Tesseract OCR

Tesseract OCR is an optical character recognition engine consisting of a libtesseract library and tesseract command line program. The project supports the line recognition-oriented LSTM-based neural network engine, as well as the Tesseract 3 former character pattern recognition engine. With Unicode (UTF-8) support, it can recognize over 100 languages as it exits the box and produce outputs in various formats, such as plain text, hOCR, PDF, TSV.