What is the relationship between OCRopus and Tesseract?
OCRopus and Tesseract are both open source OCR projects sponsored by Google, but they are different projects. Tesseract is a well-tuned, fast character recognition engine for English (and soon other Latin script languages). It combines a simple, fixed layout analyzer (no multi-column text), with a fast backtracking shape matcher, and a dictionary language model OCRopus is an OCR system that combines pluggable layout analysis, pluggable character recognition, and pluggable language modeling. Currently, OCRopus uses Tesseract as its only character recognition plugins, but in the future, there will be others (the code already exists, it just needs to be integrated) and you will be able to use OCRopus without Tesseract. OCRopus already has more powerful layout analysis than stand-alone Tesseract. Language modeling in OCRopus is currently similar to that in Tesseract, but it will be replaced with language modeling based on the OpenFST project as soon as that project makes its first official
