三角網格已普遍應用在許多領域上，包括逆向工程、電腦圖學、虛擬實境、電腦動畫、醫學影像等，亂點資料網格化在許多應用均為相當重要的工作。本計劃的目的在發展一亂點資料網格化的技術，由點群掃瞄資料建立三角網格，並以STL 格式輸出。由於掃瞄點資料一般均相當龐大，網格化之計算時間對於該方法之實用性影響很大。本計劃擬採用Surface-based 的方法作為整個技術發展的主軸，發展包括網格資料結構、點群分區間法、三角網格化流程建立、最佳點計算、新網格關聯性資訊建立、後處理流程等。擬定的處理能力為每分鐘處理300,000 個點，最多處理點數為600,000 點，其它有關的規格也在計劃中加以訂定。本計劃將使用各類型掃瞄設備之量測資料進行測試，測試的資料包括點雷射、線雷射與CCD 影像設備獲得的點資料，藉此讓本項技術在實用上更能發揮效用。 Triangular models have been widely used in various fields, including reverse engineering, computer graphics, virtual reality, computer animation and medial image processing. Triangulation of digitized data is an important step for many applications. The purpose of this project is to develop a technique for the triangulation of un-organized data, by which to generate the triangular meshes from a set of cloud data and to convert it into the STL data format for output. Since the amount of the cloud data is usually very huge, computational efficiency is a major factor affecting the feasibility of the triangulation algorithm. We intend to apply the surface-based approach for the development of the triangulation algorithm, which includes generation of the data structure, subdivision of the cloud data, establishment of the triangulation flow chart, best point computation, creation of interrelated information and development of the post-processing process. The capability of the proposed algorithm is set to be 300,000 points per minute. The maximum number of points analyzed is 600,000 points. Several digitizing devices including point-type laser, line-type laser and CCD digitizing device will be used. The scan data of the above devices will be tested and the proposed algorithm will be tuned in accordance to practical needs. 研究期間：9308 ~ 9407
