Skip to content
May 25, 2010 / Gagan

Easy way to publish raster data on web

Certainly there are many wonderful Open Source tools for web GIS application. I would add one more thing that most of these applications are more robust than propriety software. The MapTiler developed by Klokan Petr Přidal also confirms my view, by showing ease for publishing raster data on web.

You have a big image of satellite data or a plan for layout and want to show with Google maps or OpenStreetMap. If you are from remote sensing background immediately Erdas IWS comes to your  mind. But is that easy and affordable for persons like architect, town planner or for html programmer. Certainly training would be required after purchasing that.  There are other open source solution to this problem MapServer, GeoServer or mapnik.  But implementation of these software also requires a good understanding of the software as well as infrastructure for installation and running.

I had  this problem to solve, publish satellite images of  more than 1GB in size. And by googling I found the solutions leading to use of GDAL library tool gdal2tiles.py. Another GUI application developed based on gdal2tiles.py  is MapTiler. MapTiler and gdal2tiles.py are having same functionality except the GUI. Persons working on Linux would love to use the gdal2tiles.py, on other hand MapTiler allows you to process the source image step by step.

This tool MapTiler or gdal2tiles.py  process the image to be published on web application, as Google overlay or as Tile Map Service by OSGeo. An image tile cache repository is created in a directory and that can be copied to web directory for easy publishing with Google maps as tile overlay(example given in MapTiler) or with OpenLayers as TMS layer. There are many options for easy processing of the raster data. Like option to select various input image spatial reference system and formats (TIFF/GeoTIFF, MrSID, ECW, JPEG2000, Erdas HFA, NOAA BSB, JPEG etc) i.e. powered by GDAL. Another feature for zoomify effect is that gdal2tiles.py creates tile images for different zoom levels. In total in less than an hour you are ready with your satellite data processed ready for web publishing.

The feature that is missing is ‘multiple source images’, right now it accepts only single image to process. In practicality many times there are many source images because of coverage or different sensors or different layout plan. Hope this would be developed in near future.

Well a very robust and easy to use tool for Web GIS.

Advertisements

4 Comments

Leave a Comment
  1. Rakesh Modi / May 26 2010 3:42 pm

    when i try Map Tiler with image size like 1.20 GB in PNG format. It gives error:-
    unexpected errror occured:
    list index out of range

  2. gaganb / May 27 2010 10:22 am
  3. Gagan / May 27 2010 5:29 pm

    Hi,

    There seems to be a issue with png as input image. Can you convert your image into geotiff and try with geotiff.

    Gagan

  4. Rakesh Modi / Jun 7 2010 11:04 am

    hi,
    I got solved my problem with geotiff format, when you convert PNG into geotiff its size will increase.
    I got some problem when i rendered this tiles with openlayers and i would like to share what’s the problem and how i solved.
    Problem is openlayers shows pink tiles instead of images.
    how i solved :-
    1) When I run openlayers.html file (made with map tiler) it works.
    2) Copy the TMS layer and paste into my file.
    3) Copy getTiles_FromUR function and paste.
    4) Change the TMS path according to your file directory.
    5) change the folder structure like cut my tiles for 9-17 zoom level then replace this folders name 1-9 thas it……………
    🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: