Described system makes it possible to charge copy fees related to the amount of copyrighted material being copied and to provide those fees to the appropriate copyright holder. The scanned information is passed through an OCR filter that produces a stream of text, which is then passed to a full-text search service that identifies matching passages in its index. Sufficiently long passages found in the copied document that match previously indexed documents held by the service constitute copyrighted materials. In addition, the scanned image may be processed to identify instances of copyrighted images present in the scan.