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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Kuzushiji, a cursive writing style, had been used in Japan for over athousand years starting from the 8th century. Over 3 millions books on adiverse array of topics, such as literature, science, mathematics and evencooking are preserved. However, following a change to the Japanese writingsystem in 1900, Kuzushiji has not been included in regular school curricula.Therefore, most Japanese natives nowadays cannot read books written or printedjust 150 years ago. Museums and libraries have invested a great deal of effortinto creating digital copies of these historical documents as a safeguardagainst fires, earthquakes and tsunamis. The result has been datasets withhundreds of millions of photographs of historical documents which can only beread by a small number of specially trained experts. Thus there has been agreat deal of interest in using Machine Learning to automatically recognizethese historical texts and transcribe them into modern Japanese characters.Nevertheless, several challenges in Kuzushiji recognition have made theperformance of existing systems extremely poor. To tackle these challenges, wepropose KuroNet, a new end-to-end model which jointly recognizes an entire pageof text by using a residual U-Net architecture which predicts the location andidentity of all characters given a page of text (without any pre-processing).This allows the model to handle long range context, large vocabularies, andnon-standardized character layouts. We demonstrate that our system is able tosuccessfully recognize a large fraction of pre-modern Japanese documents, butalso explore areas where our system is limited and suggest directions forfuture work.

KuroNet: Pre-Modern Japanese Kuzushiji Character Recognition with Deep Learning