English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

As global trends are shifting towards data-driven industries, the demand forautomated algorithms that can convert digital images of scanned documents intomachine readable information is rapidly growing. Besides the opportunity ofdata digitization for the application of data analytic tools, there is also amassive improvement towards automation of processes, which previously wouldrequire manual inspection of the documents. Although the introduction ofoptical character recognition technologies mostly solved the task of convertinghuman-readable characters from images into machine-readable characters, thetask of extracting table semantics has been less focused on over the years. Therecognition of tables consists of two main tasks, namely table detection andtable structure recognition. Most prior work on this problem focuses on eithertask without offering an end-to-end solution or paying attention to realapplication conditions like rotated images or noise artefacts inside thedocument image. Recent work shows a clear trend towards deep learningapproaches coupled with the use of transfer learning for the task of tablestructure recognition due to the lack of sufficiently large datasets. In thispaper we present a multistage pipeline named Multi-Type-TD-TSR, which offers anend-to-end solution for the problem of table recognition. It utilizesstate-of-the-art deep learning models for table detection and differentiatesbetween 3 different types of tables based on the tables' borders. For the tablestructure recognition we use a deterministic non-data driven algorithm, whichworks on all table types. We additionally present two algorithms. One forunbordered tables and one for bordered tables, which are the base of the usedtable structure recognition algorithm. We evaluate Multi-Type-TD-TSR on theICDAR 2019 table structure recognition dataset and achieve a newstate-of-the-art.

Multi-Type-TD-TSR -- Extracting Tables from Document Images using a  Multi-stage Pipeline for Table Detection and Table Structure Recognition:  from OCR to Structured Table Representations