<div dir="ltr">Hi Akbar,<br><br>The idea of implementing the RNN inside a Python plugin for QGIS is great. Beside my experience as a QGIS user, there's already a <a href="https://plugins.qgis.org/plugins/LAPIGTools/" target="_blank">QGIS plugin</a> developed by the lab where I work, so I can take advantage of their expertise. I'll seriously consider this possibility.<br><br>About the use of the temporal series, the information held in time between observations <span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:small;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline">has a great value for some targets</span>. For example, it is quite difficult to discriminate natural grasslands and pasturelands when relying only on spatial information, but, in the other hand, the temporal signature of the pixels holds precious information to classify between those two classes. Other example is the identification of different land uses such sugar cane and crops, which present big spectral variety in time and can easily mislead a spatial model. All those targets are of great interest for research in Brazil, and I am working on models trained on that kind of classes.<br><br>Cheers, and thank you for the idea.</div><div class="gmail_extra"><br><div class="gmail_quote">2018-03-22 17:50 GMT-03:00 Akbar Gumbira <span dir="ltr"><<a href="mailto:akbargumbira@gmail.com" target="_blank">akbargumbira@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi,<div><br></div><div>It sounds like a nice idea. IMO you can also implement it as a python plugin for QGIS. If you need a simple RNN implementation, you can use Pyrenn (<a href="http://pyrenn.readthedocs.io/en/latest/" target="_blank">http://pyrenn.readthedocs.io/<wbr>en/latest/</a>) and ship it with the plugin.</div><div><br></div><div>Just wondering, why does it need to consider the temporal information? What's the relevance of the history of a spatial area to the classification? Shouldn't it just classify based on the latest data?<br><br>Cheers</div></div><div class="gmail_extra"><br><div class="gmail_quote"><div><div class="h5">On Thu, Mar 22, 2018 at 5:49 PM, Evandro Carrijo <span dir="ltr"><<a href="mailto:evandro.taquary@gmail.com" target="_blank">evandro.taquary@gmail.com</a>></span> wrote:<br></div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5"><div dir="ltr"><div style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255)">Hello there!</div><div style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255)"><br></div><div style="text-align:start;text-indent:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255)"><span style="font-size:12.8px">I'm a Computer Science Master's Degree student whose research if focused on Deep Learning algorithms applied to Remote Sensing. Currently working at the </span><a href="https://github.com/lapig-ufg" style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-transform:none;white-space:normal;word-spacing:0px" target="_blank">Laboratory of Image Processing and Geoprocessing</a><span style="font-size:12.8px"> settled at Federal University of Goiás - Brazil. I'm also member of the High Performance Computing group of the same university (more information</span><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-transform:none;white-space:normal;word-spacing:0px"> </span><a href="http://dgp.cnpq.br/dgp/espelhogrupo/7985061476854055" style="color:rgb(17,85,204);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-transform:none;white-space:normal;word-spacing:0px" target="_blank">here</a><span style="font-size:12.8px">). </span><br><br><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-transform:none;white-space:normal;word-spacing:0px">Below I present an</span><span style="font-size:12.8px"> idea to explain how I can contribute to OSGeo community and I'm seeking for mentors interested in assist my development. Please, feel free to argue me any matter about the project idea. </span><br><br><span style="font-size:12.8px">I would also appreciate a lot if you guys indicate a potential interested mentor to my project idea or a OSGeo Project suitable to it.</span><br><br><span style="font-size:12.8px">Hope there's some Interested ones out there!</span><br><br><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:small;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:600;letter-spacing:normal;text-transform:none;white-space:normal;word-spacing:0px;box-sizing:border-box">Idea</span><br></div><p style="color:rgb(34,34,34);font-family:arial,sans-serif;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;font-size:small"></p><p style="font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);box-sizing:border-box;margin-top:0px;margin-bottom:16px;color:rgb(36,41,46);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";font-size:14px">The increasing number of sensors orbiting the earth is systematically producing larger volumes of data, with better spatiotemporal resolutions. To deal with that, better accurate machine learning approaches, such as Deep Learning (DL), are needed to transform raw data into applicable Information. Several DL architectures (e.g. CNN, semantic segmentation) rely only at spatial dimension to perform, for example, land-cover/land-use (LCLU) maps, disregarding the temporal dependencies between pixels observations over the time. Also, high-res remote sensing data (e.g. Planet, Sentinel) may provide more consistent time-series, that can be use in the identification of important LCLU classes, like crop, pastureland and grasslands.</p><p style="font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);box-sizing:border-box;margin-top:0px;margin-bottom:16px;color:rgb(36,41,46);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";font-size:14px">This potential can be explored using Recurrent Neural Networks (RNN), a specific family of DL approaches which can take into account time dimension. A promising project idea would be implement a RNN approach (e.g. LSTM) to classify a Sentinel time-series, that will organize and preprocess an input data set (e.g. labeled time-series), calibrate and evaluate a RNN model, and finally classify an entire region (i.e. 2 or 3 scenes) to produce a map for one or more LCLU class. It will be great evaluate the accuracy and the spatial consistent of a map produced with a RNN approach.</p><p style="font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);box-sizing:border-box;margin-top:0px;margin-bottom:16px;color:rgb(36,41,46);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";font-size:14px">A simple example on classifying LCLU with two classes (pastureland and non-pastureland):</p><p style="font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);box-sizing:border-box;margin-top:0px;margin-bottom:16px;color:rgb(36,41,46);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";font-size:14px"><a href="https://user-images.githubusercontent.com/37085598/37687055-cc5236a8-2c78-11e8-8892-d113df44e235.jpg" style="color:rgb(3,102,214);box-sizing:border-box;background-color:transparent;text-decoration:none" target="_blank"><img src="https://user-images.githubusercontent.com/37085598/37687055-cc5236a8-2c78-11e8-8892-d113df44e235.jpg" alt="itapirapua" class="m_1648289244345588669m_6594617369945471126gmail-CToWUd" style="box-sizing:content-box;border-style:none;max-width:100%;background-color:rgb(255,255,255)"></a><br style="box-sizing:border-box"><strong style="box-sizing:border-box;font-weight:600">Target region (input)</strong></p><p style="font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);box-sizing:border-box;margin-top:0px;margin-bottom:16px;color:rgb(36,41,46);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";font-size:14px"><a href="https://user-images.githubusercontent.com/37085598/37732806-ec792782-2d24-11e8-8ad9-18867768e998.jpg" style="color:rgb(3,102,214);box-sizing:border-box;background-color:transparent;text-decoration:none" target="_blank"><img src="https://user-images.githubusercontent.com/37085598/37732806-ec792782-2d24-11e8-8ad9-18867768e998.jpg" alt="itapirapua_ref" class="m_1648289244345588669m_6594617369945471126gmail-CToWUd" style="box-sizing:content-box;border-style:none;max-width:100%;background-color:rgb(255,255,255)"></a><br style="box-sizing:border-box"><strong style="box-sizing:border-box;font-weight:600">Generated LCLU map (output)</strong></p><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;font-size:small;background-color:rgb(255,255,255);float:none;display:inline">Best,</span><br style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><br style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);float:none;display:inline">Evandro Carrijo Taquary</span><br style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:12.8px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial;background-color:rgb(255,255,255);float:none;display:inline">Federal University of Goiás</span><br></div>
<br></div></div><span class="">______________________________<wbr>_________________<br>
SoC mailing list<br>
<a href="mailto:SoC@lists.osgeo.org" target="_blank">SoC@lists.osgeo.org</a><br>
<a href="https://lists.osgeo.org/mailman/listinfo/soc" rel="noreferrer" target="_blank">https://lists.osgeo.org/mailma<wbr>n/listinfo/soc</a><br></span></blockquote></div><span class="HOEnZb"><font color="#888888"><br><br clear="all"><div><br></div>-- <br><div class="m_1648289244345588669gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div><p><b style="font-size:12.8px">Akbar Gumbira </b><br></p></div><div><b style="font-size:12.8px"><a href="http://www.akbargumbira.com" target="_blank">www.akbargumbira.com</a></b></div></div></div></div></div></div>
</font></span></div>
</blockquote></div><br></div>