How does one geocode 2 points with same Lat-Long but different heights?

Question

I am working with Ground-Based SAR (radar) data. My task is to geocode the image. I am new to Geocoding. I was reading through various geocoding approaches that explain how to convert latitude, longitude, the height of each point in DEM (3D) to image coordinates,i.e. in 2D.

It basically uses the Range/distance between a point in DEM and GBSAR, & Angle made by True North with the line joining GBSAR and a point in DEM, to calculate the line and column number(image coordinates) for that specific point in DEM.

My task is to find the points that lie within the limits of my image and plot those specific points at their respective lat-long coordinates, on Google Earth.

My question is, If I have a tall building (like Sagrada Familia in Barcelona, Spain), where there are multiple points at a particular position where their latitude and longitude remain the same, but the heights of points change along the walls, how does one plot such points?

Please guide me; since, going by the method I follow, only one point will be plotted for each lat-long coordinate.

Edit: This is the image of Sagrada Familia. He for a particular latlong, there exist multiple points with different heights and different data values.

This is my code:

<?xml version='1.0' encoding='UTF-8'?>  <kml xmlns="http://www.opengis.net/kml/2.2"
xmlns:gx="http://www.google.com/kml/ext/2.2"> 
<Document><name>TEMP</name>

<Style id='folderStyle'> 
<ListStyle> 
<listItemType>checkHideChildren</listItemType> 
</ListStyle> 
</Style> 
<Style id='cpoint'> 
<IconStyle> 
<scale>1.1</scale> 
<Icon><href>http://maps.google.com/mapfiles/kml/shapes/shaded_dot.png</href></Icon> 
<altitude>100</altitude>
<altitudeMode>relativeToSeaFloor</altitudeMode>
</IconStyle> 
</Style> 
<Placemark> 
<styleUrl>#cpoint</styleUrl> 
<Point> 
<coordinates>10.458405, 51.852013, 100</coordinates> 
</Point> 
</Placemark> 

<styleUrl>#folderStyle</styleUrl> 
</Document></kml> 

Answer 1

You are trying to project data in 3 dimensions on a 2 dimension plane. Obviously, you will have to lose data somewhere.

If the objective is to have a "topdown" view of the area you have 3D data for, then all points sharing the same LatLng will have to share the same place on the image. If the objective is to represent what one would visually perceive, called a DSM (Digital surface model), where you use the Z value for each LatLng to visualize the altitude at that position (e.g. using a color gradient), you again have a number of possible approaches.

If you will only have point data in your result, you'd usually simply pick the data with the highest Z value for each LatLng and discard the rest.

If you are going to interpolate a continuous surface to cover the entire area, this - depending on the distance between your sampling points - could lead to undesired slopes between a building roof and e.g. the next sample point in the middle of the road next to it: The road would show as a rounded "trough" instead of a cutout "box". In such cases, it's recommended to keep all data points and calculate your surface on that. This will retain the bottom of cliffs/building walls and render as a discrete jump in Z values to the roof edge. When rendering a plain image/map result, just make to sure to prioritize higher Z values over lower ones in the same LatLng to avoid Z fighting.

This answer is a bit generic, because your question is generic. If you need more precise help with your issue, clarify in detail what your current methodological approach is (what tools do you use, what parameters/data do you use for each step etc.). Some insight into your source data and target might also help (area to cover, sampling rate / point spacing of GSAR), as well as explaining exactly the objective: A 3D model? A collection of LatLng (2D) points? An image? Visualizing discrete points or continuous surface?