h a l f b a k e r yNot just a think tank. An entire army of think.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Every camera-based motion detection system I've used or
are aware of detects motion using a simple algorithm of
counting the changed pixels. And every one that I know of
suffers the same problem, which is false positives caused
when the sun moves in and out of clouds. Although there
are a
few tweaks applied to minimise this, it's still a big
issue. What I've noticed is that the pixel comparison is a
trivial one based on either RGB pixel values or intensity. A
better algorithm would be to bias the value of the change
per pixel by hue more than by intensity. So a person or
animal walking over a surface changes the colour (hue) but
a change in sunlight only changes the intensity. This would
be a very minor tweak to a program such as 'motion' or
'zoneminder' but would greatly reduce the false positive
rate.
[link]
|
| |