hrcak mascot   Srce   HID

Izvorni znanstveni članak
https://doi.org/10.2498/cit.1002168

Unsupervised Color Image Segmentation Based on Non Parametric Clustering

Imène Kirati ; Department of Computer Science, Univesity of Badji Mokhtar, Annaba, Algeria
Yamina Tlili ; Department of Computer Science, Univesity of Badji Mokhtar, Annaba, Algeria

Puni tekst: engleski, PDF (738 KB) str. 247-254 preuzimanja: 893* citiraj
APA 6th Edition
Kirati, I. i Tlili, Y. (2013). Unsupervised Color Image Segmentation Based on Non Parametric Clustering. Journal of computing and information technology, 21 (4), 247-254. https://doi.org/10.2498/cit.1002168
MLA 8th Edition
Kirati, Imène i Yamina Tlili. "Unsupervised Color Image Segmentation Based on Non Parametric Clustering." Journal of computing and information technology, vol. 21, br. 4, 2013, str. 247-254. https://doi.org/10.2498/cit.1002168. Citirano 20.09.2020.
Chicago 17th Edition
Kirati, Imène i Yamina Tlili. "Unsupervised Color Image Segmentation Based on Non Parametric Clustering." Journal of computing and information technology 21, br. 4 (2013): 247-254. https://doi.org/10.2498/cit.1002168
Harvard
Kirati, I., i Tlili, Y. (2013). 'Unsupervised Color Image Segmentation Based on Non Parametric Clustering', Journal of computing and information technology, 21(4), str. 247-254. https://doi.org/10.2498/cit.1002168
Vancouver
Kirati I, Tlili Y. Unsupervised Color Image Segmentation Based on Non Parametric Clustering. Journal of computing and information technology [Internet]. 2013 [pristupljeno 20.09.2020.];21(4):247-254. https://doi.org/10.2498/cit.1002168
IEEE
I. Kirati i Y. Tlili, "Unsupervised Color Image Segmentation Based on Non Parametric Clustering", Journal of computing and information technology, vol.21, br. 4, str. 247-254, 2013. [Online]. https://doi.org/10.2498/cit.1002168

Sažetak
Many segmentation problems have been addressed using probabilistic modeling. These methods tend to estimate the region membership probabilities for each pixel of the image. The segmentation results depend strongly on the initialization of these regions and the selection of the appropriate number of segments. In this paper we present an unsupervised segmentation method based on non parametric clustering able to deal with these two issues. After a simple splitting, a minimum variance criterion is used to generate both the initial regions and their number. The proposed model was applied on various images (synthetic, natural) showing good visual results. Finally numerical experiments demonstrate the efficiency and the robustness of the proposed model compared to other segmentation methods.

Ključne riječi
image segmentation; non parametric clustering; class initialization; homogeneity

Hrčak ID: 114776

URI
https://hrcak.srce.hr/114776

Posjeta: 1.071 *