隨著科技的進步，越來越多的嵌入式裝置被應用在許多的工廠當中。其中，嵌入式系統有著低成本、體積小與低耗電等特性。由此可見，嵌入式系統有非常大的優勢能夠取代體積大且高耗能的工業電腦。而檢測系統在半導體工廠中扮演著非常重要的角色。它用來保證產品的品質與正確性，但由於檢測系統具有龐大的影像處理運算，因此，在嵌入式處理器上演算法複雜度變得非常重要。
本碩士論文提出一嵌入式多核心之即時自動IC標誌檢測系統。其中我們將在本論文呈現基於嵌入式多核心處理器架構之IC檢測平台的系統層級設計與實現。此外，我們還提出了高效能的角度估計演算法用來進行目標IC的定位與方位旋轉角度。另外，我們所提出的角度估計演算法能夠有效的搭配經典的模板匹配演算法，進而大幅度的增加IC標誌檢測的準確度與執行速度。最後，我們更針對處理器架構來進行演算法層級的效能優化。在使用NEON單指令流多資料流和多核心技術下，系統能夠以更高的效率執行IC檢測，使能夠應用在目前許多的工廠中。經由實驗的證明，我們所提出的演算法能夠在嵌入式架構下能夠以高效率的執行能力進行運算，並且達到即時性。當目標圖像大小為640×480、模板圖像大小為80×100時，平均每張影像的IC標誌檢測可以在31微秒的時間以合理的準確度有效率的分辨出優良IC與瑕疵IC。

This thesis presents the system architecture and the algorithm of a real-time automated IC marking inspection system based on an embedded multi-core platform. To be specific, the system-level design and implementation of an IC marking inspection platform with the embedded multi-core processors is illustrated. Furthermore, an efficient angle estimation algorithm is proposed such that a rotation and location of the IC chip can be identified efficiently. Moreover, the proposed angle estimation algorithm is combined with the classic template matching algorithm to accurately inspect the marking of the targeted IC chips. In addition, we present the algorithmic optimizations based on the architecture of multi-core embedded processors and the NEON general-purpose single instruction multiple data (SIMD) engine such that the computation time is significantly reduced. A formal outline of the proposed algorithm will be given and experiment results with PC-based platform and embedded-based platform will be presented showing that, comparing with traditional IC marking inspection algorithm, our proposed algorithm can be greatly improve image processing efficiency on embedded system platform. When the size of the target image is 640×480 pixels and size of the template image is 80×100, the average inspection time of our algorithm is 31 ms. And our algorithm can distinguish good IC chips from not good IC chips with reasonable accuracy.

[1] W.-C. Wang, L.-B. Chen, W.-J. Chang and S.-M. Li, "A Machine Vision Based Automatic Optical Inspection System for Measuring Drilling Quality of Printed Circuit Boards," Accepted to be published in IEEE Access, 2017.
[2] B. Zhang, H. Yang and Z. Yin, "A Region-Based Normalized Cross Correlation Algorithm for the Vision-Based Positioning of Elongated IC Chips," IEEE Transaction on Semiconductor Manufacturing, vol. 28, no. 3, pp. 345-352, May 2015.
[3] N. Cai, J. Lin, Q. Ye, H. Wang, S. Weng and W.-K. Ling, "A New IC Solder Joint Inspection Method for an Automatic Optical Inspection System Based on an Improved Visual Background Extraction Algorithm," IEEE Transaction on Components, Packaging and Manufacturing Technology, vol. 6, no. 1, pp. 161-172, Dec. 2015.
[4] N. Amin and M. S. Khadem, "Interface Development for Cost Effective Automated IC Orientation Checking Systems," in proceeding of IEEE International Conference on Computer and Information Technology (ICCIT), Dec. 2007.
[5] B. C. Jiang, S.-L. Tasi and C.-C. Wang, "Machine Vision-Based Gray Relational Theory Applied to IC Marking Inspection," IEEE Transaction on Semiconductor Manufacturing, vol. 15, no. 4, pp. 531-539, Nov. 2002.
[6] R. Nagarajan, S. Yaacob, P. Pandian, M. Karthigayan, S. H. Amin and M. Khalid, "A Real Time Marking Inspection Scheme for Semiconductor Industries," Journal of Advanced Manufacturing Technology, vol. 34, no. 9, pp.926-932, Oct. 2007.
[7] J. Yoo, S. S. Hwang, S. D. Kim, M. S. Ki and J. Cha, "Scale-Invariant Template Matching using Histogram of Dominant Gradients," Journal of Pattern Recognition, vol. 47, no. 9, pp. 3006-3018, Sep. 2014.
[8] W.-C. Lee and C.-H. Chen, "A Fast Template Matching Method With Rotation Invariance by Combining the Circular Projection Transform Process and Bounded Partial Correlation," IEEE Signal Processing Letters, vol. 19, no. 11, pp. 737-740, Nov. 2012.
[9] S. Mattoccia, F. Tombari, L. D. Stefano, "Fast Full-Search Equivalent Template Matching by Enhanced Bounded Correlation," IEEE Transaction on Image Processing, vol. 17, no. 4, pp. 528-538, Apr. 2008.
[10] N. Jazdi, "Cyber Physical Systems in the Context of Industry 4.0," in proceeding of IEEE International Conference on Automation, Quality and Testing Robotics, May 2014.
[11] L.-M. Po and K. Guo, "Transform-Domain Fast Sum of the Squared Difference Computation for H.264/AVC Rate-Distortion Optimization," IEEE Transaction on Circuits and Systems for Video Technology, vol. 17, pp. 765-773, Jun. 2007.
[12] C.-Y. Chen, S.-Y. Chien, Y.-W. Huang, T.-C. Chen, T.-C. Wang and L.-G. Chen, "Analysis and Architecture Design of Variable Block-Size Motion Estimation for H.264/AVC," IEEE Transaction on Circuit ans Systems I, vol. 53, no. 2, pp. 578-593, Feb. 2006.
[13] H. Lv, R. Wang, J. Wan, H. Jia, Xiaodong Xie and W. Gao, "An Efficient NEON-Based Quarter-pel Interpolation Method for HEVC," in proceeding of IEEE Conference on Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Dec. 2012.
[14] H. Blume, J. von Livonius, L. Rotenberg, T. G. Noll, H. Bothe and J. Brakensiek, "OpenMP-based parallelization on an MPCore multiprocessor platform - a performance and power analysis," Journal of Systems Architecture, vol. 54, no. 11, pp. 1019–1029, Nov. 2008.
[15] F. Chang, C.-J. Chen and C.-J. Lu, "A linear-time component-labeling algorithm using contour tracing technique," Journal of Computer Vision and Images Understanding, vol. 93, no. 2, pp. 206-220, Feb. 2004.
[16] M.-k. Hu, "Visual Pattern Recognition by Moment Invariants," IRE Transaction on Information Theory, vol. 8, no. 2, pp. 179-187, Feb. 1962.
[17] H. Yang, S. Zheng, J. Lu and Z. Yin. "Polygon-Invariant Generalized Hough Transform for High-Speed Vision-Based Positioning," IEEE Transaction on Automation Science and Engineering, vol. 13, no. 3, pp. 1367-1384, Jun. 2016.
[18] Z. Yuanyuan, "Research on Automatic Visual Inspection Method for Character on Cartridge Fuse Based on Template Matching," in proceeding of IEEE International Conference on Information Science and Control Engineering (ICISCE), Jul. 2016.
[19] S.-h. Kim, S.-H. Song, J.-H. Kim, Z. Yuan and J.-D. Cho, "Fast Rotation-Invariant Template Matching with Candidate Reduction using CUDA," in proceeding of IEEE International Symposium on Consumer Electronics (ISCE), Jun. 2015.
[20] W. Chen, L.-M. Wu and S.-K. Cui, "A High Speed Image Preprocessing Method for IC wafer Inspection," in proceeding of IEEE International Conference on Electronic Packaging Technology and High Density Packaging (ICEPT-HDP), Jul. 2008.
[21] "Boradcom BCM2837 - Raspberry Pi Documentation" [Online]. Available: https://www.raspberrypi.org/
[22] ARM Cortex-A53 MPCore Processor Technical Reference Manual, Revision:r0p4, ARM, Cambridge, England, United Kingdom, 2013.