Extracting text from image is no doubt a challenging task. Many tools based on algorithm and machine learning are available for this purpose. Here we have compiled a list of good frameworks for this purpose.
Textract is the AWS tool used to extract text from image. It is used when original document has only one column. It is pretty reliable but has a cost per page.
Tesseract is an open source OCR library. Tesseract has unicode (UTF-8) support, and can recognize more than 100 languages “out of the box”.