{
  "id": 20663,
  "title": "Suitably licensed External Data allowed?",
  "url": "/competitions/draper-satellite-image-chronology/discussion/20663",
  "author_name": "",
  "post_date": "2016-05-03T11:00:23.947Z",
  "votes": null,
  "comment_count": 2,
  "views": 474,
  "content": "<p>Basically the competition seems to require learning how images taken-from-above change over time. </p>\n\n<p>Given the vast complexity and semantic knowledge possibly required for this, are we allowed to use external data to train a machine learning algorithm on what changes over time can look like for different environments?</p>\n\n<p>As we are allowed to hand-label, I assume we are allowed to use external data sets as well if they are licensed in such a way that they can be used commercially?</p>\n\n<ul>\n<li>foliage undergoes seasonality</li>\n<li>buildings are constructed (or destroyed) in a certain way</li>\n<li>tracks appear on ground faster than they disappear</li>\n<li>putting up a &quot;camp&quot; follows different sequence than pulling it down.</li>\n</ul>\n\n<p>Johannes</p>",
  "messages": [
    {
      "id": "118342",
      "postDate": "05/03/2016 11:00:23",
      "content": "<p>Basically the competition seems to require learning how images taken-from-above change over time. </p>\n\n<p>Given the vast complexity and semantic knowledge possibly required for this, are we allowed to use external data to train a machine learning algorithm on what changes over time can look like for different environments?</p>\n\n<p>As we are allowed to hand-label, I assume we are allowed to use external data sets as well if they are licensed in such a way that they can be used commercially?</p>\n\n<ul>\n<li>foliage undergoes seasonality</li>\n<li>buildings are constructed (or destroyed) in a certain way</li>\n<li>tracks appear on ground faster than they disappear</li>\n<li>putting up a &quot;camp&quot; follows different sequence than pulling it down.</li>\n</ul>\n\n<p>Johannes</p>",
      "rawMarkdown": "Basically the competition seems to require learning how images taken-from-above change over time. \r\n\r\nGiven the vast complexity and semantic knowledge possibly required for this, are we allowed to use external data to train a machine learning algorithm on what changes over time can look like for different environments?\r\n\r\nAs we are allowed to hand-label, I assume we are allowed to use external data sets as well if they are licensed in such a way that they can be used commercially?\r\n\r\n- foliage undergoes seasonality\r\n- buildings are constructed (or destroyed) in a certain way\r\n- tracks appear on ground faster than they disappear\r\n- putting up a \"camp\" follows different sequence than pulling it down.\r\n\r\nJohannes",
      "votes": null
    },
    {
      "id": "118374",
      "postDate": "05/03/2016 12:52:32",
      "content": "<p>Yes to both questions. Use of external data is permitted.</p>",
      "rawMarkdown": "Yes to both questions. Use of external data is permitted.",
      "votes": null
    },
    {
      "id": "118390",
      "postDate": "05/03/2016 13:56:31",
      "content": "<p>Please find some possible external data sets below (I haven't validated the license on all of them yet):</p>\n\n<p>Please note that the competition photos may have been taken by plane and not by satellite.</p>\n\n<p>The Landsat-8 data set seems like a promising candidate as an external data set...</p>\n\n<ul>\n<li><a href=\"http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/\">http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/</a></li>\n<li><a href=\"http://gisgeography.com/free-satellite-imagery-data-list/\">http://gisgeography.com/free-satellite-imagery-data-list/</a></li>\n<li><a href=\"http://landsat.gsfc.nasa.gov/?page_id=2391\">http://landsat.gsfc.nasa.gov/?page_id=2391</a></li>\n<li><a href=\"http://glcf.umd.edu/data/ndvi/\">http://glcf.umd.edu/data/ndvi/</a></li>\n<li><a href=\"http://www.sciencedirect.com/science/article/pii/S1364815213002648\">http://www.sciencedirect.com/science/article/pii/S1364815213002648</a></li>\n<li><a href=\"https://github.com/dutri001/bfastSpatial\">https://github.com/dutri001/bfastSpatial</a></li>\n<li><a href=\"http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/\">http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/</a></li>\n<li><a href=\"https://eros.usgs.gov/satellite-imagery\">https://eros.usgs.gov/satellite-imagery</a></li>\n<li><a href=\"http://www.pancroma.com/data.html\">http://www.pancroma.com/data.html</a></li>\n<li><a href=\"http://yceo.yale.edu/image-sources\">http://yceo.yale.edu/image-sources</a></li>\n</ul>\n\n<p>Also see:</p>\n\n<ul>\n<li><a href=\"http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html\">http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html</a></li>\n</ul>",
      "rawMarkdown": "Please find some possible external data sets below (I haven't validated the license on all of them yet):\r\n\r\nPlease note that the competition photos may have been taken by plane and not by satellite.\r\n\r\nThe Landsat-8 data set seems like a promising candidate as an external data set...\r\n\r\n- http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/\r\n- http://gisgeography.com/free-satellite-imagery-data-list/\r\n- http://landsat.gsfc.nasa.gov/?page_id=2391\r\n- http://glcf.umd.edu/data/ndvi/\r\n- http://www.sciencedirect.com/science/article/pii/S1364815213002648\r\n- https://github.com/dutri001/bfastSpatial\r\n- http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/\r\n- https://eros.usgs.gov/satellite-imagery\r\n- http://www.pancroma.com/data.html\r\n- http://yceo.yale.edu/image-sources\r\n\r\nAlso see:\r\n\r\n- http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 118374,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "05/03/2016 12:52:32",
      "content": "<p>Yes to both questions. Use of external data is permitted.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 118390,
      "author_name": "ahlmann",
      "author_url": "",
      "post_date": "05/03/2016 13:56:31",
      "content": "<p>Please find some possible external data sets below (I haven't validated the license on all of them yet):</p>\n\n<p>Please note that the competition photos may have been taken by plane and not by satellite.</p>\n\n<p>The Landsat-8 data set seems like a promising candidate as an external data set...</p>\n\n<ul>\n<li><a href=\"http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/\">http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/</a></li>\n<li><a href=\"http://gisgeography.com/free-satellite-imagery-data-list/\">http://gisgeography.com/free-satellite-imagery-data-list/</a></li>\n<li><a href=\"http://landsat.gsfc.nasa.gov/?page_id=2391\">http://landsat.gsfc.nasa.gov/?page_id=2391</a></li>\n<li><a href=\"http://glcf.umd.edu/data/ndvi/\">http://glcf.umd.edu/data/ndvi/</a></li>\n<li><a href=\"http://www.sciencedirect.com/science/article/pii/S1364815213002648\">http://www.sciencedirect.com/science/article/pii/S1364815213002648</a></li>\n<li><a href=\"https://github.com/dutri001/bfastSpatial\">https://github.com/dutri001/bfastSpatial</a></li>\n<li><a href=\"http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/\">http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/</a></li>\n<li><a href=\"https://eros.usgs.gov/satellite-imagery\">https://eros.usgs.gov/satellite-imagery</a></li>\n<li><a href=\"http://www.pancroma.com/data.html\">http://www.pancroma.com/data.html</a></li>\n<li><a href=\"http://yceo.yale.edu/image-sources\">http://yceo.yale.edu/image-sources</a></li>\n</ul>\n\n<p>Also see:</p>\n\n<ul>\n<li><a href=\"http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html\">http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html</a></li>\n</ul>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "118342": "Basically the competition seems to require learning how images taken-from-above change over time. \r\n\r\nGiven the vast complexity and semantic knowledge possibly required for this, are we allowed to use external data to train a machine learning algorithm on what changes over time can look like for different environments?\r\n\r\nAs we are allowed to hand-label, I assume we are allowed to use external data sets as well if they are licensed in such a way that they can be used commercially?\r\n\r\n- foliage undergoes seasonality\r\n- buildings are constructed (or destroyed) in a certain way\r\n- tracks appear on ground faster than they disappear\r\n- putting up a \"camp\" follows different sequence than pulling it down.\r\n\r\nJohannes",
    "118374": "Yes to both questions. Use of external data is permitted.",
    "118390": "Please find some possible external data sets below (I haven't validated the license on all of them yet):\r\n\r\nPlease note that the competition photos may have been taken by plane and not by satellite.\r\n\r\nThe Landsat-8 data set seems like a promising candidate as an external data set...\r\n\r\n- http://earthzine.org/2009/03/02/terralook-free-time-series-satellite-images-for-busy-people/\r\n- http://gisgeography.com/free-satellite-imagery-data-list/\r\n- http://landsat.gsfc.nasa.gov/?page_id=2391\r\n- http://glcf.umd.edu/data/ndvi/\r\n- http://www.sciencedirect.com/science/article/pii/S1364815213002648\r\n- https://github.com/dutri001/bfastSpatial\r\n- http://creativecommons.org.au/blog/2012/09/geoscience-australia-to-license-landsat-8-satellite-images-under-cc-by/\r\n- https://eros.usgs.gov/satellite-imagery\r\n- http://www.pancroma.com/data.html\r\n- http://yceo.yale.edu/image-sources\r\n\r\nAlso see:\r\n\r\n- http://coastal.er.usgs.gov/wetlands/ofr97-287/chapter1.html"
  },
  "source": "meta"
}