<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
{font-family:Consolas;
panose-1:2 11 6 9 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";
color:black;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
pre
{mso-style-priority:99;
mso-style-link:"Pr\00E9format\00E9 HTML Car";
margin:0cm;
margin-bottom:.0001pt;
font-size:10.0pt;
font-family:"Courier New";
color:black;}
span.PrformatHTMLCar
{mso-style-name:"Pr\00E9format\00E9 HTML Car";
mso-style-priority:99;
mso-style-link:"Pr\00E9format\00E9 HTML";
font-family:Consolas;
color:black;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:70.85pt 70.85pt 70.85pt 70.85pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body bgcolor="white" lang="FR" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal"><span lang="EN-US" style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">if it really is trying to tell the difference between a map and a photograph could you make a decision based on the presence of text, and therefore
use an OCR mechanism to judge if there are more than x words found in the image<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">-i<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<div>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal"><b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif";color:windowtext">De :</span></b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif";color:windowtext"> gdal-dev-bounces@lists.osgeo.org [mailto:gdal-dev-bounces@lists.osgeo.org]
<b>De la part de</b> Dmitriy Baryshnikov<br>
<b>Envoyé :</b> Thursday, 6 March 2014 21:16<br>
<b>À :</b> gdal-dev@lists.osgeo.org<br>
<b>Objet :</b> Re: [gdal-dev] Heuristics to classify raster data ?<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal">Hi Even,<br>
<br>
most of all depends what kind of imagery and maps you wish to classify. If the maps are classical scanned paper maps, and you want fast algorithm - the crosses of meter or degree grid can be the good pattern.<br>
But if we have areal images this will not work, as such images have crosses too. But satellites - not. May be some frame of maps can be good pattern.<br>
<br>
If you have some fragment of maps and images, I think some content analysis needed:<br>
- clustering, i.e. <a href="http://en.wikipedia.org/wiki/K-means_clustering">http://en.wikipedia.org/wiki/K-means_clustering</a><br>
-<span style="border:none windowtext 1.0pt;padding:0cm"> Neural</span> network with learning<br>
- Support vector machine i.e. <a href="http://svmlight.joachims.org/">http://svmlight.joachims.org/</a> and
<a href="http://en.wikipedia.org/wiki/Support_vector_machine">http://en.wikipedia.org/wiki/Support_vector_machine</a><br>
<br>
Also some hash comparison can be used (rather fast)<br>
- perceptual hash compare i.e. <a href="http://www.phash.org/">http://www.phash.org/</a><br>
<br>
In all cases input images should be resized to some small sizes and may be grayscaled or binarized before analysis.<br>
<br>
<br>
<o:p></o:p></p>
<pre>Best regards,<o:p></o:p></pre>
<pre> Dmitry<o:p></o:p></pre>
<p class="MsoNormal">06.03.2014 23:19, Even Rouault пишет:<o:p></o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<pre>Hi,<o:p></o:p></pre>
<pre><o:p> </o:p></pre>
<pre>I'd be interested in an algorithm to automate the classification of raster data <o:p></o:p></pre>
<pre>between maps (let's say rendering of OpenStreetMap data, or other digital <o:p></o:p></pre>
<pre>maps) one one side and aerial/satellite imagery on the other side, without <o:p></o:p></pre>
<pre>looking at metadata (bare geotiff typically). This is to help in automating <o:p></o:p></pre>
<pre>bulk of import of data from a media and establishing a first level of <o:p></o:p></pre>
<pre>classification.<o:p></o:p></pre>
<pre><o:p> </o:p></pre>
<pre>Has anyone already done that and has code and/or advice to share, or know a <o:p></o:p></pre>
<pre>software project that would do that ?<o:p></o:p></pre>
<pre><o:p> </o:p></pre>
<pre>Some ideas that came to my mind :<o:p></o:p></pre>
<pre>- maps have typically a much more reduce number of colors than imagery, but <o:p></o:p></pre>
<pre>you may have imagery that has already been transformed to 256 colors to reduce <o:p></o:p></pre>
<pre>storage space.<o:p></o:p></pre>
<pre>- maps have generally a majority color (e.g. white, green), but not in all <o:p></o:p></pre>
<pre>zones (urban zones will have more features)<o:p></o:p></pre>
<pre>- maps have higher spatial frequency (lines, text) whereas imagery will be <o:p></o:p></pre>
<pre>more continuous : use of gradient, and compute statistics on it ?<o:p></o:p></pre>
<pre><o:p> </o:p></pre>
<pre>Even<o:p></o:p></pre>
<pre><o:p> </o:p></pre>
</blockquote>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</body>
</html>