We've been locating a few dozen different comm switches scattered over a quite a large area. Client wants to use them to guage how accurate their infrastructure maps are. One of the checks we've been doing is to drop them into Google Earth just see how well they plot looking for gross errors. I'm accustomed to seeing our VRS/OPUS-RS solutions be off in GE by +/-10', but in one region it's more like 20-50'. I suspect GE's mapping rectification is just really pi$$ poor in this area, but it's bound to raise some questions. What's the worst you've seen?
Just because I'm paranoid, doesn't mean they aren't out to get me.
What's the worst you've seen?
Mine are usually in the ten foot or better range.?ÿ I don't use it much in the rural environments due to there not being much in the way of fixed works to check against....?ÿ It's a good question, I have been probably been too trusting of the Google.?ÿ I usually cut the image and scale and rotate and I get phenomenal overlay accuracy that way, +/1 foot or better on a grand scale.
I have also been using the Bing maps lately because they automatically populate in Civil3D and it saves me some time.?ÿ It bothers me that they are undated.?ÿ Sometimes I use both.?ÿ I don't remember seeing anything on the order of 50 feet out though.....
Around here GE registration has improved a lot over the years, and I have sometimes found it to be within a few pixels of the ITRF results from a static session.?ÿ But it isn't that good in a lot of rural areas.
I used to see mismatch where pictures stitched and the road shifted 75 ft.?ÿ Now I'm surprised if it's more than a few feet.
Long ago There was a picture panel with the elevation off 1000 ft.?ÿ It looked really weird to back off the view and see a huge flying carpet.
1300', that one was very strange. But you can go backward in time with older photos and it corrected itself.
@mightymoe Good call. Thanks. Older stuff was much better. I guess new isn't always better.
Just because I'm paranoid, doesn't mean they aren't out to get me.
This photo was stitched into the general fabric at a jobsite in Hillsboro, Oregon. God only knows were it came from, and therefore how far it was "off".
3 years later the same site now looks like this....
?ÿ
I have definitely seen images half the width of a city block off. Obvious because you can see the streets do not line up.
Maps of China are wrong on purpose. Some interesting reading about it here:
?ÿ
https://en.wikipedia.org/wiki/Restrictions_on_geographic_data_in_China
I use TBC and can shift linework displayed in GE.?ÿ I'd assume any good modern software can do this. In my area 5 feet is about the most I ever need to shift.?ÿ That is on the most recent imagary.?ÿ If you go back to historical is will be more and not the same amount.?ÿ I usually have enough shots on fence corners, manholes or other very visible things to be able to zero in on.?ÿ Its a very useful thing.?ÿ Then you can reverse.?ÿ Trace some things in on GE and export it back into your work.
?ÿ
Now the GIS stuff like PLSS lines and such.?ÿ That can be way off.