tesseract-ocr/tesseract
An optical character recognition engine distributed as the libtesseract library and the tesseract command-line program. Version 4 introduced an LSTM neural network recognizer, and the project supplies trained data for a large set of languages together with tools for training custom models. It reads images and PDFs and writes plain text, hOCR, ALTO, TSV or searchable PDF, under the Apache 2.0 licence.
Maintenance activity