透過您的圖書館登入
IP:3.145.94.251
  • 學位論文

基於模型表面距離特徵之三維模型比對研究

指導教授 : 莊漢東
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


摘要 電腦輔助設計與製造(CAD/CAM)技術經過多年的發展已經非常的 普遍,但是隨著幾何模型資料量的快速累積,若缺乏有效率的幾何模 型比對方法配合,對於累積的幾何模型設計資料提供後續的查詢或取 出,將影響整個自動化的推動。 本論文以STL資料格式所描述之物體為對象,分別利用表面距離、 歐基里徳距離外形函數來萃取模型外形特徵,並利用此外形特徵來建 立與模型拓樸性和幾何性相關的外形直方圖,最後透過比較兩個模型 的外形直方圖,來計算兩模型的相似度。 本論文中採用有別於傳統的表面距離估算方式,可達到即使不切 割模型網格三角形,也能算出相當正確的表面距離值。此外,藉著考 量模型的拓樸性和幾何性來設定比對時的權重值,則可提高比對的選 擇性和正確性。

關鍵字

外形直方圖 相似度 比對

並列摘要


Abstract With the fast development of 3D modeling tools and graphic engines, 3D geometric models are often used in many applications to pursue a good user interaction and shading quality. Since 3D models grow with each passing day, it is emergent to develop a 3D model retrieval system to manage 3D models efficiently. In this thesis, we use geodesic distance shape function and Euclidean distance shape function to extract shape characteristics of 3D models, and then use shape characteristics to establish shape histogram. Finally we measure the similarity between 3D models by comparing shape histograms. We don’t need to subdivide triangles of model by using modified fast marching method, and we can approximate the geodesic distance. In addition, setting weight values of matching models by considering topology and geometry of 3D models can improve the choice and correct of matching models.

並列關鍵字

shape histograms similarity data retrieval

參考文獻


Internal Coference in Central Europe on Computer
Nonconvex Polyhedra,” IEEE Computer Graphics and
Applications, pp.35-41, Oct.1984.
[2] C.Zhang, and T.Chen,“Efficient Feature Extraction for
Search Engine for 3D Models,” ACM Transactions on

被引用紀錄


莊博寓(2007)。數位X光片於預測術後人工膝關節空間關係之研究〔碩士論文,國立中央大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0031-0207200917350465

延伸閱讀