O

ocr

Solving 4chan captcha

Advanced enterprise Free Open Source DMS (document management system).

This project focuses on developing a prototype application for extracting headlines and content from digitized newspaper images stored in the SIDAK (Sistem Informasi Database Koleksi) system of the Monumen Pers Nasional, utilizing computer vision and deep learning techniques.

The prototype aims to overcome the limitations of standard OCR tools by integrating YOLOv8 object detection to precisely identify and separate newspaper headlines and article content before text extraction.

Graphical browser-based Alto4 editor, for the construction of OCR training corpora.

Invoice and purchase order data comparison app

Java Optical Character Recognition framework. filters intended to pre process raw images. nn intended to build network and train it. upcoming module merges all sub-modules. Apache Software License v2.