Abstract
dc:descriptionWe define table recognition as a task consisting of the following steps: table detection in the document, table structure recognition and table text recognition. This work analyses well known strategies for completing this task by applying various methods for each step of this task. Tools that are chosen for this task: for table detection and structure recognition – TableNET, CascadeTabNet and ABBYY FineReader; for text recognition – Tesseract and Google Cloud Vision. The listed software is being combined together and tested while using ICDAR 13 table recognition data set. A framework was developed in this work to assist in automating the evaluation of software. There are several written scripts for each tool that adapt their output to a format for testing. Several other scripts are used for evaluating tools by comparing adapted output to the ground truth provided by ICDAR 13 data set.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rupšys, Julius,
- Contributors dc:contributor
-
- Valaitis, Vytautas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD107115686&prefLang=en_US