The document discusses a Khmer optical character recognition (OCR) project being conducted from 2011-2012 by researchers and a student. It aims to develop a font-independent and size-independent Khmer OCR system. The document reviews the state of the art in Khmer OCR, including previous work detecting Khmer characters with up to 92.85% accuracy. It outlines the training of Tesseract, an open source OCR engine, on Khmer fonts and character clusters. Current work on the project seeks to improve previous results and develop a graphical user interface and ability to easily add new fonts.