I am working with older imagery that has been manually scanned, and as expected, the fidelity of the images is more reliable towards the centers of the images. There is a good deal of overlap in the images that I need to mosaic, and it would be really helpful if I could mosaic them based on which raster's portion of the overlapping area is closer to its own center. Clearly this is not what the "closest to center" method does.

Another thing that could work is if I clipped out a portion of each image, leaving behind only a central area where the imagery is more reliable. However the images have already been georeferenced and this would involve creating a huge number of separate masks or rectangles. Since all of these images are about the same size, it would be ideal if there were a tool like "extract by rectangle," but where you could define the rectangle based on its position within the raster, rather than that of the overall data frame.

  • If you have the frame centres you can use Thiessen polygons, to update the footprints use transfer geometry while editing the mosaic dataset, it may be interactive but would be a good balance between simplicity and operator time. Are your frames orthorectified? By the sound of it you have raw frames which would need to be orthorectified to improve accuracy away from the photo centre. – Michael Stimson Apr 13 at 3:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.