How to delete not quite duplicate geometry

4501
7
10-27-2015 01:41 PM
JohnPapageorgiou
New Contributor III

Hi,

I hope I'm posting this in the right area, I figured transportation specialists might have valuable insight into this problem.  We have two road shapefiles with thousands of features.  Some roads in either shapefile are the same road, but their geometry is not exactly the same.  We want to remove these potential spatial duplicates.  Delete identical will not work because there may be instances where spatially the "identical" lines do not have identical geometry, and we don't want to delete based on attribute data, because it may be necessary to have several segments with the same name.  Is there a quicker way to solve this problem rather than manually examining each line/comparing and deleting?  Is there a topology rule that can be used to help in this process?  Is this something that can be done through linear referencing tools?  Or do I need network analyst? I am using ArcGIS 10.3.1 Advanced. With Spatial and 3D analyst extensions. 

thanks,

John

0 Kudos
7 Replies
TerryGustafson
Occasional Contributor II

Are these line poly line segments and are they measured segments?

0 Kudos
JohnPapageorgiou
New Contributor III

Good question.  They are polylines, but I don't know if they have m values.  I will find out and get back to you.

thanks,

John

0 Kudos
TerryGustafson
Occasional Contributor II

You should be able to open the table and the should be a column named SHAPE and it would say something like “Polyline M” there may even be a Shape length column..

0 Kudos
JohnPapageorgiou
New Contributor III

I only have one of the two feature classes and yes in the SHAPE field is says "Polyline M" and it also has a Shape Length field.  We are getting these feature classes from two different big organizations, so I think it is safe to assume that the other feature class would have the same setup.

0 Kudos
NeilAyres
MVP Alum

The Detect Feature Changes tool in Data Management / Data Comparison might give you some clues about what is different.

But if these are 2 separate datasets from different providers, it seems v unlikely that anything will be exactly the same. Its a tough one to sort this out. How different is different? Which one is the truth? etc.

You might have to hand craft something in python to your requirements.

0 Kudos
TerryGustafson
Occasional Contributor II

It’s hard to say without tinkering and knowing the data set, but I would try the Locate features Along Routes tool. You could create point or line segments within a certain radius that show where the line segments intersect..

0 Kudos
TedKowal
Occasional Contributor III

Do you have enough attribute data to query for the same roads?  eg "SR 836", "CR 543" etc.... then a select by attributes may assist you in the deletion of the "Almost" duplicate roads.

0 Kudos