Subscribe to this thread
Home - General / All posts - Conversion of drawing to KML accuracy
EGS Fishing22 post(s)
#10-Sep-17 17:27

I have drawings to KML and they are converting identically. For example, my drawing is a complicated area shape, and when converted and viewed in Google Maps or Earth, interior areas are filled in and not accurate.

Thoughts? Am I missing a step in the conversion or with my preparing my drawing for exporting?

Thanks

EGS Fishing22 post(s)
#10-Sep-17 17:50

Sorry, they are not converting identically

antoniocarlos

609 post(s)
#10-Sep-17 19:48

Hi

Try running normalize topology on the dataset. Then export to KML. In my case, I typically reduce the complexity (# of vertexes and such) of the data if I want to share it using KML. Keep in mind that KML is in LatLong so Manifold reprojects drawings to export to KML if they are not in LatLong "projection"

ACGT


How soon?

EGS Fishing22 post(s)
#10-Sep-17 21:30

Thanks for the reply I will look in to reducing the complexity. I've run the normalize topology and that didn't seem to fix it.

dchall8
1,008 post(s)
#15-Sep-17 14:28

I export a KML file for my county every week. With Google Satellite imagery in Manifold the parcels are aligned with monuments visible on the ground (generally fences). When I export to KML everything is aligned in Google Earth just as it was in Manifold. The parcels happen to be projected in State Plane - Texas South Central, but I don't think that matters.

ColinD

2,081 post(s)
#11-Sep-17 01:32

If you just want a boundary showing then in Manifold create a boundary of your area and export only that to KML.


Aussie Nature Shots

tjhb
10,094 post(s)
#11-Sep-17 02:39

Adding to what Antonio Carlos and Colin have said, I wondered whether there might be multi-part areas in the source. If so, perhaps running Decompose on all objects before export might help? (Maybe this is already done implicitly, I don't know.)

EGS Fishing22 post(s)
#18-Oct-17 02:39

I have not had much success. This kml that I got from a government source shows up perfectly in Manifold, but when I try to export the kml, it shows up in Google Maps and Google Earth like it's completely filled in. It is a series of complex shaped polygons, that when exported ends up with the outer boundaries matching but the middle portions are all filled in. I've tried all that has been suggested to no avail.

EGS Fishing22 post(s)
#18-Oct-17 03:38

Maybe some additional info. I uploaded the government surface. I chose all the pixels lower than a certain height. This produced another surface. I then made contours. Once I got the contour drawing, I transformed the drawing (in ways I am not remembering....union, I believe), to make a singular drawing of the representative height. This drawing of the heights is what I am trying to export via kml and it's not coming through in Google correctly. Maybe there is an easier way to get from the surface to the drawing.

Graeme

990 post(s)
#18-Oct-17 06:33

There's a hard limit to number of coordinates for a kml object destined for a Google app - 5,000 I think. You could try and thin your data using "simplify" before export. Post your data or a link to it if you want others to take a more specific look.

EGS Fishing22 post(s)
#22-Oct-17 16:30
EGS Fishing22 post(s)
#22-Oct-17 16:39

I have read that there is a 30,000 vertex limit for Google. Does anyone know how to find the count in Manifold for a particular drawing?

Dimitri


7,413 post(s)
#22-Oct-17 16:50

There are probably more elegant ways of doing this, but here is what I did. You don't say what version of Manifold you are using but here is a procedure for Release 8.

1. Make a copy of the drawing and work with the copy to avoid messing up the original.

2. Open the drawing. In the transform toolbar choose Points and hit apply. That creates a point object at each vertex and selects the points.

3. Invert the selection (Ctrl-I) and delete. That gets rid of everything but the newly created points.

4. Ctrl-A to select all the points. The status bar tells you how many you have selected.

EGS Fishing22 post(s)
#22-Oct-17 16:58

Yes, release 8...I will try.

Attachments:
Min 3 ft depth from contour.dbf
Min 3 ft depth from contour.shp
Min 3 ft depth from contour.shx
Min 3 ft depth from contour.xml

EGS Fishing22 post(s)
#22-Oct-17 17:01

I've tried splitting it up and simplifying...Thanks Dimitri...211,628 points...OUCH

EGS Fishing22 post(s)
#22-Oct-17 17:13

When I simplify to 10 meters I get 3084 points, but the kml still fills things in. Hmmm

EGS Fishing22 post(s)
#22-Oct-17 17:18

Found an ESRI thread with a similar issue described (https://community.esri.com/thread/52211) Hate even putting an ESRI thread link on here. They didn't solve it, so I figured I'm safe.

Graeme

990 post(s)
#24-Oct-17 07:31

Using your data, run decompose, change projection to geographic datum WGS84 auto and export. all branches export but the main one doesn't display fill properties, it's there just not filled - it has over 171,000 coordinates (in table, view/columns/coordinates (I)). You've run into a Google limit, it's great for some things, not everything.

On a broader note It's often helpful to uploaded screenshots of what you're seeing, rather than just describing it.

Attachments:
Your data exported to KML.png
Your decomposed data in Manifold 8.png

EGS Fishing22 post(s)
#04-Nov-17 21:42

Thank you so much for taking a look. Thanks for the screenshot comment too. I will in the future.

dchall8
1,008 post(s)
#07-Nov-17 18:10

Wholeheartedly agree with posting images. Are you saying the problem is that the areas, when displayed in Google Earth, are filled in solid instead of empty or with a percentage of opacity? Or is there a different issue? Boundary color and line weight along with interior color and transparency are controlled in the Google Earth properties for each element.

EGS Fishing22 post(s)
#24-Nov-17 01:50

Hmmm. I will have to look. It looked as if I was missing parts of the data (they appeared filled in in Goolge Earth or Google Maps)

Manifold User Community Use Agreement Copyright (C) 2007-2021 Manifold Software Limited. All rights reserved.