本研究乃是利用機器視覺技術來精確估算冷陰極管經過高週波趕汞設備的高溫趕汞氣化後,其汞塊析出擴散現象的不良品管制線上檢測。研究方向主要針對冷陰極管上汞塊氣化析出量擴散分布特徵做檢測，另外為符合自動化操作,機台動力採用INDEX間歇傳送榙配鍊條上設計承置燈管V槽治具自動輸送進給方式,同時在檢測燈管前後有壹組定位缸來確保燈管位置,因此燈管檢出範圍即可自動尋找出特徵位置並計算出補償值。在線上檢測過程中，先利用目標物的燈絲特徵尋找出外型輪廓，以此為基準取50*100框像運算後找出汞塊氣化析出量。檢測進行步驟為，光源採取打背光方式，CCD(Charge-coupled Device電荷耦合元件)取像後再經程式取50*100框像來二值化處理、以灰階分佈圖分析汞析出量特徵尋找。最後本研究實際擷取30張影像作為檢測樣本，以驗證所提方法可行性並獲得高穩定度品質檢驗，做到全面的品質檢驗的目的。

This thesis is using technical of Machine Vision to estimate precisely expansion of Hg which in CCFL that passes through the High Frequency Equipment. The study is focus on the oxidization distribution characteristics of Hg which in CCFL. Besides, in order to fit with the automatic operation. Machine power uses INDEX intermittent transmission with an automatic deliver design. At the same time, there is a set of pneumatic piston to ensure tube''s position. Therefore it can automatically search the position of characteristic through the moving. In the examination, first it uses filament characteristic to find out the external sketch. Then take this result as a basis, select a 50*100 frame to analysis and find out the oxidization distribution of Hg. The step of examination is as below. First, using back-lighted light source, then the program will select 50*100 frame to the process of binary after CCD get pictures. Using the historgram to analysis the characteristic of the oxidization distribution of Hg. Finally, this research picks up 30 images as examination sample to prove the practicability of this method and acquire high stable quality.

[1] A. Rosenfeld and A.C. Kak, Digital Picture Processiny, 2 nd ed.,
Vol.2New 5ork: Academic Piess,1982.
[2]Tzag Y. Young and King-Swn Fu,Handbook of Pattern Recognition
and Imoye Processing,1986.
[3] S. Mori, et al., “Research on Machine Recognition of Handprinted
Characters”, IEEE Transactions on Pattern Analysis and Machine
Intelligence, Vol, PAMI-6, NO.4, pp386-45,July 1984
[4] T. Fukuda and T. shibata, “Theory and Applications of Nerural
Networks for Industrial Control Systems”, IEEE Trans. on IE,vol.
39, no. 6, pp. 472-489, 1992
[5] R. Oka, “Handprinted Chinede Character Recognition by using
Cellular Features”, Proc. 6th Int. Joint Conf. Pattern Recognition,
pp. 783-785,1982,
[6] k. Mori and I. Masuda, “Advances in Recognition of Chinese
Characters”, Proc. 5th ICPR.,pp. 692-702, 1980
[7] T. Fujita,M. Nakanishi, M. Niwa, “kanji character Recognition
system”, Fujitsu Scientific and Technical Journal, Sept., 1997.
[8] M. Chester, Neural Networks：A Tutorial,Prentice-Hall,1993
[9] Min Hong Han, Dongsig Jang and Joseph Foster, “Identification of Cornerpoints of Two-Dimensional Images Using A Line Search Method”, Pattern Recognition, Vol. 22, No.1, pp13-20,1989
[10] Bir Bhanu, Olivier D. Faugeras, “Shape Matching of Two Dimensional Objects”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. PAMI-6, No.2, March, 1984.
[11] P. Johnson, “Contrast Based Edge Detection ”,Pattern Recognition , Vol. 23. No. 3/4, pp.311-318, 1990.
[12] William Huang, Richard Lippmann, and Ben Gold, “A Neural Net
Approache to Speech Recognition”, IEEE ICASSP 88, S3.1, pp. 99-
102, April ,1988
[13] Todd Law, Student Member, ”Image Filtering, Edge Detection, and Edge Tracing Using Fuzzy Reasoning”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.18, No.5, May,1996.
[14] H.a. Glucksman, “Multicategory classification of patterns
represented by high-order vectors of multilevel measurements”,
IEEE Trans. Compuit., Vol. 20, pp. 1593-1598,1971.
[15] 吳成柯、戴善榮、程湘君、雲立實，數位影像處理，儒林圖書
有限公司 p12-13,1993.
[16] 王鎮雄, ”影像處理形狀辨別系統”, 中央大學機械所碩士畢業論文, 1991.
[17] 麥哲鳴, ”幾何圖形影像資料之CAD向量化研究”, 中央大學機械所碩士畢業論文, 1995.
[18] 李獻仁, ”影像引導自動對焦技術之研發”,中央大學機械所碩士畢業論文, 1996.
[19] 范勝威, ”影像引導精密定位及定向系統整合及實作”,中央大學機械所碩士畢業論文, 1996.
[20] 劉美兆, ”電腦連接器線上檢測機構之對焦系統研發”,中央大學機械所碩士畢業論文, 1997.
[21] 陳仁杰, ”CCD檢測技術在電腦連接器之應用”, 中央大學機械所碩士畢業論文, 1997.
[22] 喬珊, “CCD影像幾何圖形之辨識”, 中央大學機械所碩士畢業論文, 1998.
[23] 王進德、蕭大全,”類神經網路與模糊控制理論入門”, 全華科技圖書股份有限公司,修訂版, p. 2-12, 23-36, 136-146, 194-206,2003.
[24] 楊武智, ”影像處理與辨識”, 全華科技圖書股份有限公司, 1994.
[25] 黃衍任，”機器視覺系統”講義，中央大學，2004.
[26] 衛祖賞, ”數位影像處理”, 全華科技圖書股份有限公司
,pp220- 222, 1996.
[27] 威力盟型錄,2003.