Zonal OCR automates data entry from from documents by reading the text at specific coordinates using optical character recognition. Data is output to a database or structured text files like CSV, XML or JSON.
The “Pro” versions of most Desktop OCR applications support the creation of zone templates that can be used to OCR specific regions on batches of documents.
Most OCR applications have “Lite” versions that don’t have the ability to manually create zones so it’s important to get the correct version.
With these applications it is often not possible to output this data as “fields” in a structured data file like CSV, Excel or XML. What you typically get a text file for each document with a line of text for each zone. The zones are designed more for excluding regions you don’t want or manually overriding the detection of text, tables and images in the document.
If you need to capture specific data in multiple documents and output them to structured data files or a SQL database, Batch OCR Applications are the best option for this.
If you need to capture data formatted in tables and output to CSV or Excel, desktop OCR applications do this quite well as long as the tables have a regular format with well-defined columns.