本研究的目的在建立X光3D影像重建技術，以幫助雙面電路板BGA銲點的瑕疵檢測。透過X光非破壞性檢測，可透視被遮蔽在內部的結構。BGA銲點檢測若利用電腦斷層掃描技術（CT）來檢測，將可得到清晰的內部資訊，然而CT的缺點是取像的投影角度數量過多，造成檢測的時間過久，並且CT受限於電路板扁平的形狀，無法作環狀各個角度的投影。目前常見的檢測方式皆為X光照相術（Radiography）或薄層成像法（Laminography），然而皆無法獲得準確的3D內部影像。本研究結合薄層成像法的投影運作方式與CT的代數重建技術，以少量的投影角度數重建出待測物的內部垂直截面影像。工業檢測講求快速和準確瑕疵判斷，本文將探討影響重建品質和重建速度的數個因素。另一方面減少重建的範圍以縮短重建時間，運用上下層影像分離技術，將得到的上層BGA銲點投影影像進行重建。最後分別以自行建立的電路板模型進行電腦模擬投影，和X光取得的真實投影影像，進行影像重建。

The purpose of this study is to set up a 3D X-ray image reconstruction technique to inspect the defect of BGA joints of a double-side circuit board. Via nondestructive X-ray inspection, we can know the inside structure of BGA joint. Inspecting BGA joints by using the Computed Tomography(CT) method can help us to get the clear inner information, however one of the disadvantages of CT is that it takes long time to inspect due to so many image of different projective angles are needed. And the other is that the use of CT to inspect large or flat object like circuit board is limited because the full access around the object is necessary due to the CT algorithm. Nowadays the common inspection methods used are radiography and laminography, nevertheless, their handicap is that the reconstruction images are not clear enough. This study combines the laminography and CT algebraic reconstruction technique to reconstruct the vertical inner cross section image by using fewer images of different projective angles. Industrial inspection demands to construct the inner structure of an object fast and accurately, so in this study, the parameters which influence the reconstruction quality and speed are studied. Furthermore, in order to reduce the reconstructive computation time, we reduce the reconstructive image size by separating upper and lower images of the double-side circuit board. Thus we can reconstruct the BGA joints images.

[1] Charting a DFT Course for Limited-access Boards, Agilent Technologies.
[2] X rays expose hidden connections, Test and Measurement Europe, Vol. 8, No. 4, August-September 2000
[3] Z. des Plantes, Eine neue Methde zur Differenzierung in der Röntgenographie, Acta Radio., Vol. 13, 182, 1932.
[4] G. N. Hounsfield, Brit. J. Radiol., 46, p1016-1022, 1973.
[5] J. Zhou, “Entwicklung eines Röntgen-Computerlaminographie-Verfahrens für die Materialprüfung,” Ph.D. dissertation, Technical Faculty, University of Saarbrücken, 1995.
[6] J. Zhou, M. Maisl, H. Reiter, and W. Arnold, “ Computed Laminography for Materials Testing, Applied Physics Letters,“ Vol. 68, Nr. 24, pp.3500-3502, 1996.
[7] V.Sankaran, A.R. Kalukin, and R. P. Kraft, IEEE Transmitions on Components, Packaging, and Manufacturing Technology, Vol. 20, pp.361-366, Sept. 1997.
[8] S. M. Rooks, B. Benhabib, and K. C. Smith, “Development of an Inspection Process for Ball-Grid-Array Technology Using Scanned-Beam X-Ray Laminography,” IEEE Transmitions on Components, Packaging, and Manufacturing Technology, Vol. 18, No. 4, Dec. 1995.
[9] Claus Neubauer, “Intelligent X-ray Inspection for Quality Control of Solder Joints,” IEEE Transmitions on Components, Packaging, and Manufacturing Technology, Vol. 20, No. 2, Apr. 1997.
[10] C. Neubauer, S. Schropfer, R. Hanke, “X-ray Inspection of Solder Joints by Planar Computer Tomography（PCT）,” Electronics Manufacturing Technology Symposium, La Jolla, CA, Vol.1, p.60-64, 1994.
[11] J. Radon, Ber. Verh. Sächs. Akad. Wiss., Leipzig, Math. Phys. Kl. 69, p262, 1917.
[12] R. Gordon, R. Bender, and G. T. Herman, “Algebraic reconstruction
techniques (ART) for three-dimensional electron microscopy and X-ray photography,” J. Theoretical Biology, Vol. 29, pp.471-481, 1970.
[13] Klaus Mueller, “Fast and accurate three-dimensional reconstruction from cone-beam projection data using algebraic methods,” Ph.D. dissertation, The Ohio State University, 1998.
[14] H. Guan and R. Gordon, “A projection access order for speedy convergence of ART (algebraic reconstruction technique): a multilevel scheme for computed tomography, ” Phys. Med. Biol. Vol. 39, pp.2005-2022.
[15] C. Byrne, “Block-iterative methods for image reconstruction from projections,” IEEE Trans. Image Processing. Vol. 5, pp.792-794, May 1996.
[16] Robert L. Siddon, “Fast Calculation of the exact radiological path for a three-dimensional CT array,” Med. Phys. Vol. 12, No. 2, Mar/Apr 1985.
[17] A. Niu, C. Y. Han and W. G. Wee, “Image reconstruction from cone beam scanning using ART with cylindrical voxel model.” Nuclear Science Symposium, 1990. Conference record : Including Sessions on Nuclear Power Systems and Medical Imaging Conference, pp.1445-1451, 1990 IEEE.
[18] C. Moler J. R. Gilbert and R. Schreiber, “Sparse matrices in MATLAB: Design and implementation.” SIAM J. Matrix Anal. Appl., Vol. 13, No. 1, p333-356, Jun. 1992.
[19] Alan Chang, “The performance improvement and analysis for solving aparse linear systems,” Master, thesis, Control Engineering in National Chiao Tung University, 1994.
[20]王新榮,陳時錦,劉亞忠, ”有限元素法及其應用,” 2002.
[21]張順利, “工業CT圖像的代數重建方法研究及應用,” 西北工業大學, 2004.
[22]劉曉, “工業CT圖像重建算法的計算機模擬研究,” 四川大學, 2004.
[23]王小璞, “基於分塊迭代的快速代數重建算法研究,” 西安理工大學, 2001.
[24] A.C. Kak and M. Slaney, Principles of Computerized Tomographic Imagimg, IEEE Press, 1988.
[25] S. Kaczmarz, “Angenaherte auflosung von systemen linear gleichungen,” Bull. Acad. Pol. Sci. Lett. A, Vol. 6-8A, pp.355-357, 1937
[26] S. Gondrom, S. Schröpfer, “Digital computed laminography and tomosynthesis functional principles and industrial applications,” DGZfP Proceedings BB 67-CD, Paper 11, March 1999.
[27] R.N Bracewell, The Fourier Transform and Its Applications, 2nd ed., McGraw-Hill, Inc., 1986.
[28] A. H. Andersen and A. C. Kak, “Simultaneous algebraic reconstruction technique（SART）：A superior implementation of the ART algorithm,” Ultrason. Imaging, vol. 6, p81-94, Jan. 1984.
[29] Rohit Patnaik, Dale Thayer, “Clearvue-Tomosynthetic Reconstruction Techniques for X-Ray Inspection of Circuit Boards,” Teradyne, Inc. Confidential and Proprietary.