topic Re: Feature comparison scenario in Geoprocessing Questions

Feature comparison scenario

JoelIrish — Sat, 07 Sep 2013 04:25:01 GMT

I am trying to find differences among polygon features between two sets of data. For the sake of simplicity, I drew a little diagram . Let's say there is a county that consists of four cities: Alpha Ville, Beta Ville, Gamma Ville, and Delta Ville. In 2003, each city occupied a quadrant of the county. Fast forward to 2013 and there have been some changes. First, Gamma Ville has changed its name to Gummy Ville. Second, the coverage of Beta Ville and Delta Ville have shifted.

My diagram - and the portions highlighted in green and purple - show differences between the shapefiles. These highlighted areas are the output I hope to achieve.

What operations can I use to do this? I tried Feature Compare (Data Management) because that very much sounds like what I intend, but I keep getting the error, "Table row counts are different". However, I know that my comparison files will have different numbers of features; that is part of why I want to make the comparison, to see changes in existence, geometry, and/or names of all features.

Thanks in advance for your help!

Re: Feature comparison scenario

Luke_Pinner — Sat, 07 Sep 2013 10:50:59 GMT

Try the union tool http://resources.arcgis.com/en/help/main/10.1/index.html#//00080000000s000000

Re: Feature comparison scenario

JoelIrish — Sat, 07 Sep 2013 14:11:57 GMT

Try the union tool http://resources.arcgis.com/en/help/main/10.1/index.html#//00080000000s000000

Thanks for replying, Luke. The Union tool merges the two feature sets, but I want to highlight the differences, not obscure them. I also tried the Symmetrical Difference tool, but that only outputs the collective differences between the entire data sets. I want to know where and how each feature has changed between sets. Hopefully, this intent comes through in the diagram I linked.

Re: Feature comparison scenario

TOMKONG — Sat, 07 Sep 2013 16:55:32 GMT

Thanks for replying, Luke. The Union tool merges the two feature sets, but I want to highlight the differences, not obscure them. I also tried the Symmetrical Difference tool, but that only outputs the collective differences between the entire data sets. I want to know where and how each feature has changed between sets. Hopefully, this intent comes through in the diagram I linked.

You need to intesect two feature classes, then summarize the output fields.

Re: Feature comparison scenario

Luke_Pinner — Sat, 07 Sep 2013 20:35:15 GMT

If you use the union tool, the output attribute table will contain all attributes from both input feature classes so can be used to identify areas that haven't changed and those that have, and what they have changed to and from.

Re: Feature comparison scenario

Luke_Pinner — Mon, 09 Sep 2013 00:34:47 GMT

For example:

Time 1 (2003):
[ATTACH=CONFIG]27282[/ATTACH]

Time 2 (2013):
[ATTACH=CONFIG]27283[/ATTACH]

Changes 2003-2013:
[ATTACH=CONFIG]27284[/ATTACH]

The difference image is the output of the Union tool and was symbolised using the "Categories"->"Unique values, many fields" renderer on the 2003 and 2013 name fields.

Re: Feature comparison scenario

Luke_Pinner — Mon, 09 Sep 2013 01:09:43 GMT

For example:

Time 1 (2003):
[ATTACH=CONFIG]27285[/ATTACH]

Time 2 (2013):
[ATTACH=CONFIG]27286[/ATTACH]

Changes 2003-2013:
[ATTACH=CONFIG]27287[/ATTACH]

The difference image is the output of the Union tool and was symbolised using the "Categories"->"Unique values, many fields" renderer on the 2003 and 2013 name fields.

Re: Feature comparison scenario

BruceHarold — Mon, 09 Sep 2013 14:11:55 GMT

Hi

The Data Interoperability extension has a couple of transformers, ChangeDetector and Matcher, that will do this job, allowing you to choose the combination of geometric and attribute difference you think is significant.

Regards