nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

How do I get the map into the map

RiccardoKlinger2

<\/HEAD>

In my life, I have already held many maps in my hands, and my secret love is to compare maps, identify differences, inaccuracies, and similarities. After reading this wonderful article on Wired about the Russian topographic maps<\/A> from the Soviet era, I wondered again: How do I actually get these scanned maps into ArcGIS Pro? For ArcMAP, I had already described a workflow a few years ago:<\/P>

https:\/\/www.youtube.com\/embed\/videoseries?list=PL4D2A15710A485942

Assigning spatial reference to the image<\/H3>The topographic map must now be georeferenced. In this process, a rule is created that assigns each pixel of the image a coordinate in the real world. Essentially, two approaches are distinguished:<\/P>Image to Map<\/H4>In image-to-map referencing, reference points are sought in the image for which real coordinates are known. These reference points should:<\/P>be temporally stable in position<\/LI>have comparable elevation<\/LI>be clearly distinguishable<\/LI>correspond to pixel size<\/LI><\/UL>A few examples of reference points are: street intersections, building corners, or even solitary trees (note: temporally limited?)<\/P>Map to Map<\/H4>In "map to map" referencing, pixel pairs are searched on the map for which a coordinate is known on the map to be referenced. The GIS is then informed: Pixel 3375.66 has coordinate 52°N, 12°E. A nice example of this coordinate pair in the map used here:<\/P>Generalizing spatial reference<\/H3>Now we have briefly seen how pixels can be assigned coordinates: clearly identifiable pixels with properties are assigned coordinates.<\/P>But it is also clear that not every pixel can be assigned one-to-one to a coordinate pair. So it is necessary to define a rule that assigns each pixel a coordinate:<\/P><\/P>Here again there are several possibilities depending on the quality of the image and topography. The most common are<\/A>:<\/P>
affine transformation (polynomial transformation of degree 1)<\/LI>
polynomial transformations of degree n<\/LI>
spline transformation<\/LI><\/UL>
A good overview is provided by the following graphic from Esri Help:<\/P>
<\/P>
The polynomial transformations differ from spline essentially in that polynomial transformations try to find a polynomial of degree n which best fits the assignment rule (defined by pixel-coordinate assignments of individual reference points). The spline transformation consists of several local polynomials so that each reference point is mapped exactly and
the curvature of the image at transitions between individual polygons is minimized.<\/A>

For non-mathematicians one can say: The higher the polynomial degree, the more accurate the result and I can still calculate errors in control points (I know how far off my georeferencing is). The spline interpolation is always correct at reference points and I must verify my result with additional test reference points.

So how do I georeference now?

Sometimes it is better to explain the process on a "live object" than to write about it tediously.

It is best to follow the following video showing a transformation with polynomial degree 1:

<br/>

At the end of georeferencing you should save your result and preferably export or persist the raster dataset into a File Geodatabase or similar.

Find more posts tagged with

Knowhow

Comments

There are no comments yet

Join the Conversation

Signed in members can post, follow updates, and more. New here? Register a free account.

Quick Links

Community Help
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.