本研究提出一個沉積物高度之影像檢測系統於直接能量沉積製程，其中提出了影像扭曲校正、座標轉換、雜訊光過濾及兩種不同的檢測方式來完成此研究。
　　為了準確的量測沉積物高，本研究提出一利用影像處理的方式，來將因室內光、火花及電漿效應所產生的非需要之雜訊光除去，並將彩色影像轉至二元化影像來計算。
　　在影像感測系統中，因為視野場效應和視角差現象，所捕捉的影像產生非線性扭曲。因此，為了解決所提及之議題，本研究提出校正棒法來轉換像素值至實際尺寸，並使用多點卡法來將影像座標轉至世界座標。
　　兩個沉積物高之量測方法於DED製程描述如下。幾何擷取量測法為一定量量測法，並結合座標軸轉換來建構出一個定位式感測系統。而參考點量測法為一定性量測法，此方法之結果可利用一常數將其補償並貼近實際沉積物高。
　　其結果顯示出．本研究所提出之量測法皆與實際沉積物有著相同的趨勢。此外，其實際沉積物高與幾何擷取量測法和參考點量測法之誤差分別為0.1191和0.139mm。
本研究所提出之影像監測系統使用數位相機和影像處理理論，而量測結果與實際沉積物有著高度的一致性。各方面來說，本研究提出了一快速、方便、精準且經濟的沉積物高之監測系統。

This study proposes a vision-based inspection system for measuring deposition height in direct energy deposition process. The proposed approach include image distortion calibration, coordinate transformation, noise light filter and two deposition height measurement methods.
For measuring the deposition heights accurately, an image processing method is proposed to remove undesirable zone caused by room light, spackle and plasma effect and converts the color image into binary image.
In the vision-based inspection system, because of the field of view effect and perspective phenomenon, the captured image will have nonlinear image shape distortion. Thus, for addressing the mentioned issues, the calibration bar method to transform pixel value to real size is proposed. Then, multiple point card (MPC) method is employed to transform image coordination to world coordination.
Two methods for measuring the deposition height in DED process can be described as followings. The geometry extraction measurement method is a quantification measurement method which uses the coordinate transformation algorithm to construct a located inspection system. Another method is qualitative method named reference point measurement method, and the result of this method can be compensated by a constant to fit the deposition height of produced parts.
The results show that the deposition height of produced parts and the calculated height by using the proposed methodology express the same trend. Additionally, the error between the measurement deposition height by using the two proposed approaches: geometry extraction; reference point and deposition height of produced parts is 0.1191 and 0.139 mm, respectively.
The proposed methods use digital camera and image processing theory. The measured results demonstrate good agreements with real value of deposition height. All in all, it can be concluded that this study has proposed a rapid, convenient, accurate and economical inspection system for measuring the deposition height in DED process.

[1] B. Dutta, S. Palaniswamy, J. Choi, L. Song, and J. Mazumder, "Additive manufacturing by direct metal deposition," Advanced Materials & Processes, vol. 169, pp. 33-36, 2011.
[2] J. Bohandy, B. Kim, and F. Adrian, "Metal deposition from a supported metal film using an excimer laser," Journal of Applied Physics, vol. 60, pp. 1538-1539, 1986.
[3] G. K. Lewis and E. Schlienger, "Practical considerations and capabilities for laser assisted direct metal deposition," Materials & Design, vol. 21, pp. 417-423, 2000.
[4] A. Buongiorno, "Laser/powdered metal cladding nozzle," ed: Google Patents, 1995.
[5] I. Gibson, D. W. Rosen, and B. Stucker, Additive manufacturing technologies: rapid prototyping to direct digital manufacturing: Springer Publishing Company, Incorporated, 2009.
[6] F. Meriaudeau and F. Truchetet, "Control and optimization of the laser cladding process using matrix cameras and image processing," Journal of Laser Applications, vol. 8, pp. 317-324, 1996.
[7] G. Tapia and A. Elwany, "A Review on Process Monitoring and Control in Metal-Based Additive Manufacturing," Journal of Manufacturing Science and Engineering, vol. 136, p. 060801, 2014.
[8] D. Michael and A. Wallack, "Multiple field of view calibration plate having a reqular array of features for use in semiconductor manufacturing," ed: Google Patents, 1998.
[9] Y. Y. Tang and C. Y. Suen, "Image transformation approach to nonlinear shape restoration," IEEE transactions on Systems, Man, and Cybernetics, vol. 23, pp. 155-172, 1993.
[10] R. Hartley and A. Zisserman, Multiple view geometry in computer vision: Cambridge university press, 2003.
[11] M. Asselin, E. Toyserkani, M. Iravani-Tabrizipour, and A. Khajepour, "Development of trinocular CCD-based optical detector for real-time monitoring of laser cladding," in Mechatronics and Automation, 2005 IEEE International Conference, 2005, pp. 1190-1196.
[12] O. Faugeras, Q.-T. Luong, and T. Papadopoulo, The geometry of multiple images: the laws that govern the formation of multiple images of a scene and some of their applications: MIT press, 2004.
[13] C. Jawahar, P. K. Biswas, and A. Ray, "Investigations on fuzzy thresholding based on fuzzy clustering," Pattern Recognition, vol. 30, pp. 1605-1613, 1997.
[14] C. Jawahar, P. K. Biswas, and A. Ray, "Analysis of fuzzy thresholding schemes," Pattern Recognition, vol. 33, pp. 1339-1349, 2000.
[15] M. Sezgin, "Survey over image thresholding techniques and quantitative performance evaluation," Journal of Electronic imaging, vol. 13, pp. 146-168, 2004.
[16] N. Otsu, "A threshold selection method from gray-level histograms," IEEE transactions on systems, man, and cybernetics, vol. 9, pp. 62-66, 1979.
[17] J. T. Hofman, Development of an observation and control system for industrial laser cladding: University of Twente, 2009.
[18] D. Hu and R. Kovacevic, "Modelling and measuring the thermal behaviour of the molten pool in closed-loop controlled laser-based additive manufacturing," Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, vol. 217, pp. 441-452, 2003.
[19] D. Hu and R. Kovacevic, "Sensing, modeling and control for laser-based additive manufacturing," International Journal of Machine Tools and Manufacture, vol. 43, pp. 51-60, 2003.
[20] D. Hu, M. Labudovic, and R. Kovacevic, "On-line sensing and estimation of laser surface modification by computer vision," Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, vol. 215, pp. 1081-1090, 2001.
[21] D. Hu, Y. Wu, and R. Kovacevic, "Heat input control in 3d laser cladding based on infrared sensing," in International Mechanical Engineering Congress, Nov, 2001, pp. 11-16.
[22] J. Mazumder, D. Dutta, N. Kikuchi, and A. Ghosh, "Closed loop direct metal deposition: art to part," Optics and Lasers in Engineering, vol. 34, pp. 397-414, 2000.
[23] L. Song, V. Bagavath-Singh, B. Dutta, and J. Mazumder, "Control of melt pool temperature and deposition height during direct metal deposition process," The International Journal of Advanced Manufacturing Technology, vol. 58, pp. 247-256, 2011.
[24] L. Song and J. Mazumder, "Feedback control of melt pool temperature during laser cladding process," IEEE Transactions on Control Systems Technology, vol. 19, pp. 1349-1356, 2011.
[25] M. Iravani-Tabrizipour, "Image-Based Feature Tracking Algorithms for Real-Time Clad Height Detection in Laser Cladding," University of Waterloo, 2007.
[26] M. Iravani-Tabrizipour and E. Toyserkani, "An image-based feature tracking algorithm for real-time measurement of clad height," Machine Vision and Applications, vol. 18, pp. 343-354, 2007.
[27] A. Fathi, A. Khajepour, E. Toyserkani, and M. Durali, "Clad height control in laser solid freeform fabrication using a feedforward PID controller," The International Journal of Advanced Manufacturing Technology, vol. 35, pp. 280-292, 2006.
[28] A. Fathi and A. Mozaffari, "Vector optimization of laser solid freeform fabrication system using a hierarchical mutable smart bee-fuzzy inference system and hybrid NSGA-II/self-organizing map," Journal of Intelligent Manufacturing, vol. 25, p. 775, 2014.
[29] K. Rao, D. Kim, and J. Hwang, "Two-dimensional discrete Fourier transform," in Fast Fourier Transform-Algorithms and Applications, ed: Springer, 2010, pp. 127-184.
[30] H. Jiang, B. W. Pogue, M. S. Patterson, K. D. Paulsen, and U. L. Osterberg, "Optical image reconstruction using frequency-domain data: simulations and experiments," JOSA A, vol. 13, pp. 253-266, 1996.
[31] W. L. Luyben, Process modeling, simulation and control for chemical engineers: McGraw-Hill Higher Education, 1989.
[32] M. Mojiri, M. Karimi-Ghartemani, and A. Bakhshai, "Time-domain signal analysis using adaptive notch filter," IEEE Transactions on Signal Processing, vol. 55, pp. 85-93, 2007.
[33] N. Bassiou and C. Kotropoulos, "Color image histogram equalization by absolute discounting back-off," Computer Vision and Image Understanding, vol. 107, pp. 108-122, 2007.
[34] H. Cao and Y. Shen, "Application of MATLAB image processing technology in sewage monitoring system," in Electronic Measurement & Instruments, 2009. ICEMI'09. 9th International Conference on, 2009, pp. 3-993-3-995.
[35] G. P. Sahay, N. Sahay, and S. Samanta, "Intelligent Vehicle Braking System Using Image Processing."
[36] K. Levenberg, "A method for the solution of certain non-linear problems in least squares," Quarterly of applied mathematics, vol. 2, pp. 164-168, 1944.
[37] J. A. Suykens and J. Vandewalle, "Least squares support vector machine classifiers," Neural processing letters, vol. 9, pp. 293-300, 1999.
[38] D. G. Lowe, "Object recognition from local scale-invariant features," in Computer vision, 1999. The proceedings of the seventh IEEE international conference on, 1999, pp. 1150-1157.
[39] N. D'Apuzzo, "Overview of 3D surface digitization technologies in Europe," Three-Dimensional Image Capture and Applications, 2006.
[40] David Locke, Antonio Candel-Ruiz, "Laser metal deposition defined," in Industrial Laser Solution. Retrieved November 01,2010 from http://www.industrial-lasers.com/articles/print/ volume-250/issue-6/features/laser-metal-deposition.html.
[41] Nikon. Retrieved form http://imaging.nikon.com/lineup/sportoptics
/how_to/guide/binoculars/basic/basic_08.htm#basic001.
[42] GOM, ATOS Compact Scaner. Retrieved form http://www.gom. com /metrology-systems/atos/atos-compact-scan.html.
[43] GOM, GOM Inspect. Retrieved form http://www.gom. com / 3d-software/gom-inspect.html.