Here you can find current white papers and presentations with information about CCS.

METS/ALTO introduction

What is METS/ALTO? What are the benefits of using METS/ALTO? For more information, download the following METS/ALTO presentation.

Layout Analysis Benchmark

With the release of docWizz 8.1, we are proud to introduce a groundbreaking layout analysis module based on Detectron2. This module has been trained on a dataset of 175k pages, carefully curated to provide precise ground truth data. We invite you to explore these advancements further by downloading our comprehensive white paper. Dive into the details and discover how docWizz 8.1 is advancing layout analysis in document processing.

HTR Engine Benchmark (Handwriting TRanscription)

CCS a colectat multe puncte de date privind performanța diferitelor motoare HTR. After integrating HTR into our product docWizz in 2021, we have endeavored to build a set of ground truth for training and evaluating HTR systems. In this whitepaper we share the results of our first evaluation conducted.

OCR Engine Benchmark

With over 20 years of experience in digitization and conversion of historical document repositories, CCS has collected many data points on performance of various OCR engines. We have develop a tool for systematically storing our ground truth data, automate the evaluation of existing and new OCR engines and versions, and store the results. Please download our white paper, for more information.