🚀 go-pugleaf

In article <1993Jan8.194916.8558@black.clarku.edu> kbasye@black.clarku.edu (Ken Basye) writes:
>
>I've been wondering lately about how one might build a very simple and
>cheap color sensor.  I have in mind something that would be fairly
>reliable in providing 3 or perhaps even 4 bits of discrimination and
>which would be fairly reliable given good input, e.g., having a
>well-lit piece of colored construction paper put in front.
>
We are using some three color light emiting diodes as sensors. They appear to
work well not as sensitive as photo diodes but they will work at reasonable 
light levels. 

I think we got them form LEDTRONICS some place in Calif.
Good luck
Gordon

Gordon Couger
AB5Dg   Agriculture Engineering Oklahoma State University
gcouger@olesun.agen.okstate.edu 405-744-6514 day 744-2794 evenings   

In article <1993Jan8.194916.8558@black.clarku.edu>, kbasye@black.clarku.edu (Ken Basye) writes:
>
>I've been wondering lately about how one might build a very simple and
>cheap color sensor.  I have in mind something that would be fairly
>reliable in providing 3 or perhaps even 4 bits of discrimination and
>which would be fairly reliable given good input, e.g., having a
>well-lit piece of colored construction paper put in front.
>
>A photoresistor alone might seem to be enough, given that different
>pieces of paper will reflect different amounts of light.  So as a
>control, imagine that the sheets have been chosen so that a single
>photoresistor gets close to the same reading for each sheet, yet they
>are easily distinguished by color.
>
>My first thought was to use three photoresistors with R, G and B
>filters (just cellophane, perhaps).  Or perhaps there are tuned
>versions that would give the same effect.  Better yet, a single
>tunable photoresistor could be used with 3 consecutive readings.  
>Are there cheap ways to measure dominant frequency directly?  Etc...
>
>kbasye@black.clarku.edu     
>Ken Basye
>Dept. of Math and CS
>Clark University
>950 Main St.
>Worcester, MA  01610
>
Here's a couple ideas:

	I dug up a catalog that I get from Marlin Jones & Associates
(Phone #407-848-8236), they have a "Photo Electric Color Detector"
Part 3278-OP.  The catalog describes it as a surplus item with four
colored "filters" in front of solar cells.  Looks small. $5.00.  
(this place has a good selection of well-priced hobby elect. stuff).

	Also, (it'll be hard to get a reference on this :-), but, in
middle school, I did a science project (STOP laughing!) with a photocell
hooked up to a Commodore 64's 8-bit A/D converter.  I had a small stand
with the photocell and an light source that I would place different
colored pieces of paper under.  Significantly different amounts of light
would be detected by the CdS cell for different colors, even those with
apparently similar overall brightness.  Probably due to the non-linear
response of the CdS cell to different wavelengths and the odd spectrum
coming out of a light bulb.


>
 ----------------------------
 Ryan K. Brooks                Chip of the Month Club: TMS320C40
 2575BrooksR@vms.csd.mu.edu

In article <1993Jan8.194916.8558@black.clarku.edu>, kbasye@black.clarku.edu (Ken Basye) writes:
|> 
|> I've been wondering lately about how one might build a very simple and
|> cheap color sensor. 

|> My first thought was to use three photoresistors with R, G and B
|> filters (just cellophane, perhaps).  Or perhaps there are tuned
|> versions that would give the same effect.  Better yet, a single
|> tunable photoresistor could be used with 3 consecutive readings.  
|> Are there cheap ways to measure dominant frequency directly?  Etc...


Check out a company called Hamamatsu.  They make a sensor just like this.

It's a rectangular photoresistor about 3x8mm or so in size, with a red,
green, and blue field on it.  The sensor has two differential signals coming
out, and one we had around here seemed to work pretty well.  Don't know the
price, but I don't think it would be out of this world.

BTW,they make a bunch of other high quality sensors dealing mostly with
optoelectronics.

-Olaf

In article <1993Jan8.194916.8558@black.clarku.edu> kbasye@black.clarku.edu (Ken Basye) writes:

>I've been wondering lately about how one might build a very simple and
>cheap color sensor.

>My first thought was to use three photoresistors with R, G and B
>filters (just cellophane, perhaps).

The human RGB colour system is one of the worst among animal colour
vision systems. Most animals employ a 4 primary colour system, and one
that is more evenly spaced. The human system is (in evol terms)
recently developed and biassed towards fine discrimination of fruit
kind and status, and your robot is probably not a fruit eater!
Consider what your robot needs colour for, and give it what it needs.

-- 
Chris Malcolm    cam@uk.ac.ed.aifh          +44 (0)31 650 3085
Department of Artificial Intelligence,    Edinburgh University
5 Forrest Hill, Edinburgh, EH1 2QL, UK                DoD #205

What about a single detector (photocell/solarcell) and different colored LEDs
for illumination of the surface. Ie, what is the reflected brightness under
green/red/yellow/blue/IR light.

Jeff Bartlett