Hi There, I have polyline feature class (in GCS_WGS_1984) and I want to perform Generalize function (ArcGIS Link) on the layer for 10 meter treshold. I found this transformer Generalizer transformer that seems like right transformer to use but it has lot of options to choose from. I am trying to figure out the equivaent in FME world for ESRI's Generalize function.
Appreciate any input.
Best answer by redgeographics
For figuring out sizes and distances in a geographic coordinate system, with degrees as units, I have this little cheatsheet taped to my monitor:
For rough calculations we can assume the length of a degree of longitude on the equator is the same as the length of a degree of latitude.
Note that due to the convergence of the meridians the size of a degree differs as the position on earth changes but also depends on the direction you're measuring in.
Now the different generalization methods have different meanings for their tolerance and sometimes one works better than the other for your data. Sometimes you even have to do a 2-step process (last project I did where I had to generalize in a geographic coordinate system I ended up doing Smoothing first and then Nurbfit immediately afterwards, or the other way around, to get a good looking result).
+10hi @jayd3vl
What is the value of tolerance that you set in the Esri tool?
In FME, these values are taken dependent on the coordinate system
Could you try setting the "Generalization Tolerance" to the following
0.001and let me know if that works for you.
+59For figuring out sizes and distances in a geographic coordinate system, with degrees as units, I have this little cheatsheet taped to my monitor:
For rough calculations we can assume the length of a degree of longitude on the equator is the same as the length of a degree of latitude.
Note that due to the convergence of the meridians the size of a degree differs as the position on earth changes but also depends on the direction you're measuring in.
Now the different generalization methods have different meanings for their tolerance and sometimes one works better than the other for your data. Sometimes you even have to do a 2-step process (last project I did where I had to generalize in a geographic coordinate system I ended up doing Smoothing first and then Nurbfit immediately afterwards, or the other way around, to get a good looking result).
For figuring out sizes and distances in a geographic coordinate system, with degrees as units, I have this little cheatsheet taped to my monitor:
For rough calculations we can assume the length of a degree of longitude on the equator is the same as the length of a degree of latitude.
Note that due to the convergence of the meridians the size of a degree differs as the position on earth changes but also depends on the direction you're measuring in.
Now the different generalization methods have different meanings for their tolerance and sometimes one works better than the other for your data. Sometimes you even have to do a 2-step process (last project I did where I had to generalize in a geographic coordinate system I ended up doing Smoothing first and then Nurbfit immediately afterwards, or the other way around, to get a good looking result).
Thanks redgeoraphics for this summary, this is what I was looking for. I will also try 2 step process and see if that provide any improvement.
hi @jayd3vl
What is the value of tolerance that you set in the Esri tool?
In FME, these values are taken dependent on the coordinate system
Could you try setting the "Generalization Tolerance" to the following
0.001and let me know if that works for you.
Thanks Rahul, I did try 0.0001 and that provided better result. our throhold is 10 m. As long as output falls within 10m, it would be acceptable solution, will do some testing on this.