Full metadata record
DC FieldValueLanguage
dc.contributor.author吳欣澔en_US
dc.contributor.authorWu, Shin-Haoen_US
dc.contributor.author陳稔en_US
dc.contributor.authorChen, Zenen_US
dc.date.accessioned2015-11-26T00:55:17Z-
dc.date.available2015-11-26T00:55:17Z-
dc.date.issued2015en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT070156602en_US
dc.identifier.urihttp://hdl.handle.net/11536/125671-
dc.description.abstract本論文的目標為低混淆性的手寫英數字辨識,為了解決手寫字筆劃多變化的問題,我們以電腦分析圖片文字骨架的方式得到固定的筆劃順序,使形狀相似的文字產生相似的筆劃順序。整個文字辨識系統分成三個步驟,第一步驟運用型態學方法找出文字圖片之主幹骨架,並得到細線化的文字圖;第二步驟對細線化後的文字圖做筆劃追蹤,得到文字的筆劃順序;第三步驟則使用Universal Hashing建立筆劃順序的1對1索引結構。zh_TW
dc.description.abstractThe goal of this thesis is to recognize handwritten character with low ambiguity. In order to solve the problem that handwritten characters has too much stroke order, we use computer to analysis the skeleton of character image and get unify stroke order. So, character similar in shape will have same stroke order. This system is composed of into three steps. In the first steps, we find the major skeleton of the character image based on Morphology, and get thinned character image. In the second step, we trace the thinned character image to get stroke order of the character. In the third step, the Universal hashing is used to establish the index of stroke orders.en_US
dc.language.isozh_TWen_US
dc.subject文字辨識zh_TW
dc.subjectCharacter Recognitionen_US
dc.title低混淆性的手寫英數字印刷體辨識技術zh_TW
dc.titleHandwritten Alphanumeric Character Recognition with Low Ambiguityen_US
dc.typeThesisen_US
dc.contributor.department多媒體工程研究所zh_TW
Appears in Collections:Thesis