Spectral photographySpectral photographyΦασματική φωτογραφία Διπλωματική Εργασία Diploma Work 2018-10-012018enA common topic which concerns the color industry is the way to optimally reproduce color stimulus identical to the color we are capable of seeing. With spectroscopic technology we can measure surface spectral reflectances of Colorimetric Charts which are the golden standards in color science. Multiple light sources are acquired to form a Spectral Power Distribution Database that could help extract the spectral signatures of different illumination scenarios. Using this Database and exploiting the spectral information features of both the Visible and Near Infrared spectrum each illuminant type has, we can spectrally match the given illumination and reproduce accurately color from spectral data. Using spectral comparison metrics, illuminant estimation is achieved and color difference measurements across patches on spectrally reconstructed images will lead to the creation of true, according to human vision, computational color with illumination independency.http://creativecommons.org/licenses/by/4.0/Πολυτεχνείο Κρήτης::Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών ΥπολογιστώνGeorgiou_Georgios_Dip_2018.pdfChania [Greece]Library of TUC2018-10-01application/pdf8.8 MBembargo Georgiou Georgios Γεωργιου Γεωργιος Balas Costas Μπαλας Κωστας Samoladas Vasilis Σαμολαδας Βασιλης Zervakis Michail Ζερβακης Μιχαηλ Πολυτεχνείο Κρήτης Technical University of Crete Spectral cube Illuminant estimation NIR imaging Color reproduction Computational color