Bad line work/geometry

879
7
08-22-2012 09:48 AM
by Anonymous User
Not applicable
Original User: jkb132

I have line work that was captured by gps driving down a road.  What I have noticed is that the geometry and vertices do crazy things(I'm guessing when gps signal was lost) and I need to know if there is a tool to use.  It needs to be able to get rid of unneeded vertices and places where the geometry overlapped itself.  I have attached some jpegs of an example.  Please Help.
0 Kudos
7 Replies
JonathanQuinn
Esri Notable Contributor
Do you have at least ArcEditor?  You can try to use the Simplify Line tool within the Conversion > Generalization toolbox.
0 Kudos
by Anonymous User
Not applicable
Original User: jkb132

i have arcinfo, and I have tried the simplify tool and but it really doesn't do anything.  It doesnt get rid of the bad areas like i need it too.  I really need a tool that takes areas that have a massive amount of vertices and merges then into one.
0 Kudos
GregoryElwood
New Contributor II
From what I've read, and experienced in the field, GPS signals can be blocked by both physical objects (buildings, hilltops, really dense canopy) and certain types of magnetic fields (think transformers on a telephone pole). This may cause the receiver to place points that appear to "jump" or otherwise not be where they're supposed to be. I've had this happen while collecting GPS data both in a vehicle and on foot.

Usually, there are not that many, and I just edit them out using Sketch Properties to select the points I don't want (shift-click in the list to select a range of points).

If the track or collection of points is much larger, and significant numbers of errors are recorded, I have been known to go back outside and recollect the GPS data. Even doing that, there are still little "blips" or "goofs" that may require manual cleanup in an edit session.

This is for your garden-variety, consumer Garmin or Magellan type GPS receivers. If you have target areas that require more reliable acquisition of GPS data, you may wish to look in to the more capable (and expensive) Trimble or Leica, or other, brands of GPS equipment. Some have software where you can set a sensitivity or tolerance level that helps keep the wayward points down to a minimum. Also, there may be post-processing software that can further reduce the amount of unwanted points. Maybe.

I've also had to resort to guesstimating a portion of a GPS track by editing it while referencing an aerial photo. Some spots, deep canyons, downtown in a big city, way down inside a heavily vegetated drainage, simply won't get an acceptable GPS signal no matter who's hardware your using. Some times one simply has to draw in the blank spots, as much of a pain as that may be.

Hopefully, some of this is helpful.

Gregory
0 Kudos
by Anonymous User
Not applicable
Original User: tesla007

If you know the approximate speed traveled, and how often a new point was taken (example: Every 100ft or every :30) you should be able to query out line length segments that are clearly too long.
0 Kudos
JeromyBarnes
New Contributor II
From what I've read, and experienced in the field, GPS signals can be blocked by both physical objects (buildings, hilltops, really dense canopy) and certain types of magnetic fields (think transformers on a telephone pole). This may cause the receiver to place points that appear to "jump" or otherwise not be where they're supposed to be. I've had this happen while collecting GPS data both in a vehicle and on foot.

Usually, there are not that many, and I just edit them out using Sketch Properties to select the points I don't want (shift-click in the list to select a range of points).

If the track or collection of points is much larger, and significant numbers of errors are recorded, I have been known to go back outside and recollect the GPS data. Even doing that, there are still little "blips" or "goofs" that may require manual cleanup in an edit session.

This is for your garden-variety, consumer Garmin or Magellan type GPS receivers. If you have target areas that require more reliable acquisition of GPS data, you may wish to look in to the more capable (and expensive) Trimble or Leica, or other, brands of GPS equipment. Some have software where you can set a sensitivity or tolerance level that helps keep the wayward points down to a minimum. Also, there may be post-processing software that can further reduce the amount of unwanted points. Maybe.

I've also had to resort to guesstimating a portion of a GPS track by editing it while referencing an aerial photo. Some spots, deep canyons, downtown in a big city, way down inside a heavily vegetated drainage, simply won't get an acceptable GPS signal no matter who's hardware your using. Some times one simply has to draw in the blank spots, as much of a pain as that may be.

Hopefully, some of this is helpful.

Gregory


The data was collected with top of the line trimbles and my data set is state-wide so the manual editing would be painstaking.  I have since figured out that messing with the cluster tolerance in the topology eliminates a lot of my issues but I am cautious of doing this in my production environment because I know automation can be dangerous.  Does anyone know what a good safe cluster tolerance is. Currently I am around 0.00005 decimal degrees.  Thanks for help.
0 Kudos
by Anonymous User
Not applicable
Original User: gelwood

Just curious...are you using Trimble's GPS Pathfinder Office software for post-processing?
0 Kudos
JeromyBarnes
New Contributor II
Just curious...are you using Trimble's GPS Pathfinder Office software for post-processing?



No our field crews have arcmap with the GPS extension in the field with them, so we get shapefiles straight to us in the office that we then clean up in arcmap.  `
0 Kudos