A hybrid neural network used for the colour conversion from RGB to CMY and a new method for choosing a learning set

• Autorzy: Horzyk, A.
• Języki publikacji: EN

Abstrakty

EN

This paper presents the application of the neural network to the colour conversion problem from RGB to CMY [1]. Three different architecture of neural network are compared and the one which gives the best results is chosen, and the method of learning this network based on the back propagation method with a controlled process of change of learning parametres, particulary the number of patterns in a learning set, in order to shorten the time of learning, is presented.

Słowa kluczowe

PL

sieć neuronowa

EN

back propagation; change of the number of patterns of a learning set; CMY; colour conversion; hybrid architecture of a layer neural network; learning method; neural network; RGB

• Czasopismo: Image Processing & Communications
• Rocznik: 1998
• Tom: Vol. 4, no 1-2
• Strony: 37-46
• Opis fizyczny: Bibliogr. 5 poz.,Rys., wykr.,

Twórcy

autor

Horzyk, A.

• Institute of Automatics, University of Mining and Metallurgy, 30-835 Kraków, ul. Teligi 6/55, Poland,