Calculations for brightness

This forum is for posts that might be considered off-topic but that may be useful or interesting to members. Examples include posts about electronics or programming in general, other microcontrollers or interesting devices, useful websites, etc.
GTBecker
Posts: 616
Joined: 17 January 2006, 19:59 PM
Location: Cape Coral

Post by GTBecker »

spamiam wrote:... there are many more possible spectral combinations than color sensations.."
Yes, there are more color combinations in analog nature than anyone or anything can discern. That doesn't make those colors invisible, just apparently the same as a neighbor until a better eye or instrument separates them.
spamiam wrote:In a region where there is no overlap, then a spike is either going to be missed...
Do I misunderstand you, Tony? What regions of no-overlap of cone spectrum sensitivity exist in the normal eye - except at the short and long extremes? Where will these unobserved spikes be seen?
Tom
GTBecker
Posts: 616
Joined: 17 January 2006, 19:59 PM
Location: Cape Coral

Post by GTBecker »

DocJC wrote:... Note also that [humans] are all unique.
Sure, and the Wikipedia link that Tony provided also stresses the non-absolute nature of color perception. Still, it is clearly possible today to accurately measure and, therefore, reproduce colors; the days of the golden-eye colorist have perhaps passed. Today, we can calibrate our cameras and monitors pretty well with a $100 tool, and television can automatically balance flesh tones. These functions do not consider individual differences in perception.

Who is to say what that calibrated standard image looks like to different people? Whatever it is, I'm sure it is standard and "normal", for them. Like many, I am color-deficient; I need to be careful with resistors - the multimeter is right here - and I could never have applied to Bell Telephone as a lineman. Slate-Green-Brown-Red-Orange would have been a wrong number if I made the splice. Still, I believe I am capable of seeing most basic colors correctly without a reference, and certainly I can see the _difference_ between two "similar" colors, if I have a large enough sample.

To my eye, a red LED - essentially a point source from a distance - is probably indistinguishable from a green one (and, certainly, an orange one) at distance. If I, however, flood my eye with the LED illumination - close enough to the LED lens to see a disk of solid color, I can easily tell you if the LED is red or green, but I'd still have trouble describing one as orange - unless there is a red sample next to it. Then, I could hang a color name on each. The same effect results from removing my (near-sighted) glasses to blur the point source. A Christmas tree is more colorful if it is out-of-focus, but just points of less-colorful light when sharp; they are better when the light is in large overlapping blurred disks with astigmatic details. And fireworks are great!

My experience is that color-blindness isn't that at all; it is, instead, a color name assignment difficulty, perhaps due to an insufficient sample from compromised sensors. There is plenty of color, poorly identified.

[After some more thought:] I recognize that if there exists some color that I absolutely cannot perceive, I suppose I have no way to know that except by other's observation or instrumentation. So far, I can't recall anyone pointing to something that I could not also see because of its color.
Last edited by GTBecker on 26 July 2007, 12:51 PM, edited 1 time in total.
Tom
dkinzer
Site Admin
Posts: 3120
Joined: 03 September 2005, 13:53 PM
Location: Portland, OR

Post by dkinzer »

GTBecker wrote:My experience is that color-blindness isn't that at all; it is, instead, a color name assignment difficulty, perhaps due to an insufficient sample from compromised sensors.
That may be true in some cases but the Ishihara test would seem to identify color perception deficiency rather than color naming difficulty.

I have problems with color perception similar to what Tom described. For example I may not properly identify the color of an item that is red or brown but I can distinguish between the same two colors when they are side by side as long as the color difference is large enough.
- Don Kinzer
spamiam
Posts: 739
Joined: 13 November 2005, 6:39 AM

Post by spamiam »

GTBecker wrote:What regions of no-overlap of cone spectrum sensitivity exist in the normal eye - except at the short and long extremes? Where will these unobserved spikes be seen?
Well, the actual spectral response is not a continuous curve for each of the receptor pigments. There are areas of relatively lower response than immediately adjacent wavelengths.

In some narrow bands the response is very low. Close to zero.

Since the red and green receptors have a lot of overlap, there is little chance that a narrow spike at a certain wavelength will go totally undetected. It probably would only be misinterpreted.

But at the blue end of the spectrum, there is little overlap of the red-green pigments with the blue pigments. While I do not know of a SPECIFIC wavelength that would be unobserved, despite being within the standard visible band, I do know they exist. These are rather narrow wavelength regions, but some specific wavelengths are effectively invisible to "normal" eyes despite being within the standard visual range.

But this is a very artificial situation. In real life, most sources of light put out a band of wavelengths, even for a single "color". Therefore the light source is going to be perceived.

In the lab where it is possible to tune a device to emit a single wavelength, it is possible to find one that provokes little or no visual response while adjacent wavelengths are visible.

Why is this important? Well, it is probably not important. But LEDs emit a very narrow spectrum and laser diodes have an extremely narrow spectrum. If this happens to land in someone's dead zone, then it might not be visible.
GTBecker
Posts: 616
Joined: 17 January 2006, 19:59 PM
Location: Cape Coral

Post by GTBecker »

dkinzer wrote:Ishihara test
You're right, Don, eating my words. I remember many of those tests, and I see 21 here, the abnormal result. It would be interesting to see if the colors used in the test are discernable when larger or adjacent.

My wife will also remind me that I miss some flashes of color, like from a Jacaranda tree in bloom, until it is closer than when she gasped. Conversely, I can read street signage much farther away than she. Her world is a little more blurred, apparently, and more colorful.
Tom
GTBecker
Posts: 616
Joined: 17 January 2006, 19:59 PM
Location: Cape Coral

Post by GTBecker »

spamiam wrote:... it is possible to find one that provokes little or no visual response while adjacent wavelengths are visible. [] it is probably not important.
Unimportant? Are you kidding? Tony, if there exists an invisible frequency within the visible spectrum, it will be exploited for marketing. That's a new form of broadcasting! If that's real, pursue it!

If such frequencies exist, you will have found a new eye-safe laser industry, too. Currently, 1560nm is the shortest eye-safe frequency, I believe.
Tom
spamiam
Posts: 739
Joined: 13 November 2005, 6:39 AM

Post by spamiam »

GTBecker wrote:If such frequencies exist, you will have found a new eye-safe laser industry, too. Currently, 1560nm is the shortest eye-safe frequency, I believe.
Well, I would not think it is safe even then. Behind the retina is a layer that absorbs the light so it does not backscatter.

A well collimated beam of invisible light will still deliver enough energy to do damage as long as SOMETHING in the eye absorbs the energy.

-Tony.
Post Reply