On Sun, Jun 16, 2013 at 12:01 AM, Kavita Arora <[email protected]> wrote: > I am part of one focus group of the larger "Save Bangalore Lakes" community > initiative that aims to provide an open forum to discuss the degrading (or > improving) condition of Bangalore's Lakes. > > As part of our ongoing research to obtain as much information about the > lakes and other water bodies -- we are piloting a new data science exercise > where we plan to attempt collecting as much raw data (from scratch) about > the lakes - including catchment areas, lake peripheries and other measures. >
nice to hear. > Although some of this work has been done before, in order to be completely > objective and not introduce too much bias into our project -- we are trying > to figure out how to capture and analyze raw data collected from scratch - > such that we can come up with insights we may have overlooked before. > > Was hoping to get some ideas from some of you in this group as to how to go > about the following: > - How to go about capturing remotely sensed data > - How to convert raw data captured via Remote Sensing into shape files > - Understanding the process of building a basemap from scratch > - Annotating, tagging and modeling the adapted data > I do not know about the way you are getting the satellite data. You will need to have some idea about the topography of the place beforehand (you can proceed on the raw data otherwise too but this would make things easier) : this will help you with generation of the training set. What kind of things are you looking for ? Make a list. Can you point them out manually: if this is on target you will have no real problems. Even with Bayesian learning techniques you can get great maps. GNU/R + (views :machine learning + bayes) + grass gis Rest should be easy. Best A. Mani -- A. Mani CU, ASL, AMS, CLC, CMS http://www.logicamani.in -- For more details about this list http://datameet.org/discussions/ --- You received this message because you are subscribed to the Google Groups "datameet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
