Entropy and Gibbs distribution in image processing : an historical perspective

• Autorzy: Battiato S.
• Języki publikacji: EN

Abstrakty

EN

This paper presents an historical overview about the entropy and its applications for the solution of inferential statistical problems in image processing. This survey covers some of the more important entropy-based research approaches. A brief introduction to the mathematical details and foundations about the basic concepts of Markov Random Fields (MRF} and related Gibbs sampling is also given. The information entropy is a mathematical measure of information or uncertainty derived from a probabilistic model. The paper starting from the seminal works of C. Shannon and of E.T. Javnes and of S. Geman and D. German discusses results obtained using different related techniques in image restoration, analysis and synthesis of textures and saliency maps construction. The paper moreover gives useful suggestions about the trend of development in future research

Słowa kluczowe

EN

Gibbs distribution; image processing; Markov random field; Shannon's entropy

PL

entropia Shannona; pole losowe Markowa; przetwarzanie obrazów; rozkład Gibbsa

• Wydawca: Instytut Telekomunikacji i Informatyki Uniwersytetu Technologiczno-Przyrodniczego w Bydgoszczy
• Czasopismo: Image Processing & Communications
• Rocznik: 2001
• Tom: Vol. 7, no 3-4
• Strony: 35--50
• Opis fizyczny: Bibliogr. 34 poz.,Rys., wykr.,

Twórcy

autor

Battiato S.