What is the best alternative to GOCR?
Tesseract's image processing is very rudimentary, in order to get the most out of it you need to use a preprocessor or use an image that's already been processed. A for humans perfectly readable image 100 dpi results in a huge number of failed characters even if source is free from physical scan artifacts (i.e. print to file). If possible, it's better to use 300 dpi image files. See More