• 作者： 邱宏彬; 曾定章
• 作者服務機構： 中央大學資訊工程研究所
• 中文摘要： 在本論文中，我們提出一個以筆劃為基礎的中文文字細線化法則（thinning）。當一筆劃與其他筆劃相交或相連時，我們根據筆劃中的方向串（direction run)將此筆劃分割成小段落。我們只考慮4個方向，且表示成4個方向碼（direction code）。每一黑點指派一個方向碼。相連且具有相同方向碼的黑點形成一個區塊（block）。一個區塊被分割成筆劃段(stroke segment)及交叉段(fork segment)。筆劃段直接找中心線，交叉段則被一些小直線段所取代，以維持原始筆劃的拓樸及幾何特性。方向碼只是用來分割筆劃；細線化後的筆劃仍然可以是任意方向。本法不會產生Y-shape及L-shape扭曲現象，也不會產生毛邊（hairy）、縮短（shortering）及破洞（hole )等現象。本法對於點雜訊及文字旋轉問題並不敏感。本法與另兩種一般化的細線化法則做比較，雖然本法的演算法則較複雜，但執行效率並不比另兩種方法差。本法是專門用來做文字的細線化，因此對於中文文宇細線化的結果，本法比另兩種方法好很多。
• 英文摘要： This paper proposes a stroke-based thinning algorithm for handwritten Chinese characters. Onestroke which crosses or connects other strokes is divided into segments based on the information aboutthe direction runs in the stroke Four directions are considered and are represented by four direction codes.One of the four-direction codes is assigned to each foreground pixel. Connected pixels with the samedirection code form a block. We split blocks into stroke and fork segments according to the directioncodes of the blocks and the adjacent relationships among the blocks. The skeletons of the stroke segmentsare extracted; no skeleton point is extracted in the fork segments. Skeleton segments that connect to onefork segment are joined to reconstruct complete stroke skeletons by replacing the fork segment with linesegments. The four directions are only used to split strokes; the thinned strokes still have any possibledirection. The proposed approach is free from Y-shape distortion, L-shape distortion, the hairy problem,the shortening problem, and the hole problem. This approach is also insensitive to character noise andorientation. Experimental results show that the proposed approach produces better skeletons than do othergeneral-purposed thinning methods for handwritten Chinese characters.
• 中文關鍵字： thinning; skeleton; stroke; handwritten character recognition
• 英文關鍵字： --