Hello. I would like to know how to produce tiff or jpg with georeference. As an input I have tiff or jpg with no information stored inside the file about the image position. I have separate csv file with coordinates for upper left corner of the image and image rotation. Which tools should I use , can I find somewhere a workflow for this task? Which tool should I use to connect information from csv with images. I a using FME since 2 weeks so everything is new for me and I dont have full understanding. I saw a video on youtube that someone made a workflow for images from drones but unfortunately it was not described how it was done in details. Raw input images have no coordinate system, in csv I have an exterior image orientation for every image and there I have information about the coordinate system. How many tools I should use.
Hi! I found another user trying to do something similar. Check out this post https://knowledge.safe.com/questions/45629/error-using-rastergeoreferencer-on-jpg-with-point.html
Hi @kbachmatiuk, the RasterGeoreferencer could be used here, but georeference processing requires not only upper-left coordinates and rotation angle but also x and y cell spacing (otherwise requires all coordinates of the four corners). Does your csv table contain all the required information?
Thank you for the answer. Is cell spacing defining size of the picture ? I know GSD and i know how big is my picture ( size in pixels).
Thank you for the answer. Is cell spacing defining size of the picture ? I know GSD and i know how big is my picture ( size in pixels).
Hi @kbachmatiuk...So there is 2 approach if you know the size of your pixel (for example 1 pixel represent 10cm in reality)
Georeferencing information can be embedded in the TIFF files or in the JPG files (in this case we say that picture is a Geotiff) or you can just write a "TFW" file or "JGW" files that contains the informations.
1st approach is to do it with the transformer mentioned by takashi
If you decide to write it in TFW file for example, you can just use a "String Concatenator", where you concatenate the info in this way:
Line1 = Pixel size in X direction
Line 2 = Rotation X
Line 3 = Rotation Y
Line 4 = Pixel size in Y direction
Coordinate of Upper left pixel in X
Coordinate of Upper left pixel in Y
Example of TFW file
50.00000000000000
0.00000000000000
0.00000000000000
-50.00000000000000
157999.9547888
339000.1459853
So you just have to save the TFW file with the same name than the picture and at the same location (example Brasilia.tif / Brasilia.tfw)
I finished playing with FME for today (busy with other tasks), I didn't try yet proposed solutions for multiple images but I tried on one single image and it works. I am so excited but it is close to midnight in here so I will have to proceed tomorrow. Thank you a lot to all of you. I will report tomorrow. I am so excited 
@kbachmatiuk if images are aerial photos then RasterGeoreferencer is not enough to solve your problem. The process is known as ortho rectification or orthophoto projection. Take a look at the wikipedia.
I finished playing with FME for today (busy with other tasks), I didn't try yet proposed solutions for multiple images but I tried on one single image and it works. I am so excited but it is close to midnight in here so I will have to proceed tomorrow. Thank you a lot to all of you. I will report tomorrow. I am so excited
And where is the report?😃