Re: [Flashcoders] visually similar spectrum
I'm not sure if this will help, but try looking up hvs on apple's website. I know they use such a theory in calabrating their monitors. Karl Sent from losPhone On Nov 3, 2009, at 4:48 PM, Anthony Pace wrote: Well kind of a mix really, but I guess you are right for part of it; yet, I would like to know the math and theory involved, so I can make educated adjustments. So I guess the steps will be: -to reduce the quality of only certain sections of the frame; for, I want to decrease quality based on the amount of motion in a section in relation to the the previous frame, and the level of visual interest, such as contrast, or line, in a given section (I guess the rules can be adhered to by separating the image into a grid of smaller images and running the filter based on location within the frame) -remove pixels that have not changed colour, and remove pixels for a frame that have changed colour but are too visually similar for the human eye to distinguish. Suggestions? Matt S. wrote: Arent you essentially looking for the compression algorithms used by GIF/PNG/JPG images? For example, when you turn down the quality of ― or number of colors on ― a gif the like-colors increasingly tend to merge? .m On Tue, Nov 3, 2009 at 4:53 PM, Anthony Pace > wrote: After doing a little reading, I think, but I don't know for sure, that what I am referring to is called HVS(human visual system) compression. Any books, detailed articles, formulas or code? ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
Re: [Flashcoders] visually similar spectrum
Arent you essentially looking for the compression algorithms used by GIF/PNG/JPG images? For example, when you turn down the quality of — or number of colors on — a gif the like-colors increasingly tend to merge? .m On Tue, Nov 3, 2009 at 4:53 PM, Anthony Pace wrote: > After doing a little reading, I think, but I don't know for sure, that what > I am referring to is called HVS(human visual system) compression. > > Any books, detailed articles, formulas or code? > ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
Re: [Flashcoders] visually similar spectrum
After doing a little reading, I think, but I don't know for sure, that what I am referring to is called HVS(human visual system) compression. Any books, detailed articles, formulas or code? Henrik Andersson wrote: Anthony Pace wrote: I am not sure how to distinguish at what range or threshold a colour becomes too visibly similar to the eye to detect a difference, and I am really hoping there is a formula out there that I can be directed to. It depends on the context. If they are right next to each other and someone uses a magnifying glass, it is much more clear. But if they are two blobs that move and are on the opposite of a screen, it might be very hard. The eyes plays tricks on you. ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
Re: [Flashcoders] visually similar spectrum
> Does anyone have the math for at what range or threshold colours become > too visibly similar to separate by the eye? Is there a good book or > formula you know of? I think it's a little more complex--the human eye can distinguish colors in certain ranges much better than other ranges. For example, I think the human eye can detect smaller variations in earth tones than in, say, blues. Cordially, Kerry Thompson ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
Re: [Flashcoders] visually similar spectrum
Well kind of a mix really, but I guess you are right for part of it; yet, I would like to know the math and theory involved, so I can make educated adjustments. So I guess the steps will be: -to reduce the quality of only certain sections of the frame; for, I want to decrease quality based on the amount of motion in a section in relation to the the previous frame, and the level of visual interest, such as contrast, or line, in a given section (I guess the rules can be adhered to by separating the image into a grid of smaller images and running the filter based on location within the frame) -remove pixels that have not changed colour, and remove pixels for a frame that have changed colour but are too visually similar for the human eye to distinguish. Suggestions? Matt S. wrote: Arent you essentially looking for the compression algorithms used by GIF/PNG/JPG images? For example, when you turn down the quality of — or number of colors on — a gif the like-colors increasingly tend to merge? .m On Tue, Nov 3, 2009 at 4:53 PM, Anthony Pace wrote: After doing a little reading, I think, but I don't know for sure, that what I am referring to is called HVS(human visual system) compression. Any books, detailed articles, formulas or code? ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
Re: [Flashcoders] visually similar spectrum
Anthony Pace wrote: I am not sure how to distinguish at what range or threshold a colour becomes too visibly similar to the eye to detect a difference, and I am really hoping there is a formula out there that I can be directed to. It depends on the context. If they are right next to each other and someone uses a magnifying glass, it is much more clear. But if they are two blobs that move and are on the opposite of a screen, it might be very hard. The eyes plays tricks on you. ___ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
