Dear Dilip:
the projection depends upon the source. In most cases, the only source
that exists is the Census district handbook maps. These in turn have
been prepared from taluka level maps that the Directorate of Land
Records and Survey Settlement creates from village-level cadastral maps.
Now this method by definition means there is no original projection
system, and that these maps are not geo-corrected (not
terrain-corrected) as the originals are based on chain survey and plain
table surveys, not theodolite surveys. So unless someone has done some
correction work before creating the maps you used, they would have
significant limitations in terms of geo-referencing. Our experience (we
created the Karnataka layer from scratch, by digitizing the individual
taluka-level village boundary maps by hand) tells us RMS errors of 2-3km
for the state as a whole would be normal. What we did was to
georeference by matching the state boundary so created to the Survey of
India state boundary, which is available and is a properly projected
layer. (Simply creating a prj file for the existing map does not work,
because the scans will be anyway in centimetre or mm or inch units. So
you need pucca geo-rectification.) Fairly easily done in QGIS.
I hope this helps. Am still digging for the layers we supposedly have.
Sharad
On 18-01-2016 12:57, Dilip Damle wrote:
Hi,
I think I will share the data anyway after doing some checks. I am not
attempting the attribute linking right now because that will be a
bigger effort.
I tried to georeference it.
My approach was as follows.
So far I have not used any transformations which distorts the shape as
it was in the PDF file.
I imported each district separately and have been able to match them
with each other to make the whole of Gujarat using only MOVE, SCALE
(EQUAL XY) and ROTATE.
Practically I used the Autocad ALIGN command.
Which means I have exactly same shape as in PDF files.
Then I thought I will guess the original projection and see if it
matches and then reproject it rather than Doing any
rubbersheeting/other arbritrary distortions to shape.
I tried LCC projection for India , UTM 43N, and LL84 in that order.
They do not seem to match.
Anyone may suggest any other possible projection.
Otherwise I will check it and share it as it is.
This is the summary of the Data :
33 districts : 114 Polylines in Autocad (because of Islands and enclaves)
379 Taluka Polylines
20128 Village polylines
15804 Text names
16137 Markers manly the Village dot.
On Monday, January 18, 2016 at 12:41:51 AM UTC+5:30, Sharad Lele wrote:
Dilip:
before you put in more effort, could you pause for 24hrs? I think
I may have the Gujarat village boundary file for 1991 in vector
(un-georeferenced) format. Will know by tomorrow.
Sharad
On 18-01-2016 00:35, Dilip Damle wrote:
HI,
I have been able to put together Gujarat boundaries from the PDF
files in the links above.
Right now it is an Autocad Drawing with Village, Taluk, and
District boundaries.
I will try to georeference it and share as .dwg somewhere.
It is about 55 MB file.
Shape file may be a long way away since there is no point in it
without at least village name attributes.
Links names may take take some time. Some one else also can try
it may be faster than me.
enclosed two snapshots. DWG file tomorrow/day after.
I will create a new discussion thread for sharing/discussing the
dwg file.
On Friday, December 18, 2015 at 12:16:12 PM UTC+5:30, Nisha
Thompson wrote:
Hi All,
There have been tons of threads on getting more local area
maps. So lets try to work together and get it done.
We started a github wiki to collect links to PDFs we have
found that need to be digitalized.
https://github.com/datameet/maps/wiki/Maps-we-found-that-need-to-be-digitalized
<https://github.com/datameet/maps/wiki/Maps-we-found-that-need-to-be-digitalized>.
How should we do this? Who can help?
I think we do some skill sharing hangouts on how to
digitalize if people are interested and slowly work on
freeing sets of maps.
What do you all think?
Nisha
--
Nisha Thompson
DataMeet.org
ni...@datameet.org
skype: nishaqt
mobile: 962-061-2245
--
Datameet is a community of Data Science enthusiasts in India.
Know more about us by visiting http://datameet.org
---
You received this message because you are subscribed to a topic
in the Google Groups "datameet" group.
To unsubscribe from this topic, visit
https://groups.google.com/d/topic/datameet/WF6eTOatjZ0/unsubscribe
<https://groups.google.com/d/topic/datameet/WF6eTOatjZ0/unsubscribe>.
To unsubscribe from this group and all its topics, send an email
to datameet+u...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/d/optout
<https://groups.google.com/d/optout>.
--
Datameet is a community of Data Science enthusiasts in India. Know
more about us by visiting http://datameet.org
---
You received this message because you are subscribed to a topic in the
Google Groups "datameet" group.
To unsubscribe from this topic, visit
https://groups.google.com/d/topic/datameet/WF6eTOatjZ0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to
datameet+unsubscr...@googlegroups.com
<mailto:datameet+unsubscr...@googlegroups.com>.
For more options, visit https://groups.google.com/d/optout.
--
Datameet is a community of Data Science enthusiasts in India. Know more about
us by visiting http://datameet.org
---
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 datameet+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
