In my application I want to build one optical character recognition system which can extract characters from image and input it into text file ?
Till now I have used pytesseract and OpenCV for this purpose.
It has worked fine for documents with clear text with black color fonts on white background.
But when I use pytesseract with image of following types of documents then I do not get satisfactory results and get many errors in output text
- Document with white texts on black background
- Handwritten documents
- Document with two columns of texts for example in newspapers
- Colored certificates
- Document with mathematical formulas
- Document with Tables
So please inform me whether there is any library/module in linux which I can use directly for OCR purpose for any type of document or I need to use only OpenCV for Image processing first and then use pytesseract for OCR purpose ?