Choosing serial tests for discrimination of textures in biomedical images

• Autorzy: Kulikowski J.K. , Wierzbicka D.
• Języki publikacji: EN

Abstrakty

EN

There are described several methods of the design of non-parametric serial statistical tests used for discrimination of textures. It is assumed that different textures are subjected to different but a priori unknown statistical distributions. The methods are based on linear ordering of multidimensional observation space. There are proposed several types of linear ordering based on the concepts of: reversible scanning, rosettes and permuted scale coding. The effectiveness of tests based on the proposed types of observation space ordering has been evaluated in a series of numerical experiments performed on real images of textures; the results of experiments are given in the paper with corresponding comments.

Słowa kluczowe

EN

statistical tests; image processing; discrimination of textures; linear ordering of observation space

PL

test statystyczny; przetwarzanie obrazu; dyskryminacja tekstur

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2005
• Tom: Vol. 25, no. 3
• Strony: 65--77
• Opis fizyczny: Bibliogr. 3 poz., rys., tab.

Twórcy

autor

Kulikowski J.K.

• Institute of Biocybernetics and Biomedical Engineering, Polish Academy of Sciences, 4 Ks. Trojdena Str., 02-109 Warsaw, Poland

autor

Wierzbicka D.

• Institute of Biocybernetics and Biomedical Engineering, Polish Academy of Sciences, Warsaw, Poland

Bibliografia

• [1] Wilks S.S.: Mathematical Statistics (Russian translation). Nauka, Moscow, 1967.
• [2] Runyon R.P.: Nonparametric Statistics. A Contemporary Approach. Addison-Wesley Publishing Company, Reading, Mass., 1977.
• [3] Kulikowski J.L., Wierzbicka D.: Texture Analysis Based on Application of Non-Parametric Serial Statistical Tests. Biocybemetics and Biomedical Engineering, 2004, 24, 2, 27-39.