為了減低使用電腦輔助製造軟體所需之能力與時間，特徵辨識是一項重要且關鍵之技術，一般而言，特徵在不同加工領域有不同的加工方式，因此難有一套標準的規則來定義特徵，但木屋樑柱的特徵明確且易於辨識，因此本論文針對木屋樑柱加工提出特徵辨識的方法，並針對每個特徵自動化產生加工刀具路徑。本論文之運作流程是先輸入木屋樑柱之3D幾何模型，判斷該模型之凹凸邊，並搜尋特徵所屬的平面或圓弧面，將特徵初步分類為凸特徵與凹特徵；接著產生特徵編碼，與預先建立的標準型特徵庫比較，若編碼符合，則該特徵為標準型特徵，若編碼不符合，則由凹邊存在的狀態與面與面的方向性辨識各種特徵，最後系統共分類出凸塊型、口袋型、孔型、槽型與階梯型五大類特徵；最後確立各類特徵之加工法（例如圓孔使用鑽孔加工、深槽使用鋸片加工等）、切削區域和刀軸方向。
本論文除詳述如何以拓樸與幾何資訊分類凸特徵與凹特徵、以規則方式分類各種特徵及以特徵種類確立加工條件的方法外，並且進行電腦系統開發，以三個案例驗證所開發系統的實用性。

Feature recognition is an important and crucial technology to reduce the operation time and human resources required by a computer aided manufacturing software. Generally speaking, geometric features have different material processing methods in different manufacturing fields. As a result, it is difficult to have a set of standard rules to define features for subsequent material processing procedures. Nonetheless, the feature types of a wooden house beams are distinct and easy to identify. Therefore, this thesis proposes the methodology of recognizing geometric features for wooden house beams, as well as generating tool paths to machine each individual feature. The first step is to classify the topological information by examining the concave and convex edges of the beam’s 3D geometric model, along with the search of planar and cylindrical surfaces belonging to features. Feature codes are then generated and compared with the previously established feature library. If the feature codes match any item in the feature library, the feature can be classified as a standard feature. On the contrary, the feature is classified as non-standard and will be identified by the aspect of the convex edges and the surface normal directions. Non-standard features have five different categories, including boss, pocket, hole, slot and step. In addition, the material processing conditions, the cutting region and the cutting tool path can be established by the geometric data from the identified standard and non-standard features.
Aside from elaborating on the methodology of identifying protrusion and cavity features, classifying machining features by pre-defined codes, and establishing the material processing conditions, this thesis also develops a computer system based on the proposed methodology and uses three case examples to verify the practicability of the system.

[1] Grayer, A.R. (1976), ‘‘A computer link between design and manufacture,’’ Ph.D. thesis, Department of Computer Science and Technology, University of Cambridge, Cambridge, UK.
[2] Joshi, S. (2001), ‘‘A discourse on geometric feature recognition from cad models’’ Journal of Computing and Information Science in Engineering, Vol. 1, No. 1, pp. 41-51.
[3] Lee, Y. and Fu, K. (1987), ‘‘Machine understanding of CSG: extraction and unification of manufacturing features,’’ IEEE Computer Graphics and Applications, Vol. 7, No.1, pp. 22-32.
[4] Kyprianou, L. (1980), ‘‘Shape classification in computer-aided design,’’ Ph.D. thesis, Department of Computer Science and Technology, University of Cambridge, Cambridge, UK.
[5] Joshi, S. and Chang, T. (1988), ‘‘Graph-based heuristics for recognition of machined featured features from a 3D solid model,’’ Computer-Aided Design, Vol. 20, No. 2, pp. 58–66.
[6] Corney, J.R. (1993), ‘‘Graph-based feature recognition,’’ Ph.D. thesis, Department of Mechanical Engineering, Heriot-Watt University, Edinburgh, UK.
[7] Kim, Y.S. (1992), ‘‘Recognition of form features using convex decomposition,’’ Computer-Aided Design, Vol. 24, No. 9, pp. 461–476.
[8] Henderson, M. and Anderson, D. (1984), ‘‘Computer recognition and extraction of form features: a CAD/CAM link,’’ Computers in Industry, Vol. 5, No. 4, pp. 329–339.
[9] Verma, A.K. and Rajotia, S. (2004), ‘‘Feature vector: a graph-based feature recognition methodology,’’ International Journal of Production Research , Vol. 42, No. 16, pp. 3219–3234.
[10] Pinilla, J., Finger, S. and Prinz, F. (1989), ‘‘Shape feature description using an augmented topology graph grammar,’’ Preprints: NSF Engineering Design Research Conference, June 11–14, 1989, Amherst, MA, USA, pp. 285–300.
[11] Narayan, S.V. and Ling, Z.K. (1994), ‘‘Heuristics based feature recognition: a graph approach,’’ Advances in Design Automation, Vol. 1, No. 1, pp. 299-306.
[12] Wang, M.T. (1990), ‘‘A geometric reasoning methodology for manufacturing feature extraction from a 3D CAD model,’’ Ph.D. thesis, School of Industrial Engineering, Purdue University, West Lafayette Indiana, IN, USA.
[13] Madurai, S.S. and Lin, L. (1992), ‘‘Rule-Based automatic part feature extraction and recognition from CAD data,’’ Computers & Industrial Engineering, Vol. 22, No. 1, pp. 49-62.
[14] Riona, I. (2016), ‘‘A rule-based two-level classification approach for recognition of machining features from 3d solid models.’’, Master’s thesis, Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei.
[15] Prabhakar, S. and Henderson, M.R. (1992), ‘‘Automatic form-feature recognition using neural-network-based techniques on boundary representations of solid models,’’ Computer-Aided Design, Vol. 24, No. 7, pp. 381–393.
[16] Shi, Y., Zhang, Y. and Harlik, P. (2020), ‘‘Manufacturing feature recognition with a 2D convolutional neural network,’’ CIRP Journal of Manufacturing Science and Technology, Vol. 30, August 2020, pp. 36–57.
[17] 張閔淳，「以製程需求表達法辨認切削特徵」（2017），碩士論文，交通大學工業工程與管理系所，新竹市。
[18] Malleswari, V.N., Valli, P.M. and Sarcar, M.M.M. (2013), ‘‘Automatic recognition of machining features using STEP files,’’ International Journal of Engineering Research & Technology, Vol. 2, No. 3.
[19] Khusna, D. and Hideki, A. (2014), ‘‘Basic study on process planning for turning-milling center based on machining feature recognition,’’ Journal of Advanced Mechanical Design, Systems, and Manufacturing, Vol. 8, No. 4, Pages jamdsm0058.
[20] Zhang, X., Nassehi, A. and Newman, S.T. (2014), ‘‘Feature recognition from CNC part programs for milling operations,’’ International Journal of Advanced Manufacturing Technology, Vol. 70, No. 1, pp. 397–412.
[21] Elmokadem, A. (2017), ‘‘Towards an efficient process planning of the V-bending process: an enhanced automated feature recognition system,’’ International Journal of Advanced Manufacturing Technology, Vol. 9, No. 1, pp. 4163–4181.
[22] https://www.luoow.com/dc_tw/109106725