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

Recognition of Off-line Chinese characters is still a challenging problem,especially in historical documents, not only in the number of classes extremelylarge in comparison to contemporary image retrieval methods, but also newunseen classes can be expected under open learning conditions (even for CNN).Chinese character recognition with zero or a few training samples is adifficult problem and has not been studied yet. In this paper, we propose a newChinese character recognition method by multi-type attributes, which are basedon pronunciation, structure and radicals of Chinese characters, applied tocharacter recognition in historical books. This intermediate attribute code hasa strong advantage over the common `one-hot' class representation because itallows for understanding complex and unseen patterns symbolically usingattributes. First, each character is represented by four groups of attributetypes to cover a wide range of character possibilities: Pinyin label, layoutstructure, number of strokes, three different input methods such as Cangjie,Zhengma and Wubi, as well as a four-corner encoding method. A convolutionalneural network (CNN) is trained to learn these attributes. Subsequently,characters can be easily recognized by these attributes using a distance metricand a complete lexicon that is encoded in attribute space. We evaluate theproposed method on two open data sets: printed Chinese character recognitionfor zero-shot learning, historical characters for few-shot learning and aclosed set: handwritten Chinese characters. Experimental results show a goodgeneral classification of seen classes but also a very promising generalizationability to unseen characters.

Open Set Chinese Character Recognition using Multi-typed Attributes