DocumentCode :
538085
Title :
Multiscale segmentation based on mode-shift clustering
Author :
Tarnawski, Wojciech ; Miroslaw, Lukasz ; Pawlikowski, Roman ; Ociepa, Krzysztof
Author_Institution :
Syst. & Comput. Networks, Wroclaw Univ. of Technol., Wroclaw, Poland
fYear :
2010
fDate :
18-20 Oct. 2010
Firstpage :
129
Lastpage :
133
Abstract :
We present a novel segmentation technique that effectively segments natural images. The method is designed for the purpose of image retrieval and follows the principle of clustering the regions visible in the image. The concept is based on the multiscale approach where the image undergoes a number of diffusions. The algorithm has been visually compared with a reference segmentation.
Keywords :
image retrieval; image segmentation; pattern clustering; image retrieval; mode shift clustering; natural image segmentation; Anisotropic magnetoresistance; Image color analysis; Image segmentation; Kernel; Measurement; Pixel; Prototypes;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Information Technology (IMCSIT), Proceedings of the 2010 International Multiconference on
Conference_Location :
Wisla
ISSN :
2157-5525
Print_ISBN :
978-1-4244-6432-6
Type :
conf
DOI :
10.1109/IMCSIT.2010.5679957
Filename :
5679957
Link To Document :
بازگشت