Searching \ for '[EE]: Video: Color to B/W conversion' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/displays.htm?key=video
Search entire site for: 'Video: Color to B/W conversion'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Video: Color to B/W conversion'
2000\11\29@003257 by Sean H. Breheny

face picon face
Hi all,

Quick video question: I want to try taking a color video signal and use a
high speed comparator to compare the brightness of pixels to some set value
(trying to detect a dark object against a light background). I want to
reject the color information in this process (luma only). How can I do
this? I know it must be fairly simple since B/W TV's can accept color signals.

Thanks,

Sean

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@004544 by Stuart O'Reilly

flavicon
face
Just run the video signal through a notch filter. The filter should be
set for 4.43 MHz for PAL or 3.58 MHz for NTSC.
Regards
Stuart

Sean H. Breheny wrote:
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@005847 by Stephen B Webb

flavicon
face
> Quick video question: I want to try taking a color video signal and use a
> high speed comparator to compare the brightness of pixels to some set value
> (trying to detect a dark object against a light background). I want to
> reject the color information in this process (luma only). How can I do
> this? I know it must be fairly simple since B/W TV's can accept color signals.

If you have access to an S-Video type input, you can split off the luma
signal and just let the color info spill onto the floor.

If you have to split a combined (composite) signal, you will tend to get
inferior results.

(they make cables which go from s-video to dual RCA type connectors, one
for chroma, one for luma)

-Steve

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@010505 by Sean H. Breheny

face picon face
Hi Stuart,

Thanks! Could I low pass filter it instead of using a notch filter? It
might be a bit simpler. How do old B/W TVs reject the color subcarrier?
Since they didn't know about it when they designed them, they couldn't
have  used a notch filter for 3.58 MHz (yes, I'm dealing with an NTSC signal).

Also, in response to Stephen, unfortunately, I will be using a cheap little
CCD board camera, so I don't think it will have S-video output.

Sean

At 04:47 PM 11/29/00 +1100, you wrote:
>Just run the video signal through a notch filter. The filter should be
>set for 4.43 MHz for PAL or 3.58 MHz for NTSC.
>Regards
>Stuart

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@015603 by David VanHorn

flavicon
face
At 12:33 AM 11/29/00 -0500, Sean H. Breheny wrote:
>Hi all,
>
>Quick video question: I want to try taking a color video signal and use a
>high speed comparator to compare the brightness of pixels to some set value
>(trying to detect a dark object against a light background). I want to
>reject the color information in this process (luma only). How can I do
>this? I know it must be fairly simple since B/W TV's can accept color
signals.

You're already there.
NTSC carries chroma on a subcarrier.
The baseband signal is only lumiance and sync.
B/W TVs don't pay any attention to the subcarrier, and neither should you.


--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@020416 by Sean H. Breheny
face picon face
Hi Dave,

Thanks,but I want to ask a further question because I don't totally
understand how that's true. If I'm looking at an NTSC color composite video
signal, isn't the subcarrier just added in to the signal? If so, why
wouldn't a comparator pick it up (i.e., if the B/W video level is -0.25,
and the subcarrier instantaneous voltage is 0.5 V, then why doesn't the
comparator act like the level is 0.25 V ?)

Thanks,

Sean

At 01:29 AM 11/29/00 -0500, you wrote:
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@022512 by Robert Rolf

picon face
"Sean H. Breheny" wrote:
> Thanks! Could I low pass filter it instead of using a notch filter? It

Yes, if you don't mind loosing some of your resolution.

> might be a bit simpler. How do old B/W TVs reject the color subcarrier?

Trap filter (not quite as deep as a notch but sufficient).

> Since they didn't know about it when they designed them, they couldn't
> have  used a notch filter for 3.58 MHz (yes, I'm dealing with an NTSC signal).

Depends on when/how it was made. The really cheap TV's don't bother
with a filter, and you'll see all kinds of dot crawl from strong
color information. (Color bars are really good for showing this).
Better ones do have this filter, and the really good ones use a
'comb' filter to cleanly remove the color info without affecting
the B&W bandwidth.

The NTSC signal was designed to be compatible with old B&W TV's so the
color phase changes every scan line so that you end up with a four field
sequence in NTSC. This visually averages out the color info dots since
every second field (odds and evens) has opposite chroma polarity.
3.579545Mhz /227.5= 15734.263 H rate /262.5 =59.94Hz=V rate
You will notice that the color subcarrier is 'interlaced' just like
the scan lines.

>
> Also, in response to Stephen, unfortunately, I will be using a cheap little
> CCD board camera, so I don't think it will have S-video output.

You could wiretap the luminance info since it is processed separately
from the chroma and usually combined at the output stage.
You may be able to just cut a trace to nuke the chroma addition function
so that you get B&W only output, with full bandwidth.

Chip info is available on the Sony and Sharp semi sites.
If you poke around the last transistor at the output you'll probably
find a SMT cap that is coupling the chroma signal into the base.
Remove it and you have a B&W camera.

Don't forget to gate your 'dark compartor' since sync is blacker
than black. And be sure to clamp your video so that scene brightness
changes don't throw off your comparator setpoints.

Robert

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@023141 by Nigel Goodwin

flavicon
face
In message <spam_OUT4.3.2.7.2.20001129003122.00c607d0TakeThisOuTspampostoffice2.mail.cornell.e> du>, Sean H. Breheny <.....shb7KILLspamspam@spam@CORNELL.EDU> writes
>Hi all,
>
>Quick video question: I want to try taking a color video signal and use a
>high speed comparator to compare the brightness of pixels to some set value
>(trying to detect a dark object against a light background). I want to
>reject the color information in this process (luma only). How can I do
>this? I know it must be fairly simple since B/W TV's can accept color signals.

The colour (color for you!) encoding systems were designed to be
backwards compatible with B/W TV's - the colour is encoded on a
4.44361875Mz sub-carrier (for PAL) and 3.58?Mz for NTSC. An old B/W set
will simply not display anything from this sub-carrier, later sets also
include a notch filter to remove the chroma from the video path, as do
colour sets in their luminance processing. All you need is a simple
notch filter, or it may just work without doing anything!.
--

Nigel.

       /--------------------------------------------------------------\
       | Nigel Goodwin   | Internet : nigelgspamKILLspamlpilsley.co.uk           |
       | Lower Pilsley   | Web Page : http://www.lpilsley.co.uk       |
       | Chesterfield    | Official site for Shin Ki and New Spirit   |
       | England         |                 Ju Jitsu                   |
       \--------------------------------------------------------------/

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@030200 by Sean H. Breheny

face picon face
Hi Robert,

Thanks for the info. I am going to gate the comparator by simply having my
code (PIC, of course :-) just reject the input during sync periods (I will
have an LM1881 sync separator in the circuit to tell the PIC when the sync
is happening). I'm not sure what you mean by "clamp", though. If you mean
just limit the video signal swing, I don't understand how that would
achieve what you are talking about.

Sean

At 12:04 AM 11/29/00 -0700, you wrote:
>Don't forget to gate your 'dark compartor' since sync is blacker
>than black. And be sure to clamp your video so that scene brightness
>changes don't throw off your comparator setpoints.
>
>Robert
>
>--
>http://www.piclist.com hint: The list server can filter out subtopics
>(like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@033808 by David VanHorn

flavicon
face
At 02:02 AM 11/29/00 -0500, you wrote:
>Hi Dave,
>
>Thanks,but I want to ask a further question because I don't totally
>understand how that's true. If I'm looking at an NTSC color composite video
>signal, isn't the subcarrier just added in to the signal? If so, why
>wouldn't a comparator pick it up (i.e., if the B/W video level is -0.25,
>and the subcarrier instantaneous voltage is 0.5 V, then why doesn't the
>comparator act like the level is 0.25 V ?)

The subcarrier is phase modulated, and low amplitude.
If this were a problem, you'd see it in a B/W tv, as distortion of the
picture.

Also, the speed of your comparator is critical here.
You could lowpass your video, but this will get you phase distortion.


--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@040341 by Michael Rigby-Jones

flavicon
face
       At 12:04 AM 11/29/00 -0700, you wrote:
       >Don't forget to gate your 'dark compartor' since sync is blacker
       >than black. And be sure to clamp your video so that scene
brightness
       >changes don't throw off your comparator setpoints.
       >
       >Robert


> {Original Message removed}

2000\11\29@072308 by Bob Ammerman

picon face
If you don't need to detect 'darkness' with 'pixel-level' resolution, you
can indeed use a low-pass filter.

Since each scan line is 1/15750 of a second, a cutoff around 15750*100 =
1.575 MHz should get you pretty good resolution.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@072519 by Bob Ammerman

picon face
> Hi Robert,
>
> Thanks for the info. I am going to gate the comparator by simply having my
> code (PIC, of course :-) just reject the input during sync periods (I will
> have an LM1881 sync separator in the circuit to tell the PIC when the sync
> is happening). I'm not sure what you mean by "clamp", though. If you mean
> just limit the video signal swing, I don't understand how that would
> achieve what you are talking about.
>

Video is normally AC coupled.  Clamping, also called DC restoration,
rebiases the signal so that the sync tips are at a known voltage.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@090334 by Olin Lathrop

face picon face
> Quick video question: I want to try taking a color video signal and use a
> high speed comparator to compare the brightness of pixels to some set
value
> (trying to detect a dark object against a light background). I want to
> reject the color information in this process (luma only). How can I do
> this? I know it must be fairly simple since B/W TV's can accept color
signals.

This is very easy if it is a composite signal.  The magnitude of that signal
is the luminance channel, which is the black and white part.  Color, if
present, is encoded such that an old black and white TV can just use the
luminance and without any knowledge of color.

Color requires two additional degrees of freedom.  These are encoded as the
amplitude and phase shift of a carrier whose frequency is just a bit above
the official upper limit of the luminance signal.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, .....olinKILLspamspam.....embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@090533 by Olin Lathrop

face picon face
> >Just run the video signal through a notch filter. The filter should be
> >set for 4.43 MHz for PAL or 3.58 MHz for NTSC.
>
> Thanks! Could I low pass filter it instead of using a notch filter?

Yes, the luminance signal is all below the color carrier.  But obviously the
slower the filter cutoff, the more luminance signal you will be throwing
out.  This will look like horizontal blurring.

> It
> might be a bit simpler. How do old B/W TVs reject the color subcarrier?

They don't.  The beam intensity does get modulated by the color subcarrier,
but so what?  There is no way you can see this accross the room, and
probably not even by carefully examining the picture.  The color subcarrier
frequency is deliberately just a bit higher than the luminance upper
frequency limit.  If you had special high performance B+W TV, the color
subcarrier would show up as very tightly spaced horizontal ripples.  Most
old black and white TVs can't resolve this, and your eyes can't either from
a normal viewing distance.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, EraseMEolinspam_OUTspamTakeThisOuTembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@093210 by Olin Lathrop

face picon face
> Thanks,but I want to ask a further question because I don't totally
> understand how that's true. If I'm looking at an NTSC color composite
video
> signal, isn't the subcarrier just added in to the signal?

Yes.

> If so, why
> wouldn't a comparator pick it up (i.e., if the B/W video level is -0.25,
> and the subcarrier instantaneous voltage is 0.5 V, then why doesn't the
> comparator act like the level is 0.25 V ?)

It would if the comparator is fast enough.  The color subcarrier was meant
to be unnoticeable to humans watching old B+W TVs at a normal viewing
distance.  It is not necessarily unnoticable to a custom circuit that
analyses the video signal.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, olinspamspam_OUTembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@111401 by Robert Rolf

picon face
Most Video circuits are AC coupled (especially cheap cameras).
This means that level you've
chosen to detect (say .8V) will move as the AVERAGE scene brightness
changes. Clamping uses the sync signal to force the known black
part of the video (just after sync) to a fixed level. It can
also be as simple as a diode that gets biased on for sync tips since the
relationship between sync (0V) and video is supposed to be constant.

The LM1881 has a clamp output (back porch) which you'd use to
turn on a transistor switch for to force your
downstream (past the AC coupling cap) video to black level (0v
since you want to keep your dynamic range as wide as possible for
noise considerations).

If you look a changing video scene with a scope you will see that
the trace moves around as the average scene brightness changes.
This will play havoc will your comparator setpoint unless you clamp
the video (having striped it of the chroma subcarrier before you
do this).

Robert


"Sean H. Breheny" wrote:
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@111801 by Robert Rolf

picon face
David VanHorn wrote:
>
> At 02:02 AM 11/29/00 -0500, you wrote:
> >Hi Dave,
> >
> >Thanks,but I want to ask a further question because I don't totally
> >understand how that's true. If I'm looking at an NTSC color composite video
> >signal, isn't the subcarrier just added in to the signal? If so, why
> >wouldn't a comparator pick it up (i.e., if the B/W video level is -0.25,
> >and the subcarrier instantaneous voltage is 0.5 V, then why doesn't the
> >comparator act like the level is 0.25 V ?)
>
> The subcarrier is phase modulated, and low amplitude.

It's not low amplitude if he's shooting against a strong colored
background. The subcarrier can be a large as the video signal
for 'saturated' color backgrounds.


> If this were a problem, you'd see it in a B/W tv, as distortion of the
> picture.

NO! What you see is 'dot crawl'. Ever look at color bars on a B&W
TV? Check out the boundaries between the bars to see what I mean.(TV
dependant of course. The cheaper the TV the more likely you'll
see the crawling).

> Also, the speed of your comparator is critical here.

Yes, it should be fast, but that also depends on what he's trying
to detect. A flying insect or an intruder.

> You could lowpass your video, but this will get you phase distortion.

Not enough to be an issue here.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@114648 by Harold M Hallikainen

picon face
       The color subcarrier frequency is chosen such that the black and white
dots produced by the color subcarrier appear as white and black dots in
the next frame. Your eye then averages these out removing the effects of
the color subcarrier on b/w television receivers.

Harold


On Wed, 29 Nov 2000 01:05:36 -0500 "Sean H. Breheny" <@spam@shb7KILLspamspamCORNELL.EDU>
writes:
{Quote hidden}

FCC Rules Online at http://hallikainen.com/FccRules
Lighting control for theatre and television at http://www.dovesystems.com

________________________________________________________________
GET INTERNET ACCESS FROM JUNO!
Juno offers FREE or PREMIUM Internet access for less!
Join Juno today!  For your FREE software, visit:
dl.http://www.juno.com/get/tagj.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@131809 by Dan Larson

flavicon
face
On Wed, 29 Nov 2000 00:04:57 -0700, Robert Rolf wrote:

>
>Depends on when/how it was made. The really cheap TV's don't bother
>with a filter, and you'll see all kinds of dot crawl from strong
>color information. (Color bars are really good for showing this).
>Better ones do have this filter, and the really good ones use a
>'comb' filter to cleanly remove the color info without affecting
>the B&W bandwidth.

Many *good* color TVs had comb filters as well. They got a better picture
by separating the color and luminance early on using the comb filter to remove
the color signal from the composite and then subtracting the result from the
composite again to get the chroma. Much of the good color data can be derived
from the sideband signals of the color subcarrier that are sort of interleaved
throughout the spectrum with the luminance and are lost  with a simple
notch, or trap filter. This would be the higher detail color data that spills over
like that.

Dan

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@172121 by Matthew Fries

flavicon
face
Speaking of Comb filers... I was looking at televisions at Best Buy and
one of them boasted a "Digital Comb Filter". How can this be? Isn't the
comb filter deep in analog territory? What can be digital about it?

These were NTSC televisions, not the HDTV variety either.



On Wed, 29 Nov 2000, Dan Larson wrote:

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@181352 by Nigel Goodwin

flavicon
face
In message <KILLspamPine.GSO.4.21.0011291618400.4374-100000KILLspamspamisis.visi.com>,> Matthew Fries <RemoveMEfreezeTakeThisOuTspamVISI.COM> writes
>Speaking of Comb filers... I was looking at televisions at Best Buy and
>one of them boasted a "Digital Comb Filter". How can this be? Isn't the
>comb filter deep in analog territory? What can be digital about it?
>
>These were NTSC televisions, not the HDTV variety either.

It's quite simple, all the processing is done digitally inside chips,
they just implement a comb filter algorithm in the software - probably
using a DSP.
--

Nigel.

       /--------------------------------------------------------------\
       | Nigel Goodwin   | Internet : spamBeGonenigelgspamBeGonespamlpilsley.co.uk           |
       | Lower Pilsley   | Web Page : http://www.lpilsley.co.uk       |
       | Chesterfield    | Official site for Shin Ki and New Spirit   |
       | England         |                 Ju Jitsu                   |
       \--------------------------------------------------------------/

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@185504 by Bob Ammerman

picon face
----- Original Message -----
From: Nigel Goodwin <TakeThisOuTnigelgEraseMEspamspam_OUTLPILSLEY.CO.UK>
To: <RemoveMEPICLISTspamTakeThisOuTMITVMA.MIT.EDU>
Sent: Wednesday, November 29, 2000 6:10 PM
Subject: Re: [EE]: Video: Color to B/W conversion


{Quote hidden}

At video frequencies!! ?? !!

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\29@235827 by Robert Rolf

picon face
Bob Ammerman wrote:
> From: Nigel Goodwin <RemoveMEnigelgEraseMEspamEraseMELPILSLEY.CO.UK>
> To: <RemoveMEPICLISTspam_OUTspamKILLspamMITVMA.MIT.EDU>
> Sent: Wednesday, November 29, 2000 6:10 PM
> Subject: Re: [EE]: Video: Color to B/W conversion
> > Matthew Fries <RemoveMEfreezeTakeThisOuTspamspamVISI.COM> writes
> > >Speaking of Comb filers... I was looking at televisions at Best Buy and
> > >one of them boasted a "Digital Comb Filter". How can this be? Isn't the
> > >comb filter deep in analog territory? What can be digital about it?
> > >
> > >These were NTSC televisions, not the HDTV variety either.
> >
> > It's quite simple, all the processing is done digitally inside chips,
> > they just implement a comb filter algorithm in the software - probably
> > using a DSP.
>
> At video frequencies!! ?? !!

Yep. It's a trivially simple algorithm. Sample line at a minimum of
4*Fsc, delay one H, subtract new samples from delayed line samples and
output. The early comb filters used an ultrasonic subcarrier (10.7Mhz
was common) into a glass slab that bounced the signal back and forth
numerous times to get the delay they needed. That what those funny
tall (2cm), thin (.5cm) brightly color slabs were for.

Robert

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\11\30@004035 by Sean H. Breheny

face picon face
Except that what you mention, while it may be discrete-time signal
processing, doesn't qualify as DSP, and it especially doesn't qualify as
"using A DSP" because that implies an actual DSP processor.

Sean

At 09:58 PM 11/29/00 -0700, you wrote:

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2000\11\30@014440 by Nigel Goodwin

flavicon
face
In message <006001c05a5f$10b58b60$EraseME08e9c540spamspamspamBeGonesciencekit.com>, Bob Ammerman
<RemoveMERAMMERMANKILLspamspamPRODIGY.NET> writes
>{Original Message removed}

2000\11\30@054036 by mike

flavicon
face
On Wed, 29 Nov 2000 18:49:29 -0500, you wrote:

>----- Original Message -----
>From: Nigel Goodwin <nigelgSTOPspamspamspam_OUTLPILSLEY.CO.UK>
>To: <spamBeGonePICLISTSTOPspamspamEraseMEMITVMA.MIT.EDU>
>Sent: Wednesday, November 29, 2000 6:10 PM
>Subject: Re: [EE]: Video: Color to B/W conversion
>
>
>> In message <KILLspamPine.GSO.4.21.0011291618400.4374-100000spamBeGonespamisis.visi.com>,> >> Matthew Fries <EraseMEfreezespamEraseMEVISI.COM> writes
>> >Speaking of Comb filers... I was looking at televisions at Best Buy and
>> >one of them boasted a "Digital Comb Filter". How can this be? Isn't the
>> >comb filter deep in analog territory? What can be digital about it?
>> >
>> >These were NTSC televisions, not the HDTV variety either.
>>
>> It's quite simple, all the processing is done digitally inside chips,
>> they just implement a comb filter algorithm in the software - probably
>> using a DSP.
>
>At video frequencies!! ?? !!
Yes - take a look at any PC video capture card - you won't see any big
analogue delay lines or filters - all done digitally. Check out the
Brooktree website for data if you want to read more.
--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2000\11\30@074656 by Bob Ammerman

picon face
> Bob Ammerman wrote:
> > From: Nigel Goodwin <@spam@nigelg@spam@spamspam_OUTLPILSLEY.CO.UK>
> > To: <spamBeGonePICLISTspamKILLspamMITVMA.MIT.EDU>
> > Sent: Wednesday, November 29, 2000 6:10 PM
> > Subject: Re: [EE]: Video: Color to B/W conversion
> > > Matthew Fries <.....freezespam_OUTspamVISI.COM> writes
> > > >Speaking of Comb filers... I was looking at televisions at Best Buy
and
> > > >one of them boasted a "Digital Comb Filter". How can this be? Isn't
the
{Quote hidden}

Ah... the light goes on. I was envisioning a typical FIR/IIR type digital
filtering with multiple poles...

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2000\11\30@095518 by Olin Lathrop

face picon face
> Speaking of Comb filers... I was looking at televisions at Best Buy and
> one of them boasted a "Digital Comb Filter". How can this be? Isn't the
> comb filter deep in analog territory? What can be digital about it?

Think of the comb filter as a mathematical formula.  It can be realized
(approximated) by either analog or digital means.  In this particular case
the bandwidth is fairly high so a digital implementation is not trivial.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, TakeThisOuTolin.....spamTakeThisOuTembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.



'[EE]: Video: Color to B/W conversion'
2000\12\01@042026 by Peter L. Peres
picon face
>digital comb filter

Basically it's a video speed ADC, asic or dsp implementing the filter,
followed by a DAC (or three - usually all the processing is done in
digital after that, followed by a RGB treble DAC driving the CRT guns
almost directly).

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email TakeThisOuTlistservKILLspamspamspammitvma.mit.edu with SET PICList DIGEST in the body


2000\12\01@042032 by Peter L. Peres

picon face
Sean, small BW TV's don't do anything about rejecting color, but they do
reject it because they usually do not have the bandwidth to display it,
either in the IF or in the video amplifier sections. Professional monitors
do have rejection circuitry and the bandwidth and when the former does not
work then the color signal manifests itself in a very peculiar disturbance
pattern on the monitor (you can tell immediately what is on).

The reverse is also true, high resolution color images rendered on a
monitor while using composite video signal (CVBS) can have image details
that are b/w appear in color on a monitor because their features generate
color carrier frequencies in the Y channel.

So the way to get rid of color is to use a simple tank circuit (LCR) tuned
to 3.57 MHz (in America), with a 6dB bandwidth of about 350 kHz (which is
about 10% of Fo). However due to the items detailed above, you also need
to limit the bandwidth of the video signal before comparing, to lower than
3.57-(0.35/2) MHz.

The best way is, to modify the source of video signal so it does not
output color or burst at all. Then you have all the bandwidth and no
filtering problems. This is trivial in most cases (find the YC mixer in
the camera or VCR and remove the color signal coupling capacitor - this
works in most cases). TV tuners can be tuned 'low'. This puts the color
signal on the high attenuation upper slope of the SAW filter and will
remove color (and possibly sound in single IF TV tuner/if units). Use the
AFC control to do this on an assembled set.

good luck,

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservspamRemoveMEmitvma.mit.edu with SET PICList DIGEST in the body


2000\12\01@042107 by Peter L. Peres

picon face
>what color carrier does to comparator

Under certain conditions the comparator will output a pulse train of the
frequency of the color carrier if you do not remove it.

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservspamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body


2000\12\01@165646 by Robert Rolf

picon face
"Peter L. Peres" wrote:

GUESSED? Speculated? Mispoke?

> Sean, small BW TV's don't do anything about rejecting color, but they do
> reject it because they usually do not have the bandwidth to display it,
> either in the IF or in the video amplifier sections.

If this were true Peter, you wouldn't have any sound. For most
TV standards the sound carrier is ABOVE the chroma carrier so any
bandwidth constraint in the IF  that killed the color carrier would also
kill the sound carrier. For NTSC sound is at 4.5Mhz, color at 3.58Mhz.

>  Professional monitors
> do have rejection circuitry and the bandwidth and when the former does not
> work then the color signal manifests itself in a very peculiar disturbance
> pattern on the monitor (you can tell immediately what is on).

That pattern (herringbone or 'stipe noise') occurs when the raw
luma video signal (vertical stripes in a scene) produces frequencies
that match the color subcarrier. Modern cameras trap this out so
that you rarely see this effect.

> The reverse is also true, high resolution color images rendered on a
> monitor while using composite video signal (CVBS) can have image details
> that are b/w appear in color on a monitor because their features generate
> color carrier frequencies in the Y channel.

A trick that the old Apple II computer expoited to simplify it's
video circuitry.

> So the way to get rid of color is to use a simple tank circuit (LCR) tuned
> to 3.57 MHz (in America), with a 6dB bandwidth of about 350 kHz (which is
> about 10% of Fo). However due to the items detailed above, you also need
> to limit the bandwidth of the video signal before comparing, to lower than
> 3.57-(0.35/2) MHz.
>
> The best way is, to modify the source of video signal so it does not
> output color or burst at all. Then you have all the bandwidth and no

Removing color "burst" will simply make a color TV display a B&W image.
It will do NOTHING to prevent the color subcarrier from being included
in the video signal.

> filtering problems. This is trivial in most cases (find the YC mixer in
> the camera or VCR and remove the color signal coupling capacitor - this

But this is removing the chroma signal, not necessarily the burst, and
you may be surprised to find that you can get great 'color black' signal
(composite sync with burst) by breaking the one path, without actually
affecting the other (I know, it happened to me in an old Sony camera).

> works in most cases). TV tuners can be tuned 'low'. This puts the color
> signal on the high attenuation upper slope of the SAW filter and will
> remove color (and possibly sound in single IF TV tuner/if units). Use the
> AFC control to do this on an assembled set.

The simplest solution is an LC low pass filter, if resolution loss
is tolerable. Otherwise a  parallel LC trap tuned to the color
subcarrier.

Robert

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spamBeGonelistserv@spam@spamspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body


2000\12\03@124844 by Sean H. Breheny

face picon face
Thanks to all who responded about the video question. As it turns out, I
can tolerate a reduction in resolution (in fact, it will actually help me
by preventing a single dark pixel from being detected as a whole dark
object), so I plan on using a low pass filter with a cutoff somewhere
around 1 MHz (I have to figure out exactly how much resolution reduction I
can tolerate).

I tried a very quick test of the idea a few days ago, just using a fast
comparator (without sync gating or DC restoration, as I said, just a very
crude test) and it looks like this will probably work. I have to wait a
couple of weeks or so until I get a chance to actually try a better experiment.

As for DC restoration, I'm afraid I don't understand what is being
suggested. My thought was to capacitively couple the input, sample the
signal level when the LM1881 says that the negative sync tip should
occur(or just use a negative peak detector), and then use an op-amp to set
the average level such that the negative tip ends up where I want it (the
inverting input would be connected to the sampled voltage, the
non-inverting input to a voltage reference, and the output would go to the
signal line via a resistor. A small capacitor would go from output to
inverting input to allow constant feedback for high frequencies).

When I see the circuit for the Dirt Cheap Frame Grabber (which Mike
suggested), however, my circuit seems WAY too complicated. I don't
understand how that guy's circuit (just the capacitor, 2 resistors, a
diode, and a 2N2222) does what he says,or what is necessary. He talks about
the capacitor clamping the voltage. I don't see any way that the cap is
going to affect the average level (which it must because the average level
is changing and we don't want it to).

Thanks again,

Sean

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2000\12\03@182821 by Olin Lathrop

face picon face
> As for DC restoration, I'm afraid I don't understand what is being
> suggested. My thought was to capacitively couple the input, sample the
> signal level when the LM1881 says that the negative sync tip should
> occur(or just use a negative peak detector),

You shouldn't use the sync tip for DC restoration.  It can work if you're
always receiving video from the same known device, but I would definitely
not use it accross video sources.  The scan line front porch is provided for
the purpose of DC restoration.  It occurs after the sync tip and before the
displayable part of the scan line.  This is also where the color burst is,
so you have to be a little careful.  You can also use the back porch, but
usually finding the front porch is a little easier with simple circuitry.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, TakeThisOuTolinspamspamembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2000\12\04@022454 by Roman Black

flavicon
face
Sean H. Breheny wrote:
{Quote hidden}

Sean, see if you can find a TV trades textbook, an older
one will be better. These simple circuits have been used
for many years for finding the sync pulses, dc restoration,
black/white level detection etc etc. Any good TV textbook
will have chapters devoted to this with simple circuits,
formulas and examples. If you are going to be doing a
lot of work with video signal this may be worth getting.
-Roman

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2000\12\07@012404 by Harold Hallikainen

picon face
On Sun, 3 Dec 2000 18:17:16 -0500 Olin Lathrop
<olin_piclistEraseMEspamEMBEDINC.COM> writes:
{Quote hidden}

       By the way, waveforms for NTSC television are shown in section 73.699 of
the FCC rules. This is available as text (without the images) at
www.hallikainen.com/cgi-bin/section.pl?section=73.699
and as a pdf (with images) at
frwebgate.access.gpo.gov/cgi-bin/get-cfr.cgi?TITLE=47&PART=73&SECT
ION=699&TYPE=PDF

Watch for URL wrap. Also, on the pdf, you may have to save the file, then
rename it with pdf extension to view it.

Harold


FCC Rules Online at http://www.hallikainen.com/FccRules/

________________________________________________________________
GET INTERNET ACCESS FROM JUNO!
Juno offers FREE or PREMIUM Internet access for less!
Join Juno today!  For your FREE software, visit:
dl.http://www.juno.com/get/tagj.

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu


More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...