<p>Low-Memory GEOgraphic information system library - lmgeo</p>

<p>Lmgeo is a Python raster GIS library with low memory requirements. Aim is to provide software building blocks with a low-memory footprint that can be used to automate specific tasks without necessarily requiring common GIS software like ArcGIS or QGIS. </p>

<p>These building blocks can be used in the following ways:</p>

<ul>
<li>as basis for easy-to-use GIS applications with an interface designed according to the so-called LIME approach (less is more). Non-expert users can be taught how to use such applications for recurring tasks which would otherwise require the intervention of expert users.</li>
<li>as part of integrated software that deals with more than only the domain of GIS</li>
<li>to automate tasks which would cause ArcGIS / QGIS to reserve a lot of memory and would slow down the execution </li>
<li>to automate tasks that are error-prone in ArcGIS - e.g. certain raster GIS operations.</li>
</ul>

<p>The folllowing principles were used in the implementation of lmgeo:</p>

<ul>
<li>should work on any platform (Windows, Linux or Mac)</li>
<li>the library was implemented as much as possible in pure python with as few dependencies as possible</li>
<li>modular design, allowing easy replacement of components </li>
<li>a common interface for all classes dedicated to image formats</li>
<li>reading, processing and writing of image data as much as possible line by line, to limit memory usage.</li>
</ul>

<p>Supported image formats:</p>

<ul>
<li>ASCII grid (*.asc)</li>
<li>Floating point rasters (*.flt)</li>
<li>Band interleaved by line (*.bil)</li>
<li>Band sequential (*.bsq)</li>
<li>GeoTiff (*.tif)</li>
<li>HDF5 (*.hdf5)</li>
<li>NetCDF4 (*.nc4).</li>
</ul>

<p>Dependencies:</p>

<ul>
<li>numpy</li>
<li>pylibtiff</li>
<li>PyTables</li>
<li>netCDF4.</li>
</ul>

<p>More documentation is in preparation. Feel free to ask if you need assistance.</p>
