Texture alteration detection in bitemporal images of lesions with psoriasis

AbstractThe objective of this work is to explore the feasibility of quantifying textural change between pairs of segmented patterns without registering them. The Multi-variate Alteration Detection (M.A.D.) Transform is applied to a texture model constructed with the data of segmented psoriasis lesions images. The texture model is Haralick's co-occurrence matrix, which is computed and normalized for each single band with the equalized data of a given lesion. The contribution of each single color band to the textural change is analyzed.
Keywordsco-occurrence matrix, M.A.D. Transform, Multivariate Alteration Detection, texture model, dermatological images, psoriasis
TypeTechnical report
Year2003    pp. 30
PublisherInformatics and Mathematical Modelling, Technical University of Denmark, DTU
AddressRichard Petersens Plads, Building 321, DK-2800 Kgs. Lyngby
SeriesIMM-Technical Report-2003-8
Electronic version(s)[pdf]
BibTeX data [bibtex]
IMM Group(s)Image Analysis & Computer Graphics