Hi! William Levak wrote: > You have to be careful here not to adhere too strictly to the > mathematical. The human eye and brain does not interpret colors > according to what is mathematically the best fit. Yellows and flesh > tones are interpreted by the eye as wrong when only slightly off. > > Most TV's distort the color a little to emphasize flesh tones at the > expense of other colors. Well, I'd be happy if this thing worked just on the pure mathematical basis ;-). Later it could be possible to enhance it (since, as I understand, perception should be possible to model with some tricks where the program calculates differences (errors) between the colors. Regarding the TV color distortion: I haven't noticed this on PAL TVs (PAL has another artifacts, especially at areas where Y has high freq components). I can sometimes recognize TV programs that were recorded in NTSC (American or Asian TV transmissions). The color difference is noticeable, especially (as you ? said) in the green area (somehow, the colors seem to be 'green'ish, and the picture misses 'highlights' (sky is 'darker' than on PAL records), but that may also be because of signal conversion). This program works in pure RGB color space; HSB would be more close to the human eye, but I'm not familiar with it, and as I said I'm currently rather focusing on presenting a 'working' routine; RGB is simpler to deal with. (But if someone knows how to deal with HSB, I'd be happy to know that). Best regards, L. - This message was sent through the cbm-hackers mailing list. To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tml.hut.fi.
Archive generated by hypermail 2.1.1.