Colour recovery from B&W kinescope films

[Aussie Bloke]

G'day all.

A few months ago I came across an awesome article on a program Richard Russell in the UK has developed to recover PAL colour from B&W 16mm kinescope film recordings of colour programs and it's success rate has proven very good on some of the kinescopes. The articles can be seen here:
http://www.techmind.org/colrec/
http://colour-recovery.wikispaces.co...olour+recovery

Makes me wonder on the possibility of someone in USA developing a program like this to recover NTSC colour from 1950s B&W kinescopes of colour programs, now that would be really cool!

[W.B.]

I was just about to ask the same question about B&W kinnies of NTSC material. But you beat me to it. Well, somebody had to bring it up . . .

Maybe one could coordinate with Livefeed or something . . .

[NewVista]

So all this is made possible because they goofed up in the mastering process: (the chroma artifacts should have been filtered out prior to display as they can shift the correct grayscale levels)

[Joel Cairo]

As I mentioned in another thread, due to the technical nature of how color is encoded in PAL vs. NTSC, as well as differences in how the respective telerecordings for each format are created, Richard Russell's astounding program will not work with NTSC material. In order to develop a similar process for NTSC, it would be necessary (at a minimum) to re-create the original color burst signal of the kinescoped program, which is not feasible at this point in time.

- Kevin

[wa2ise]

Originally Posted by Joel Cairo

As I mentioned in another thread, due to the technical nature of how color is encoded in PAL vs. NTSC, as well as differences in how the respective telerecordings for each format are created, Richard Russell's astounding program will not work with NTSC material. In order to develop a similar process for NTSC, it would be necessary (at a minimum) to re-create the original color burst signal of the kinescoped program, which is not feasible at this point in time.

- Kevin

If there's an area of the picture with a constant color (in a scene, like sky) one could infer a burst. It would have to be done by hand to adjust the tint at least for every scene. Adjust until you get the faces right. Labor intensive.

Another complication for NTSC is if the CRT display in the kinescope has some horizontal nonlinearity. Which would look like a gradual shift in tint if you demodulated the photographed chroma subcarrier. But one could model such nonlinearity in the decoder program if there's say a title screen of some constant color at the beginning or end of the show. PAL is more tolerant of such phase errors, as the phasing flips from line to line. SECAM material should be immune to this, SECAM may be the easiest to recover the color here. Oh, no burst to tell U from V, but just shift the line alternation of the decoder if the colors come out weird.

Of course this is dependent on someone running the kinescope capture process not bothering to filter out the chroma subcarrier. Probably using equipment built before color TV was installed elsewhere in the station. "Oh, we'll just low pass filter the film to video converter later.".

[Joel Cairo]

Originally Posted by wa2ise

If there's an area of the picture with a constant color (in a scene, like sky) one could infer a burst. It would have to be done by hand to adjust the tint at least for every scene. Adjust until you get the faces right. Labor intensive.

Another complication for NTSC is if the CRT display in the kinescope has some horizontal nonlinearity. Which would look like a gradual shift in tint if you demodulated the photographed chroma subcarrier. But one could model such nonlinearity in the decoder program if there's say a title screen of some constant color at the beginning or end of the show. PAL is more tolerant of such phase errors, as the phasing flips from line to line. SECAM material should be immune to this, SECAM may be the easiest to recover the color here. Oh, no burst to tell U from V, but just shift the line alternation of the decoder if the colors come out weird.

Of course this is dependent on someone running the kinescope capture process not bothering to filter out the chroma subcarrier. Probably using equipment built before color TV was installed elsewhere in the station. "Oh, we'll just low pass filter the film to video converter later.".

Well, the bigger issue with NTSC kinescopes of color programs is that by and large, they are not made the same way that UK telerecordings were: which is to say from a MONOCHROME monitor source that displayed (with sufficient resolution) the fine patterning of the accidentally-included color portion of the transmission signal. It is that patterning that allows Mr. Russell's process to recreate and decode the color portion of the PAL signal.

Unfortunately, here in the US, most B&W kinescopes of color programs were shot off of COLOR monitors that had shadow masks-- which, (along with NTSC's inherently lower resolution) effectively killed any patterning due to the resultant dot-crawl artifacts and softness of the image.

-Kevin

[ceebee23]

This really is a sort of geeky magic... amazing ... and you never know what results the system will bring!

BUT re: NTSC ... let's hope somewhere ..somehow someone finds the time and original materials that just happen to have been created in the right way and somehow an NTSC variant of the software can somehow be made to work!

Like the BBC material, this is actually part of our cultural and social history we need to retrieve what we can.

[Sandy G]

Well, fiddle-dee-dee....(grin) Too bad, it was a neat idea, though...

[newhallone]

Here is a link to the Dr.Who restoration team website with some great articles. It's amazing some of the things they can do now. And as time goes on more will be possible.

http://www.restoration-team.co.uk/

[NewVista]

Originally Posted by Aussie Bloke

http://www.techmind.org/colrec/
Makes me wonder ..a program ..to recover.. NTSC

Check out the diagonal lines on link foto; I'd forgotten how bad PAL was. With NTSC, dots don't align but are neatly 180 deg out of phase so TVs/ video effects can extract most luminance detail with a simple comb filter (can't do that with PAL). Also PAL decoding loses half chroma V resolution and every generation of co-dec displaces the chroma down one line! At least it's not as bad as SECAM which effects manufacturers wouldn't support.

[andy]

...

[NewVista]

Originally Posted by andy

The dots in PAL also line up 180 degrees out of phase on adjacent lines. .

It would have been good if they did, then PAL TV's could have implemented analog Y/C comb filters for sharper pictures since the 80's like the RCA Dimensia premium TV's since 1984

The NTSC developers foresaw this in the early 50s, but - 10 years later - the PAL developers screwed up - incorporating a 25Hz offset (to minimize dot crawl) resulting in an awkward 8-field 'Color Frame' sequence and really nasty artifacts like on those screen shots.

To comb filter a PAL TV, it requires expensive frame store digital processing chip-sets--not available until just before the phase out of PAL . Germany - where it started - ditched PAL for DTV a number of years ago.


Interesting link:
http://www.burnworld.com/dvd/primer/ntsc.htm

"Another important factor in choosing the new exact frame rate was to make sure that the color signal phase would be shifted exactly 180 degrees for each scanline. There are two reasons why this is important. First, the chroma signal does cause some distortion to older TV sets, especially those that were used at the time of the introduction of color TV and which didn't have notch filters to filter out the chroma information. In addition, early color tv sets (and newer cheap ones) suffer from imperfect luminance and chrominance separation, causing dots to appear near strong-colored edges. These dots are called creepy crawlies or, more commonly, dot crawl. They are particularly visible along vertical lines in the transmitted video, especially when SMPTE color bars are transmitted. The phase shift makes these dots non-stationary and thus reduces their visibility. The second reason to the phase shift is that it makes it possible to use a comb filter, which allows separating chrominance and luminance information with much better fidelity. While an exect 180 degree phase shift per scanline is not an absolute necessity for a comb filter to work, it makes implementation easier and also gives the best potential quality. This is a lesson that was later forgotten when developing the PAL color coding scheme. This probably didn't seem like a big omission at the time, since comb filters didn't become widely available in NTSC television sets before the 1980's (and, because of huge implementation difficulties, high-end PAL 100 Hz TV sets didn't get comb filters before the late 1990s). Nevertheless, the theoretical groundwork that made comb filters possible was there from the beginning"

[andy]

...

[old_tv_nut]

Slight correction about PAL chroma phasing:

The B-Y alternates from line to line due to the subcarrier frequency (plus has a slight offset due to the 25Hz frequency offset, which makes it slip phase slowly from the top to the bottom of the screen); the R-Y, due to "Phase Alternation by Line," does NOT alternate 180 degrees from line to line, and has only the slight diagonal offset due to the 25 Hz. This consistent phase should be very helpful in recovering the chroma from the subcarrier dots, I think.

The R-Y line-up can be analyzed in frequency domain also (perhaps making your brain hurt!) as modulation by a half-line-scan-rate (15625/2 Hz) square wave, which suppresses the original color subcarrier and sideband frequencies and creates double the number of sidebands, which are offset by +/- 7812.5 Hz. This can be shown (in either the frequency or time domain) to show that a simple one-line-delay comb filter will work for B-Y, but not for R-Y, since the R-Y dots line up just as though they were luminance stripes (except for that 25 Hz frame offset, of course).

You can draw some really pretty pictures of the 3-dimensional spectra (horizontal, vertical, and temporal frequency axes) of NTSC color and PAL color which show just what is necessary for comb filtering. but I'm not going to try that here. These also show all the flickering cross-color aliases for the two systems and show why some of the PAL crosscolor is more noticeable.

Edit: I forgot that the subcarrier frequency for PAL is also offset by a quarter cycle per line period to prevent the vertical line-up of the R-Y. Makes the cross-color not as bad as it could have been, but still a problem for comb filtering.

[dtuomi]

Actually most B&W kine's of color shows in the U.S. were made with B&W picture tubes. I had an engineer mentor of mine who used to get stuck babysitting one when he was a bad boy at ABC. Apparently they used blue phosphors in the tube because the film they used was most sensitive to that color (even though it was B&W).

Anyway, NTSC is real easy to notch filter out, and everyone did it to prevent the color info from causing any unwanted "random" patterns on the films. So almost no kinescopes of American color television would even have any color information in them at all.

David