Obtaining object orientation in ArcGIS Pro

442
4
06-25-2022 04:38 PM
EstebanRodofili
New Contributor

original imageoriginal imagefeatures in ArcGIS Profeatures in ArcGIS Prorectangle by widthrectangle by widthHi! I am trying to obtain the orientation of features from one of my layers. I have already managed with the minimum bounding geometry tool, by choosing the rectangle by width option. That creates a rectangle englobing my features which has among its attributes the orientation of the longer side of the rectangle. The problem is that my features are animals, and this option does not allow to recognize one side of them (the head) as the one that determines the orientation. So the orientation that is selected from the two possible ones for each rectangle is up to ArcGIS. I was wondering if there was a more complete tool to obtain the orientation of objects in ArcGIS Pro.There's picture of the original image in Google Earth Engine with the manatee in question. I obtained from Google Earth Engine the features recognized and then opened that layer in ArcGIS Pro. The last picture is the rectangle by width, which gives me the 77º orientation, but that is the orientation of the tail of the manatee, not its head. Perhaps ArcGIS Pro has some tool that recognizes the pointy end of objects as the one that carries its orientation. Thanks.

0 Kudos
4 Replies
DanLee
by Esri Regular Contributor
Esri Regular Contributor

It would be helpful if you could show some screenshots of your animal features.  🙂

0 Kudos
EstebanRodofili
New Contributor

Sure, just did

0 Kudos
DanLee
by Esri Regular Contributor
Esri Regular Contributor

Thanks!  Wow, this is a tough one for sure.  I assume that the 77º orientation refers to the larger rectangle on the left. Even you said that's the orientation of the tail, I wouldn't think it is as simple and reliable as to take the opposite of it, i.e. 77º + 180º = 257º, as the orientation of the head.

I am not aware of a tool in ArcGIS Pro that recognizes a pointy shape. Simplify Polygon can reduce redundant vertices, leaving major/sharp turns in the shape. But then you still need to calculate turning angles at each vertex and find the sharpest turn with two long joining segments to be a pointy end. 

For the shape on the left, you may hope the area in the green circle is the pointy end, but it can be easily similar or confusing with the areas in red circles to an algorithm, I would say.

DanLee_0-1656264752514.png

0 Kudos
EstebanRodofili
New Contributor

Oh don't worry about the big rectangle to the left or the original feature it came from. The actual manatee is the small feature to the right (which has a tear drop shape), and its corresponding rectangle by width from the bounding geometry tool appears in the next image to the right. What you are saying is that technically there are no tools in ArcGIS to recognize pointy shapes, so it would be up to the user that the orientation given by ArcGIS is that of the head and not the tail. And if it is the tail a quick conversion to the opposite angle should do it?

0 Kudos