[mythtv] New idea for commercial flagging...

Nigel Pearson nigel at ind.tansu.com.au
Tue Jan 4 01:38:18 EST 2005


	I forgot to say at the time - I think this is a great idea!

	The same principle could maybe be applied to station watermark
detection - do hashing on the luminance channel in the bottom
right hand corner of the frame (or wherever the station puts it).


> Any thoughts on a sensible hashing function?  Might be almost a 
> perfect identical frame for DVB users, but for analogue users there is 
> going to be sufficient "noise" to make the detection of identical 
> frames tricky.  My first thought would be to break the frame up into, 
> say, 8x8 pixel blocks, average the colour and then quantise this 
> colour to a smaller range (say 1-8 instead of 1-256 for each colour).

	That would have an advantage of working well for both analogue
and digital streams (and compressed streams conveniently already
divide the picture up into pixel blocks).

>   Does this sound "fuzzy" enough to spot identical frames?

	Yep.


	Here are a few more simple algorithm ideas:

* Analogue: average about 8 bytes horizontally in the middle
   of each even scanline (hopefully minimise effect of noise/ghosting),
   and hash that across each "underscan safe" scanline in the field

* Digital: just add every 20th byte (or every 100th, or every 1000th)
   into an 8bit variable.


> Anyone read any research papers on doing this that might have some 
> better ideas..?  Anyone got a couple of hours free to try coding this 
> up...?

	Free time? Good joke!

--
Nigel Pearson, nigel at ind.tansu.com.au|  "Let's see how Spike is going"
Telstra BI&D, Sydney, Australia      |              ...
Office: 8255 4222    Fax:  8255 3153 |"I'd like to keep Spike as my 
pet."
Mobile: 0408 664435  Home: 9792 6998 |        Illyria - Angel



More information about the mythtv-dev mailing list