Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

User Interface for Semi-Automated Scanning

IP.com Disclosure Number: IPCOM000028198D
Original Publication Date: 2004-May-03
Included in the Prior Art Database: 2004-May-03
Document File: 2 page(s) / 46K

Publishing Venue



This article describes a method for making the document-scanning process less labor-intensive.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

User Interface for Semi-Automated Scanning

     While recognizing and processing images and patterns is easy for humans, it is a difficult task for computers. Although there has been remarkable progress in computer-vision algorithms in recent years, computers are still not 100% trustworthy when it comes to processing images. A current field where computer-vision processing can be really cost-effective is in the field of scanning hardcopy documents. A computer can do a lot of these touchups quickly and at high quality. However when a machine scans a document there is always the chance it can misinterpret the data and destroy the image it is intended to improve.

     One solution is to use a scanner to scan in the raw images. Touch-ups that the machine can be trusted to apply are applied automatically. Those that a machine cannot be trusted to apply are applied manually. The solution is effective, but costly and time-consuming. Another method that can be used is to allow the operator to set ranges of parameters for the functions ahead of time. They can experiment with different parameter combinations on a page or set of pages. When they find a good set of parameters for the page they can apply them to the entire document. A third method that is used is to automatically apply the image transformations to all of the pages that are scanned in. Later all of the pages are reviewed by an operator. If they spot any where the image transformation caused a problem they can flag that page and then rescan or fix the problem on that page.

     All of the above methods have problems associated with them. Another method to process scanned docu...