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
Feature extraction from images with augmented feature inputs
KTH, School of Computer Science and Communication (CSC).
2017 (English)Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE creditsStudent thesisAlternative title
Särdragsextrahering från bilder med förstärkt särdragsinmatning (Swedish)
Abstract [en]

Machine learning models for visual recognition tasks such as image recognition is a common research area as of lately. However, not much research has been made when multiple features are to be extracted from the same input. This thesis researches if and how knowledge about one feature influences model performance of a model classifying another feature, as well as how the similarity and generality of the feature data distributions influences model performance.

Incorporating augmentation inputs in the form of extra feature information in image models was found to yield different results depending on feature data distribution similarity and level of generality. Care must be taken when augmenting with features in order for the feature not to be completely redundant or to completely take over in the learning process. Selecting reasonable augmentation inputs might yield desired synergy effects which influences model performance to the better.

Abstract [sv]

Maskininlärningsmodeller för uppgifter inom visuellt igenkännande så som bildigenkänning har på senaste tiden varit ett vanligt forskningsområde. Dock har inte mycket forskning fokuserats på att extrahera multipla särdrag från samma inmatning. Detta examensarbete syftar till att undersöka hur kunskap om ett särdrag influerar en modells prestanda som syftar till att klassificera ett annat särdrag, men även hur likhet och generalitet i särdragens datadistribution influerar modellprestanda.

Integrering av förstärkande inmatning i form av extra särdragsinformation i bildklassificeringsmodeller visades ge olika resultat beroende på likhet och generalitet av distribution av särdragsdata. Hänsyn måste tas när förstärkande särdrag används för att de förstärkande särdragen inte ska bli helt redundanta eller helt ta över under träningsprocessen. Väljande av rimliga förstärkningssärdrag kan medföra önskade synergieffekter vilket påverkar modellprestandan till det bättre.

Place, publisher, year, edition, pages
2017.
Keywords [en]
neural networks, machine learning, feature augmentation
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:kth:diva-219073OAI: oai:DiVA.org:kth-219073DiVA, id: diva2:1161950
External cooperation
Sellpy
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2017-12-21 Created: 2017-12-01 Last updated: 2018-01-13Bibliographically approved

Open Access in DiVA

fulltext(7467 kB)54 downloads
File information
File name FULLTEXT01.pdfFile size 7467 kBChecksum SHA-512
9eecdb20a5926acf3e6eec88a6909cf0d392735deaab9b6a3c21d7b876b5b1de6a09436f0b5146397841241e15fa2c51f4e8ac4739aa96935e3cee3f686818d0
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Drangel, Andreas
By organisation
School of Computer Science and Communication (CSC)
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 54 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: 106 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