering av dolda Markov-modeller kan användas för att konstruera modeller för beteenden hos möss. Vi använde ett neuralt nätverk för att extrahera positionsdata för olika kroppsdelar i videoinspelningar av enskilda möss i en inhängnad. Därefter beräknade vi egenskaper såsom hastighet och kroppslängd ur positionsdatan och använde en implementation av Baum-Welch-algoritmen för att passa dolda Markov-modeller till egenskapsdatan. Vi kunde identifiera återkommande beteenden såsom "springer längs med väggen" och "undersöker väggen" som estimerade tillstånd hos flera möss, vilket överenstämmer med vad vi kunde iakta i videomaterialet. Därigenom visar vi att estimering av dolda Markov-modeller med hjälp av Baum-Welch-algoritmen kan användas för att automatiskt hitta modeller av beteenden hos möss.

 

Abstract [en]

In this report we demonstrate the usefulness of hidden Markov model estimation as a method to construct models of mouse behavior. We used a neural network to retrieve positional data of different body parts from overhead video recordings of lone mice in an enclosure. We then extracted features such as velocity and elongation from the positional data and used an implementation of the Baum-Welch algorithm to fit hidden Markov models to the feature data. We could identify recurring behaviors such as "running next to wall" and "investigating wall" among the estimated states in several different mice, which was consistent with what we could see in the actual videos. We thereby demonstrate that hidden Markov model estimation by the Baum-Welch algorithm can be utilized to automatically find models of mouse behavior.

 

Place, publisher, year, edition, pages
2019.
Series
TRITA-SCI-GRU ; 2019:125
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:kth:diva-255851OAI: oai:DiVA.org:kth-255851DiVA, id: diva2:1342442
Supervisors
Examiners
Available from: 2019-08-13 Created: 2019-08-13 Last updated: 2019-08-13Bibliographically approved

Open Access in DiVA

fulltext(8687 kB)10 downloads
File information
File name FULLTEXT01.pdfFile size 8687 kBChecksum SHA-512
08700e4b3cb0f92742d5c82fac8d36d8c17833817a0fc544edd8ad241bf9d9eb5e7d3fbae5a39db5d6cbf2ef3cc02bded14d09ada24eb278806c9a7a1d3c3e4f
Type fulltextMimetype application/pdf

By organisation
School of Engineering Sciences (SCI)
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 10 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

urn-nbn

Altmetric score

urn-nbn
Total: 25 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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