5

I’m not an expert in GIS but recently I studied a bit, came on different sites. Among others QGIS, MangoMap, ArcGIS, … etc

A group of 6 small towns in Belgium wants to join efforts for bringing their graveyard online.

I’m wondering if the following procedure could be effective…

  • Take a photograph with coordinates from the sites (with a drone instead of an airplane…?)
  • Import the picture as a layer in AUTOCAD
  • In another layer draw Lines/Polylines over the existing graves
  • Add drawing for known future implantation of graves (not much)
  • Export the AUTOCAD files to shape
  • Upload the files to f.e. MangoMaps or another server
  • Place one POINT per deceased person on the graves (unique identification)
  • Make an Excel file with the same unique identifier per deceased person
  • Connect the Excel file with the files in MangoMap (upload)
  • Extra persons can easily be added in the file and re-uploaded

One town (mine) has a shapefile of the site. I can import this file in AUTOCAD and add the new graves. Some errors also have to be corrected. Can also be done in QGis but not sure what’s easier.

Towns who have no shape files don’t need them. Just:

  • upload the picture
  • put the number of POINTS (deceased persons) on the graves (can be several points on one grave)
  • export the file as Excel (believe that’s possible)
  • add the deceased persons in the EXCEL file at the correct unique identifier
  • upload the Excel file

Could this be a good way to do it?

  • 2
    QGIS + PostGIS for free option. Database tables could be something like : area, line, point, person, road where area,line and point table have rowid, geometry, type... and person table has rowid, person details and geometry id for id in area table. same logic with other types. After that it is just using SQL tools – simplexio Mar 31 '15 at 8:14
  • The shapefile is the native vector format for QGIS and if you have a decent one for the area of interest I'd suggest that QGIS is the way to go. If the shapefile is not confidential it would help if you could post a link to it. – nhopton Mar 31 '15 at 11:38
  • PatrickVanDenNoortgaete, is the purpose of the GIS merely for the towns to add a location context to their interment data, or is the goal to map it on the web? If it's the latter, I'll volunteer to help you stand up a customized Leaflet + Geoserver + PostGIS website complete with a name search / zoom-to grave feature. As this is something I've wanted to get into, myself, and you're going the exact same direction I wanted to go with it--i.e. using a drone to get high-res imagery of the cemetery grounds. .....one thing to be mindful of, you'll want leaf-off imagery! – elrobis Mar 31 '15 at 14:52
  • 1
    @elrobis We need a groundplan. I thought the easiest way is the picture. Otherwise we need to go out and measure and draw it in AUTOCAD. – Patrick Van Den Noortgaete Mar 31 '15 at 15:17
  • @elrobis Nice to offer the help. I'm just starting top explore, so don't know yet the best way. Each town will need a groundplan. I thought the easiest way is the picture. Otherwise we need to go out, measure and draw it in AUTOCAD. Or probably directly in QGis. With the picture and importing it as a layer I just have to hover over it and click to have the drawing. Then it is just a matter of linking the data. The cost for a drone picture may be worth it, didn't ask prices yet. But measuring the whole graveyard would also cost some time and $$$. – Patrick Van Den Noortgaete Mar 31 '15 at 15:24
2

This is something I've given some thought to so I'll offer some reactions and feedback.

Common, Recognized Coordinate Systems

First--it's perfectly fine to start the interment dataset in Excel. My guess is you have some "army" of students, genealogy enthusiasts, and possibly some cemetery employees building the point data with GPS points. To avoid any frustrations down the road, just make sure everyone is taking LAT/LON points. I've seen people take points in UTM before because they didn't know any better and it's like, "great.. now I've got to surmise what UTM zone they were in.." These days most units seem to be in Lat/Lon by default, but the scenario I'm recalling was an older Garmin unit. Once you're ready to publish your data, you'll want to save your Excel data as a CSV file type. ..given all the European names, I wonder how that may impact any diacritics? That's a potential character set encoding issue. Perhaps someone will comment on that point..

Second--I suppose you've got AutoCAD and you know it well which is why you want to use it for the line work. That's fine too, just make sure to setup your project to use a common, recognized coordinate system, this will ensure simple transfer into other GIS systems when you're ready to publish your data. I don't know AutoCAD, so I don't know how to advise you on this point, but assuming you're solid with CAD, I expect you probably know what I'm talking about. UTM 31 North with a WGS 84 spheroid might be a good fit--but you may be aware of a better CRS. Just make sure it's a popular CRS before you're knee-deep in the project, otherwise you'll have growing pains when you're ready to move the data into PostGIS.

BTW, I realize I just said to stay in lat/lon for the points, then recommended UTM 31N for the line work. But that's only for simplicity during the capture/drawing phases. For anyone collecting points it will be more intuitive, and preventing confusion there is important. Then, for CAD, I'm assuming you need a Cartesian grid. Once you're ready to publish the data, you can merge everything into the same system.

Web-facing Architecture

Finally--once you're ready to publish--this is where I would deviate from what you proposed in some respects. I agree with translating the CAD data to shapefile to make it more portable. But were it me, I'd use OGR2OGR to load both the Excel data (as CSV) and the AutoCAD data (as SHP) into PostGIS. Next, I'd stand-up a Geoserver instance to serve your drone imagery. With the data ready, I'd use the Leaflet library to handle the web cartography.

Lastly, and admittedly some may disagree with me on this point--I'd write PHP methods to fetch the vector point/line data from PostGIS directly and push them into the map as interactive data layers. Where interactive means the end user can hover/click on the grave plots/points and get super-quick feedback. There are other, less interactive ways to load in that data with Geoserver, itself, but I like to keep the site feeling fast and rich, and it's easy enough to do this with some "bare metal" PHP endpoints.

As mentioned in my comment, above, I'd furture recommend acquiring your drone imagery in the winter so that you can see through the trees (if that's a problem)! Otherwise, while your point data and your line work will still represent the grave site, you won't be able to see grave itself. On some smaller cemeteries with a great number of mature trees, this creates a seemingly incomplete user experience. (Not sure how to describe that, really.)

To put some icing on the cake, it's pretty easy to put together a little search-by-name component that spits out search suggestions as the user types. For an example, check out this demo site I built awhile back for a friend: http://www.solposition.com/ Unfortunately it's a bit dated now, built in Flash using the OpenScales library, but in the search bar, type "wanna" and watch it spit out options corresponding to that street. That's the idea. In fact, in many respects that SolPosition site is exactly what I'm recommending for you, only with your custom interment data, rather than the solar stuff.

  • Need to digest a little what you wrote :-) – Patrick Van Den Noortgaete Mar 31 '15 at 15:34
  • Sure. If you have any questions or feedback later put them here as a comment and I'll see. If nothing else I'd be interested in following your project. – elrobis Mar 31 '15 at 15:54
  • <My guess is you have some "army" of <…SNIP…> building the point data with GPS points.> Unfortunately, most here are analfabetics when it comes to computers and related technology. But that’s good for me, so I can play around and learn and explore a bit  Just recently a joined effort was started up, but it is not me who is in charge. But I have an interest in putting this together. This said, I think I do not really need people to go around on the graveyard with GPS locators… More in next... (not enough characters left...) – Patrick Van Den Noortgaete Apr 1 '15 at 7:01
  • 1
    I simply loaded the specific shapefile with the “graves as polygones” in QGIS, added another layer where I put POINTS on the graves (one per deceased person), some graves will have 2, 3 or even more. There is even a pillar with the names of people whose ashes are scattered in a piece of land of 10 by 10 m. Doesn't matter, I put one POINT for each in the grass area or even on the POLY of the pillar. Now I need to connect my EXCEL file to each POINT with the "unique identifier". A little struggling with it as it is a few months ago that I have been studying QGis. Will have to start it over… – Patrick Van Den Noortgaete Apr 1 '15 at 7:02
1

The work should be based on a good data model. You should make sure your objects adhere to national standards - if available. In Norway a lot of work has gone into standardizing graveyards.

A PDF document is available describing this standard. It is unfortunately only available in Nrwegian, but Google translate should be able to help you out. From the foreword:

Version 4.5 has been developed by a project SOSI Graveyard where KA church employers and organization has project ownership, and where there are participants from the previous workgroup in addition to several Joint Council / local authorities and commercial operators.

From the text describing the scope of the standard:

The catalog includes objects that are peculiar to graveyards, their characteristics and what relationships that can exist between them. It shall not cover objects found at burial sites, that are already covered by other disciplines in SOSI. In the product specification, such occurrences can be retracted so that this will emerge as a unified tool to use for mapping of graveyards. According to the Regulations to the Act relating to cemeteries, cremation and burial (funeral law) § 4, paragraph 3 cultural heritage objects shall be indicated on the grave map. SOSI Graveyard includes not own objects of cultural heritage because it is a separate SOSI standard for localizing cultural heritage.

This is elaborated in the PDF-document. Not sure if this document is available in English or other major European languages.

0

I think ArcGIS and ArcGIS online can handle all of that itself, but it's perhaps one of the more pricey software solutions.

ArcGIS will let you import any image of the site and georeference it, so you may be able to avoid costly drone procedures. You can draw features very easily on top of images. You can then link tables with some features to add attributes, such as people's names, date of death etc.. For example, a polygon grave plot will have data associated to it.

I don't know much about the other software options you mentioned, but having done a bit of drone imagery before I'll just say it's pretty tough and pricey! Most developed countries have fairly accurate aerial data you can purchase fairly cheaply, and some of the big map companies allow you to purchase small maps as well. If you could use a diagram rather than an actual image, it is probably much easier to make a diagram yourself, starting with a few waypoints at key spots loaded from a phone or handheld GPS unit. I work doing biology research, and I easily created a map of our path system in the forest by walking along each path a few times taking lots of points and averaging the resulting lines.

  • Our town is a little remote to Belgian standards :-) Lots of imagery, even on GoogleMaps, is outdated. There is even an area now equipped with a tunnel etc, which still shows the old Railway crossing... Imagery may be an idea for the towns who have no *.shp files. I will probably have to draw floorplans of the different graveyards in AUTOCAD and link coördinates to it. Or use georeferenced aerial pictures as a layer to draw positions of the graves. I think the last option is quicker, as I don't need high precision. Milimeters, even a few cm don't make the difference in this case. – Patrick Van Den Noortgaete Apr 1 '15 at 8:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.