<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 13/02/2015 3:38 AM, Greg Troxel
wrote:<br>
</div>
<blockquote cite="mid:rmiiof7kruc.fsf@fnord.ir.bbn.com" type="cite">
<pre wrap="">
Peter Wendorff <a class="moz-txt-link-rfc2396E" href="mailto:wendorff@uni-paderborn.de"><wendorff@uni-paderborn.de></a> writes:
</pre>
<blockquote type="cite">
<pre wrap="">Am 18.01.2015 um 07:14 schrieb John F. Eldredge:
</pre>
<blockquote type="cite">
<pre wrap="">You could use a light meter to measure how bright the light is. That
isn't the only factor in the suitability of the lighting, but it is
objective.
</pre>
</blockquote>
<pre wrap="">... provided that you measure on a dark night without moon and stars,
without cars driving on the road and no house next to it enlightened?
Even then:
Where do you measure the brightness? directly below the street lamp or
in the middle of two subsequent ones?
If that's the case, what's "brightness" when you compare different light
sources? which light temperature do you use as reference (which
wavelengths to measure at?, as the visibility of some colors are quite
different when looked at on yellow versus white street lamps (not to
mention the more subtile differences within those broad color categories).
</pre>
</blockquote>
<pre wrap="">
This is all well established in the professional lighting community.
You use a meter which reads in lux, which is the SI unit of
illuminance. This is calibrated to human vision, so the various color
issue is taken care of.
The basic two things to measure are the maximum illuminance on the
roadway, and the minimum.
It's true that this is tricky with stray light. However, the real
question is how much light is on the ground, not where it comes from.
So I would suggest that tagging be of min and max illuminance. People
who don't have meters can self-calibrate and guess - it's not super hard
to learn.
Another approach is to define a few ranges in terms of lux, and have
people tag those. Some tags will be wrong, but it takes us out of
fuzzy.
</pre>
<br>
</blockquote>
<big><br>
The problem is the human eye self compensates for different light
levels... and you get incorrect data from people who have little
experience with light. I'd hoped the smart phone application
could be used .. but not so. Try it yourself? I'm yet to do some
readings with dedicated cameras to see what they say. Will need to
go from ASA, speed and shutter to lux .. that should be ok.
Dedicated light meters are not common so it is unlikely that a
mapper will have one. </big><br>
</body>
</html>