Texture classification from single uncalibrated images: Random matrix theory approach

Publikation: Kapitel i bog/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Abstrakt

We studied the problem of classifying textured-materials from their single-imaged appearance, under general viewing and illumination conditions, using the theory of random matrices. To evaluate the performance of our algorithm, two distinct databases of images were used: The CUReT database and our database of colorectal polyp images collected from patients undergoing colon capsule endoscopy for early cancer detection. During the learning stage, our classifier algorithm established the universality laws for the empirical spectral density of the largest singular value and normalized largest singular value of the image intensity matrix adapted to the eigenvalues of the information-plus-noise model. We showed that these two densities converge to the generalized extreme value (GEV-Frechet) and Gaussian G 1 distribution with rate O(N1/2), respectively. To validate the algorithm, we introduced a set of unseen images to the algorithm. Misclassification rate of approximately 1%-6%, depending on the database, was obtained, which is superior to the reported values of 5%-45% in previous research studies.

OriginalsprogEngelsk
Titel27th IEEE International Workshop on Machine Learning for Signal Processing
Antal sider6
ForlagIEEE Press
Publikationsdato2017
ISBN (Trykt)978-1-5090-6342-0
ISBN (Elektronisk)978-1-5090-6341-3
DOI
StatusUdgivet - 2017
Begivenhed27th International Workshop on Machine Learning for Signal Processing - Tokyo, Japan
Varighed: 25. sep. 201728. sep. 2017
Konferencens nummer: 27

Workshop

Workshop27th International Workshop on Machine Learning for Signal Processing
Nummer27
Land/OmrådeJapan
ByTokyo
Periode25/09/201728/09/2017
NavnMachine Learning for Signal Processing
Vol/bind2017
ISSN1551-2541

Fingeraftryk

Dyk ned i forskningsemnerne om 'Texture classification from single uncalibrated images: Random matrix theory approach'. Sammen danner de et unikt fingeraftryk.

Citationsformater