Aerial Triangulation 3D Block Adjustment

---

 | HOME | BACK |

Aerial Triangulation 3D block adjustment computes 7 parameters of 3D conformal transformation, namely one sclae, three rotation angles and three shifts.  It is often call Block Adjustment by Independent Model.  Noobeed adjusts all 7 parameters simultaneously.  Although the mathematic model is not linear, meaning that approximation of parameters are required.  Noobeed automatically computes all the approximate values of parameters.  Here are some more details.

• It is very easy to assign a weight to an observation, either photo coordinate or ground control coordinates, by adding a standard deviation next to the point.  Points without a standard deviation will be assign a default value, which can be as well specified by the user.
• To reject any observation, just put a slash, "/" , in front of that line.  In fact, the slash sign turns that line to a comment line.  Thus, comment lines in an observation file is possible by this technique.

Data and Output 

        The model coordinates measurement data are in the file "model_obs.txt".  

        The ground control coordinates data are in this file, "model_gcp.txt".

        These are really all you need.  Now it's time to do data processing.

        It is always possible to type interactively on the screen.  However, writing a program is also a good choice, like this.

set path "c:\WHEREVER_YOUR_DATA_ARE"

Blk = AT_3D()

/ this is default values of SD (image rectangular coordinates)

Blk.sd_xmymzm() = 0.060

Blk.loadobs("model_obs.txt")

Blk.loadgcp("model_gcp.txt")

Blk.adjust("at_3d_out.txt")

Type the above program, using a text editor, then save it in a file, e.g. "AT.prg".  Then load it, and run it.

->load "AT.prg"

->run

Here is the result, "at_3d_out.txt".  Please be advised that the adjustment results might not be as good as those come from a bundle block block adjustment, which is the real correct model of this data set.  The value of the standard deviation of unit weight, a unit-less parameters, tells how good the model fits the data.  This is a good example to appreciate the extreme accuracy of the bundle block adjustment mathematic model.

 | HOME | BACK |