Detailed Information on Publication Record
2014
RSurf Texture Descriptor
MAJTNER, Tomáš, Roman STOKLASA and David SVOBODABasic information
Original name
RSurf Texture Descriptor
Name in Czech
Popisovač textúry RSurf
Authors
MAJTNER, Tomáš (703 Slovakia, belonging to the institution), Roman STOKLASA (703 Slovakia, belonging to the institution) and David SVOBODA (203 Czech Republic, guarantor, belonging to the institution)
Edition
2014
Other information
Language
English
Type of outcome
Software
Field of Study
20206 Computer hardware and architecture
Country of publisher
Czech Republic
Confidentiality degree
není předmětem státního či obchodního tajemství
RIV identification code
RIV/00216224:14330/14:00074012
Organization unit
Faculty of Informatics
Keywords in English
image descriptor; Pattern recognition; RSurf; HEp-2 cells;
Technical parameters
Software pre výpis vlastností textúry snímku, ktoré je možné použiť pre rozpoznanie a následnú klasifikáciu vstupného obrázku. Program bol vyvinutý pre spracovanie obrázkov HEp-2 buniek nasnímaných fluorescenčným mikroskopom ale obecne je možné ho použiť pre ľubovolný vstup. Implementácia je realizovaná v jazyku C++.
Zodpovedné osoby: Tomáš Majtner <majtner@ics.muni.cz> a Roman Stoklasa <rstoki@seznam.cz>
Adresa: Fakulta informatiky Masarykovy univerzity, Botanická 68a, 602 00 Brno.
Tags
Tags
International impact
Změněno: 29/10/2014 13:48, RNDr. Ing. Bc. Tomáš Majtner, Ph.D.
Abstract
V originále
In biomedical image analysis, object description and classification tasks are very common. Our work relates to the problem of classification of Human Epithelial (HEp-2) cells. Since the crucial part of each classification process is the feature extraction and selection, much attention should be concentrated to the development of proper image descriptors. In this article, we introduce a new efficient texture-based image descriptor for HEp-2 images. We compare proposed descriptor with LBP, Haralick features (GLCM statistics) and Tamura features using the public MIVIA HEp-2 Images Dataset. Our descriptor outperforms all previously mentioned approaches and the kNN classifier based solely on the proposed descriptor achieve the accuracy as high as 91.1%.
Links
GBP302/12/G157, research and development project |
| ||
MUNI/A/0855/2013, interní kód MU |
|