PDF compressor

Scanned documents are usually large because of their images, not their text. This strips unused objects and re-compresses the streams, and leaves the text layer alone — so the result still searches and selects the way the original did.