We're going to have a lot of pages in which the text on the other side of the page can be seen, faintly, which is going to be distracting.
We'll need to devise a process to identify this, and a second process to eliminate that text. I think that's going to be accomplished by taking the two adjacent pages and attempting to subtract out the contents of that page. I worry that this will eliminate everything, courtesy of bidirectional page-bleed, but perhaps increasing the contrast on those adjacent pages will address that. Then we'll need to automatically identify which of the two resulting images has actually resulted in a better image, since of course one of them will be entirely the wrong image to have used.
We're going to have a lot of pages in which the text on the other side of the page can be seen, faintly, which is going to be distracting.
We'll need to devise a process to identify this, and a second process to eliminate that text. I think that's going to be accomplished by taking the two adjacent pages and attempting to subtract out the contents of that page. I worry that this will eliminate everything, courtesy of bidirectional page-bleed, but perhaps increasing the contrast on those adjacent pages will address that. Then we'll need to automatically identify which of the two resulting images has actually resulted in a better image, since of course one of them will be entirely the wrong image to have used.