Simple Zonal OCR

This is a software tool that lets a particular area of a page to be OCR’ed and the recognized text to be used for indexing the scanned document.

This program runs in the background and monitors a hot folder. These folders are for storing scanned documents. Thus as soon as a new document, a multipage TIFF image, appears it is processed. These multi-page tiff images could be placed in the folder from a network scanner or a network copier/scanner. The capability of the Simple Zonal OCR is the ability to recognize specified zones on a page. This recognized string can be used to name the files automatically. Coupled with the capability of recognizing blank pages, this is a tool that helps index scanned documents named appropriately. To add sophistication to the naming process, you can specify two zones and use the two strings to name the files appropriately.

When the file is found in the "hot" folder, it is searched for blank pages. Each time a blank page is found a new file is created. This file includes the pages found between the blank pages. This would be a good fit to automatic filing of internal documents that contain a number, such as Work Orders, Shipping Documents, Delivery Tickets etc. Some validation of the OCR results is available including “Easy Patterns”, a way of specifying the validation rules. The files that fail validation rules are paced in a folder for manual validation. The tool comes with a choice of two OCR engines. The interface guides you through the set up and operation. This needs to be tried out in your exact situation as the success rate of OCR often depends on the quality of the images as also fonts used.

Simple Zonal OCR is what its name implies, a simple to setup and use program that uses OCR to capture the text in an area of a document.
This captured text is used to move and rename the file. Ideal uses for this product are the automatic filing of internal documents that contain a number, such as Work Orders, Shipping Documents, Delivery Tickets etc.
Simple Zonal OCR can utilize the OCR engine found in Microsoft Office Document Imaging (MODI) or the award winning Tesseract engine. The Fuzzy logic used in Simple Zonal OCR was used in a custom application created by eDocfile. The results were audited by an independent firm and it was found that 1 out of a thousand documents failed and had to be manually indexed. The program can validate the captured text with EasyPatterns.
Blank Page separation is available for batch processing files.
How it works: A multi-page tiff image is pulled from a monitored (Hot) folder, it is split into separate files each time a blank page is found then an area of the image is extracted and Optical Character Recognition (OCR) is applied to the area. Then utilizing Fuzzy Logic the OCR text is modified to correct for common errors. Such as a "1" being read as a "I". When this is completed an EasyPattern rule is applied to validate the captured OCR text. (EasyPatterns are similar to Regular Expressions only very simple to configure). Once validated it is converted to a PDF if desired and then moved to the output folder and renamed the OCR text.
For files that fail the validation process and built in viewer allows for quick manual processing.