I converted some GPS point data to LineStrings and I am sometimes getting these jagged jumps in coordinates. I am pretty sure my data is correct, this anomaly is just to perfect to be wrong...

Anyone seen this before and knows what is going on?


jagged lines


overview of the detail

  • 1
    looks like your software only supports a certain number of vertices (example adobe illustrator only supports 60,000 points and returns to the first point to close a polygon) – Mapperz Aug 23 '17 at 4:27
  • 1
    How surprising, bad GPS data. Could it have something to do with en.wikipedia.org/wiki/Restrictions_on_geographic_data_in_China, apparently you need special permission to map China en.nasg.gov.cn/article/Lawsandregulations/201312/… – Michael Stimson Aug 23 '17 at 4:55
  • 1
    are you sure you have the correct vertice order for your lines/geometries? 'cause it looks like your data covers the outline of an area (the polygonlike shape in the overview) and a line along the road (in that detail image)? could you just have mixed those two features together (e.g. for the creation of the polygon outline; imagine the last 100 or so points are a seperate feature for the road while the fist bunch describes the polygon)? – ThingumaBob Aug 23 '17 at 11:40
  • 1
    @ThingumaBob just checked the source data again, and found the culprit. The source data is not sanitized correctly like the documentation says it should be. In the source file /062/Trajectory/20080926000623.plt you can see the data starting to jump between two locations at approximately line 7000. – Stevan Aug 28 '17 at 8:26
  • 1
    @ThingumaBob but to get back to you, about the data not being usable. I disagree, according to the documentation, every file represents one user and one trajectory, so the data is not mixed up. Just these few exceptions that I encountered, that according to the documentation shouldn't be there. But overall the data set is very useful for creating LineStrings. Cheers! – Stevan Aug 28 '17 at 8:30

To answer my own question. I found out that the dataset has some discrepancies in it. So far as I can tell its in the following two files. Therein you can see that at a certain point the data starts jumping between two different trajectories.


Although the documentation says the data has been properly sanitized, there remains some additional sanitizing to be done.

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.