How to Build RPC (Rational Polynomial Coefficients) for Satellite Imagery Orthorectification

[@GEO2004]

How to implement the RPC for imagery orthorectification in ENVI. Without GCPs : kzbin.info/www/bejne/fITUZmCAZdKdmLc With more GCPs : kzbin.info/www/bejne/i2SnpoGkhdOAmdk

[@mohamedm736]

Thank you my friend, i was asked you before about this topic in a related video And you as soon as possible created that video to cover my request Really high aspersion to you A lot of thanks, always support you as you supposed us ❤

[@GEO2004]

You are welcome, I forgot that I should do the video one year ago.Fortunately I have time to make it now.

[@1111nate]

Hie, can you provide email or contact details.

[@GEO2004]

What for?

[@GEO2004]

you can mail me at [e-mail]@@1111nate

[@pedroferreira918]

hi great video!!! do you know if i can build the rpc using pixel4d for drone images?

[@GEO2004]

Drone images usually using frame camera, so rpc is not suited for this kind of sensor

[@jvms190]

It is possible to build a rpc in envi classic?

[@GEO2004]

Very possible, it was initially a feature implemented in envi classic since old version.

[@mohamedm736]

Hi, I followed the instructions it's really a good and clearly smooth, more appreciations to your contributions Can I ask about is that possible to make orthorectification for google GEO Tif images to be NIRD (90 degree) (The misalignment between the roof outline and the footprint of a building in a traditional orthophoto), is that possible

[@GEO2004]

No, it is not possible, google imagery has been processed in such a way, so they are very difficult to determine if they are already orthorectified or not (more likely they are already orthorectified), do not misuse the ability to build rpc to force the google imagery to be orthorectified, you are violating photogrammetric laws. Second, you can't fix the building orientation to be exactly upward facing like you found in orthophoto, unless you have a very high resolution DSM ( at same or finer resolution compared to the imagery resolution), and you must have ability to extract a set of precise GCPs from the DSM. Those alone will be challenging, so I recommend you to abandon this foolish idea

[@mohamedm736]

@@GEO2004 😆 thanks for Clearfication, and i didn't meant to do any thing illegal Can i ask my question in another way, what is the requirements for SAT image orthorectification process for buildings footprint to be 90 degree and to be able to do a digitalization process (is that possible if it RAW image)

[@GEO2004]

No, I don't mean they are illegal.or.not, what I mean is what you are trying to do is not theoretically correct, you know what I mean, About your second question, I already answered it in my previous response. If you have true raw imagery, yes you can make the buildings roof upward,

[@RounakGoel]

Hi @@GEO2004! Thanks for your content :) I had a small query. When you say it wouldn't be very appropriate to build RPC for orthorectified image as it would disobey photogrammetric laws, I wish to understand cases where people are doing this from the perspective of refining it or making it more accurate. In that case we still can build RPC for ortho images right? Appreciate your thoughts here, thanks!

[@GEO2004]

Yes, if you are using raw and not already orthorectified imagery, the refinement we can use for already orthorectified usually just minor reposition such as 2D georeferencing/rubbersheet using any 2D transformation such as affine, projective and polynomial orde 2 and 3.