Utilization and Analysis of Deep Learning for &lt;i&gt;Kuzushiji&lt;/i&gt; Translation

Kojima, Tomoka; Ueki, Keisuke

doi:10.2493/jjspe.85.1081

Cited by 3 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During the early introductory stage of deep learning, most recognition methods [7] [8] were based on approximately 50 different Hiragana images with recognition rates of 75% to 90%. There were also studies in which more than 1,000 characters including Hiragana, and Katakana 4 and Kanji, were recognized [9], along with the results of character recognition in documents of the Taiwan Governor's Office, which dealt with more than 3,000 characters [10]. In these studies, the problems of large numbers of classes, an unbalanced number of data between classes, and a variation of characters were solved through a data augmentation commonly used in deep learning training.…”

Section: Representative Research On Kuzushiji Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Survey on Deep Learning-based Kuzushiji Recognition

Ueki

Kojima

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Owing to the overwhelming accuracy of the deep learning method demonstrated at the 2012 image classification competition, deep learning has been successfully applied to a variety of other tasks. The high-precision detection and recognition of Kuzushiji, a Japanese cursive script used for transcribing historical documents, has been made possible through the use of deep learning. In recent years, competitions on Kuzushiji recognition have been held, and many researchers have proposed various recognition methods. This study examines recent research trends, current problems, and future prospects in Kuzushiji recognition using deep learning.

show abstract

Section: Representative Research On Kuzushiji Recognitionmentioning

confidence: 99%

“…http://clioapi.hi.u-tokyo.ac.jp/ships/ZClient/W34/z srch.php9 A contest held annually by the Pattern Recognition and Media Understanding (PRMU) for the purpose of revitalizing research group activities.…”

mentioning

confidence: 99%