<p>U okviru disertacije predloženo je rešenje za određivanje kolorimetrijskih<br />vrednosti otisaka štampanih goniohromatskim pigmentima na osnovu odziva<br />digitalne kamere. Predmet rada predstavljao je razvoj modela karakterizacije<br />kamere prilagođenog za fitovanje više-ugaonih podataka, kao i ispitivanje<br />uticaja parametara predložene metodologije na tačnost procene vrednosti<br />boja kamerom. Razvijeni model, baziran na veštačkim neuronskim mrežama,<br />omogućio je postizanje zadovoljavajuće preciznosti merenja boja, procenu<br />vrednosti boja svih testiranih mernih geometrija na osnovu snimaka u jednom<br />mernom uglu, a pokazao je i visok stepen adaptivnosti na promenu osvetljenja<br />koje se prilikom merenja koristi. Model je optimizovan primenom genetskog<br />algoritma, čime je njegova efikasnost znatno unapređena.</p> / <p>The thesis proposes а solution for colorimetric characterization of prints enhanced<br />with goniochromatic pigments by means of a digital camera. The subject of the<br />research was the development of a camera characterization model adapted to fit<br />multi-angular data and the assessment of the proposed framework parameters<br />impact on the accuracy of camera-based color measurement. The developed<br />model, based on artificial neural networks, enabled accurate color measurement<br />with a satisfactory level of accuracy, estimation of color values of all analyzed<br />measurement geometries on the basis of images obtained in one detection angle,<br />and was proved to be very adaptive to the change of the illuminant used during the<br />measurement. The model was optimized by means of a genetic algorithm, which<br />led to the significant improvement of its efficiency.</p>