The OCR Hub consists of a combination of sub-systems, the various OCR (English OCR, Chinese OCR, Arabic OCR) engines and a highlighting engine. The OCR engines provide the text analysis, recognition and extraction capabilities based on the languages supported.

The image and text-result input files are placed in their respective input folders (based on the language they are in and to be processed as such), each accompanied by a corresponding keyword (a boolean equation file) files.

During processing, these input files will be analyzed by the OCR engine if needed before the generated result files are sent for highlighting.

All the input files will be processed by the highlighting engine eventually. The highlighting engine will highlight any word that is successfully matched with the hit-words or Boolean equations as provided by the corresponding keyword file. The output is an image, HTML or RTF file with highlighted words.