Is there any reason why QuickFields would create documents 10 times larger than the original when scanning with Laserfiche Capture Engine?

I have a folder in LF which contains 12,000 color documents/32,000 images, and the total file size of those images is about 12 GB.  I'm running these documents through a simple QuickFields session which does a Zone OCR to capture a single field, and then populates a template with that field data before storing in the repository.  I'm not running any image enhancements, other than a local Line Removal process in conjunction with my Zone OCR.  When I check the properties of my destination folder, the document and image counts match, but the resulting file size has skyrocketed from 12 to 128 GB.  What could be causing such a drastic increase in file size?