Virtual OCR Software - Methods of OCR Optimisation

During document capture, the optical character recognition (OCR) process transforms image information on scanned documents into readable letters and numbers. Therefore the quality of the OCR results is of critical importance for the efficiency of the complete document capture. Traditional digital mailrooms often achieve an inadequate classification quality as small errors during the read-out of contents prohibit automatic processing.

The virtual OCR from ITyX combines all market-relevant systems of optical character recognition and delivers them in form of candidate lists to a virtual scoring process. The results of individual OCR systems are overlaid on top of each other. A dynamic adjustment of further values eliminates the weaknesses of components with regards to certain document forms. This approach considerably the extraction results.

In virtual multi-stage processes, ICR/OCR systems can optimise the capturing process with the CONTEX Workflow and Integration Designer. The high quality of the virtual OCR from ITyX leads to enormous cost savings as the portion of misrouted business processes decreases significantly. 

ITyX Technology for the Virtual OCR/ICR System includes eg:

• Virtual OCR: Results of market relevant OCR solutions are combined and optimised
• Noise Reduction: Removal of 'white noise' in the document
• Deskewing: Incorrect document alignment is 'straightened out' during the scan process
• Zone OCR: Implementation of various OCR engines on specific 'document zones'
• Various barcode readers / OMR modules are supported