<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<font size="-1">I<font size="-1"> sometimes work <font size="-1">on
Senegal from Bing imagery,<font size="-1"> and I'll </font>hopefully
convinc<font size="-1">e</font> some <font size="-1">Senegalese
friends </font></font></font>to help me with names and POI.
I took a look at the competition's progr<font size="-1">ess and I
have been very impressed : whereas <font size="-1">O</font>penstreetmap</font>
co<font size="-1">uld formerly claim better coverage in Africa, <font
size="-1">it is now lagging in volume behin<font size="-1">d <font
size="-1">Google - take a look at <a class="moz-txt-link-freetext" href="http://goo.gl/maps/mH3pK">http://goo.gl/maps/mH3pK</a>
(equivalent area in OSM at <a class="moz-txt-link-freetext" href="http://osm.org/go/azSB9oX">http://osm.org/go/azSB9oX</a>--).
Google now features impressive mileage of dirt tracks</font></font></font></font>
and residential dirt streets in obscure backwaters.<br>
<br>
Surely the Google su<font size="-1">rvey cars are not roaming
those places - or I would be astonished. <font size="-1">D</font>oes
</font>anyone knows how they do it ?<font size="-1"> A couple of
years ago, I had posted my suspicions about <font size="-1">them
using automated tracing techniques in general and street grid
detection in particular:
<a class="moz-txt-link-freetext" href="http://lists.openstreetmap.org/pipermail/talk/2010-January/046539.html">http://lists.openstreetmap.org/pipermail/talk/2010-January/046539.html</a></font></font>
- but since then it seems that their techniques have improved a
lot. Does anyone have i<font size="-1">nf<font size="-1">ormation
about the technology they use and how it could benefit OSM ? </font></font></font><font
size="-1"><font size="-1"><font size="-1">Maybe they just have an
army of tracers<font size="-1">, but <font size="-1">automation
seems more like <font size="-1">how Google solves
problems.<br>
<br>
<font size="-1">Last time <font size="-1">played with
it, the results of the Bing road detect API did not
look satisfactory </font></font>to me<font
size="-1"> - but <font size="-1">there may have been
<font size="-1">progress since those early days. </font></font></font>But
<font size="-1">wouldn't road detection perform better
using imagery in different bands -<font size="-1">
multispectral or even hyperspectral i<font size="-1">magery
? The Bing satellite imagery render</font></font></font>
<font size="-1"><font size="-1">uses <font size="-1">false
color </font>images fit for pleasant human use<font
size="-1"><font size="-1"> - but <font size="-1">surely
the raw<font size="-1"> captured data</font></font></font></font></font></font>
is richer. Are there any remote sensing specialist here
to enlighten about t<font size="-1">hat ?<br>
<br>
</font></font></font></font></font></font></font>
</body>
</html>