臺灣博碩士論文加值系統

本論文係有關一種分析與辨認小型盒裝扁方形(QFP) IC之方位的快速影像處理方法。尤其是有關放置在包裝盤上之小型陣列式IC中具有錯誤方位IC的檢測。在小型IC封裝處理過程中因抖動或其他誤差而容易使IC不被放正甚至被放反。因為該等IC之體積小並且一個包裝盤可能有數十個至數百個IC，因此產品作業員不易以目測發覺並且不能有效地在適當的時間中找出未被放正之IC，以至於在後面的電路板製作步驟中導致錯誤以及損失。因此本計畫提出關於用以分析並且辨認被放置在包裝盤上之小型已封裝IC之放置方位的品檢自動化之快速影像處理方法。
在所提方法中，首先將以現場工程師之經驗配合系統之學習功能建立各種常見方位之邏輯代表模式，並且進一步地整合各種表示模式成為一種知識庫。接著在辨認程序中使用二維影像以快速地處理而判定盒中數十個至數百個IC的方位是否有誤，並且與知識庫內所建立各種方位模式相比較以便進行方位之檢測。各種被檢測出錯誤方位之結果除了提供給現場作業員進行更正之外，同時也將被加以統計分類以供作業改進之參考。
本論文將探討把所發展出的方法結合統計程序管制系統而作為IC方位有誤之判定依據，並且經適當化簡整理後供硬體電路之製作。
本論文預定之成果可應用於小型盒裝QFP IC之包裝程序中，以即時方式判定數十個或數百個放置於包裝盤上之小型IC的方位。本計劃的成果也可應用於類似陣列式物品之生產自動化的故障檢測程序。

This proposal relates to a method for analyzing and detecting the orientation of the packaged, small sized IC’s. Especially relates to the packaged IC’s placed on the tray. Generally in the final manufacturing process, a plurality of small sized IC’s, for example, having a surface dimension of 7mm by 7mm, are placed upon a tray for ease of transportation. However, such small IC’s are susceptible to vibration or errors in the manipulation of the IC’s, and may therefore, become wrongly oriented due to its small size.
When the robotic arm picks up a wrongly oriented IC and attaches it onto a circuit board, the circuit board will malfunction. The malfunction circuit board will be rejected, thus increasing the cost of manufacturing. This kind of loss can be avoided if wrongly oriented IC’s can be detected and corrected in advance.
The proposal provides a method for analyzing the image of a multitude of IC’s on a tray and detecting the positions of wrongly oriented IC’s for correction.
In the proposed method, the knowledge of technicians is combined with a learning function in the system to establish suitable representations of the ideal IC images. The representations are further integrated to form a knowledge base. Then, in the analyzing phase, two dimensional images of the IC’s are compared with the ideal IC images to determine the presence of wrongly oriented IC’s. The statistics of the wrong orientations will be collected and categorized for further improvement of the IC manipulation process.
The developed method will also be simplified to suit the implementation of using hardware circuits.
It is expected that the proposed method and apparatus can be applied to the detection of small sized IC orientations in the IC manipulation process. This proposal can also be applied for detecting the mark defections on a single larger sized IC with little modification.

摘要...........................................................I
Abstract.....................................................III
目錄..........................................................VI
圖目錄.....................................................VIIII
表目錄......................................................XIVI
第一章 前言....................................................1
1.1 簡介.......................................................1
1.2 研究動機...................................................2
1.3 論文架構...................................................4
第二章 IC封裝與QFP.............................................6
2.1 IC封裝產業概況.............................................6
2.2 構裝製程簡介..............................................10
2.3 構裝的製造技術............................................14
2.4 構裝型態之介紹............................................16
2.5 結論......................................................18
第三章 相關運算與影像處理.....................................19
3.1 一維相關運算..............................................19
3.2 二維相關運算..............................................22
3.3 相關匹配..................................................25
第四章 QFP之影像處理..........................................27
4.1 影像處理的基本步驟........................................27
4.2 取像......................................................28
4.3 前處理....................................................30
4.4 相關運算於影像處理之運用..................................32
第五章 錯誤方位檢測...........................................35
5.1 一維相關運算的模擬結果....................................35
5.2 二維相關運算的模擬結果...................................41
5.3 傅立葉譜之相關運算模擬結果................................46
第六章 實際模擬結果...........................................57
第七章 結論...................................................62
參考文獻......................................................63

1. Madhav Moganti and Fikret Ercal, “A Subpattern Level
Inspection System for Printed Circuit Boards,” Computer
Vision and Image Understanding, Vol. 70, pp. 51-62, 1998.
2. A. A. Vasiliev, V. V. Dadeshidze, I. N. Kompanets and G. A.
Lunyakova, “Optical Pattern Recognition Using Correlation
Procedure for Rotation and Scale Invariant Feature
Extraction,” IEEE, pp. 1260-1264, 1988.
3. Yo Horikawa, “Pattern Recognition with Invariance to
Similarity Transformations Based on the Third-order
Correlation,” Proceedings of ICPR’96, IEEE Proc. ICPR, pp.
200-204, 1996.
4. Mohammad S. Alam and Yuhui A. Gu, “Joint Transform
Correlation Using Correlation Plane Data Fusion,” IEEE, pp.
694-701, 1996.
5. Jiangsheng You, Weiguo Lu, Jian Li, Gene Gindi and Zhengrong
Liang, “Image Matching for translation, rotation and
uniform scaling by the Radon Transform,” IEEE, pp. 1009-
1012, 1998.
6. D. V. Satish Chandra, “Target Orientation Estimation Using
Fourier Energy Spectrum”, IEEE Trans. Aerospace and
Electronic Systems, Vol. 34, No. 3, pp. 703-712, 1997.
7. Vinod Chandran, Brett Carswell, Boualem Boashash and Steve
Elgar, “Pattern Recognition Using Invariants Defined from
Higher Oder Spetra: 2-D Image Input”, IEEE Transaction on
Image Processing, Vol.6, No5, pp. 788-791, 1997.
8. Xiaoyan Zhu, Yifan Shi and Song Wang, “A New Distinguishing
Algorithm of Connected Character Image Based on Fourier
Transform”, IEEE, pp. 571-576, 1999.
9. N. G. Bourbakis and A. T. Gumahad, “Knowledge Based Text
Character Recognition Using Fourier Transform”, IEEE, pp.
847-851, 1990.
10. Neng-Chung Hu and Kuo-Kan Yu, “Shape Recognition Using
DRFT-based correlator”, Electronics Letters, Vol. 34, No.
11, pp. 1093-1094, 1998.
11. P. Premaratne and C. C. Ko, “Image blur recognition using
under-sampled discrete Fourier transform”, Electronics
Letters, Vol. 35, No. 11, pp. 889-890, 1999.
12. 吳成柯, 程湘君, 戴善榮, 雲立實譯, 數位影像處理, 儒林圖書有限
公司, 1996.
13. 繆紹綱, 數位影像處理-Matlab, 全華科技圖書股份有限公司, 1999.
14. 陳飛龍, 半導體製造自動化-線上教材, 國立清華大學工業工程所,
1997.