在科技業中，製程上需要進行瑕疵檢測，而這些都需要大量的人力來執行，花費許多人力及時間。所以現今自動光學檢測(Automated Optical Inspection, AOI)廣泛應用於產業中，其運用機器視覺的方式來解決傳統人工檢測的缺點，藉此降低人力成本及時間成本，進而使產線達到自動化的需求。
本研究的目的為就物體表面的條紋特徵提出一種瑕疵檢測方法，瑕疵種類包含刮痕及污漬。一開始確認影像的角度及區塊位置的正確性，接著獨立出影像的區塊及白色間隔，在區塊部分，利用一維中值濾波器生成標準的無瑕疵影像，找出原影像及無瑕疵影像的影像差異，藉此分割出區塊內的瑕疵位置；在白色間隔部分，則透過適當的門檻值分割出瑕疵。接著取得區塊結果及白色間隔結果的聯集，即得到整張影像的檢測結果。最後利用影像合成的方式生成多張瑕疵影像，以65張生成影像及20張真實影像，總數為85張影像來評估演算法的性能。結果顯示，以區域比對的方式評估，準確率達94.6%；以像素標籤的方式評估，準確率達88.3%。

In today's industry, defect detection is required in the manufacturing process. However, much manual work and consumes much time are usually needed. As a result, automated optical inspection (AOI) has been widely introduced and implemented in the industry. AOI uses machine vision to overcome the difficulties associated with the detection by human eyes, including labor reduction and time-saving. Another advantage of AOI is that it can also be integrated into an automatic production system.
In this study, a defect detection algorithm to detect the defects with the striped background was proposed. The defect types include scratches and stains. First, we rotated the images to make the stripes horizontal. Then we separated the images into two portions: the blocks and the white intervals. For the blocks, we used a one-dimensional median filter to generate standard defect-free images, found the difference between the original images and the defect-free images, and segmented the positions of the defects. For the white intervals, we separated the defects by setting the appropriate threshold values. Finally, we combine the results to generate the final images with defects identified. We evaluated the performance of the algorithm with 65 images we synthesized and 20 original images. The accuracy was 94.6% in terms of defect location, and the accuracy was 88.3% in terms of pixel label.

[1] 彭健鈞, "一種適應外觀多樣性變化的 晶圓瑕疵檢測技術," 碩士, 電機工程學系, 中華大學, 新竹市, 2017.
[2] 劉得政, "封裝製程之IC表面缺陷自動檢測," 碩士, 資訊工程學系, 元智大學, 桃園縣, 2018.
[3] 謝汶龍, "錫球缺陷檢測演算法," 碩士, 光機電整合產業專班, 國立聯合大學, 苗栗縣, 2013.
[4] 朱俊霖, "利用影像處理技術在晶圓缺陷及硬幣影像對位之應用," 碩士, 資訊管理系碩士班, 國立雲林科技大學, 雲林縣, 2007.
[5] C. C. Wang, B. C. Jiang, J. Y. Lin, and C. C. Chu, "Machine Vision-Based Defect Detection in IC Images Using the Partial Information Correlation Coefficient," IEEE Transactions on Semiconductor Manufacturing, vol. 26, no. 3, pp. 378-384, 2013.
[6] C. H. Li, C. Y. Chang, and M. Jeng, "Applying Regional Level-Set Formulation to Postsawing Four-Element LED Wafer Inspection," IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 41, no. 6, pp. 842-853, 2011.
[7] H. Liu, W. Zhou, Q. Kuang, L. Cao, and B. Gao, "Defect Detection of IC Wafer Based on Spectral Subtraction," IEEE Transactions on Semiconductor Manufacturing, vol. 23, no. 1, pp. 141-147, 2010.
[8] X. Bai, Y. Fang, W. Lin, L. Wang, and B. F. Ju, "Saliency-Based Defect Detection in Industrial Images by Using Phase Spectrum," IEEE Transactions on Industrial Informatics, vol. 10, no. 4, pp. 2135-2145, 2014.
[9] C. H. Yeh, F. C. Wu, W. L. Ji, and C. Y. Huang, "A Wavelet-Based Approach in Detecting Visual Defects on Semiconductor Wafer Dies," IEEE Transactions on Semiconductor Manufacturing, vol. 23, no. 2, pp. 284-292, 2010.
[10] 黃敏霖, "灰階權重式密度分群演算法與LAO晶圓表面瑕疵檢測應用," 碩士, 自動化及控制研究所, 國立臺灣科技大學, 台北市, 2012.
[11] C. Y. Chang, C. H. Li, Y. C. Chang, and M. Jeng, "Wafer defect inspection by neural analysis of region features," Journal of Intelligent Manufacturing, vol. 22, no. 6, pp. 953-964, 2009.
[12] K. Kyeong and H. Kim, "Classification of Mixed-Type Defect Patterns in Wafer Bin Maps Using Convolutional Neural Networks," IEEE Transactions on Semiconductor Manufacturing, vol. 31, no. 3, pp. 395-402, 2018.
[13] V. Singh and A. K. Misra, "Detection of plant leaf diseases using image segmentation and soft computing techniques," Information Processing in Agriculture, vol. 4, no. 1, pp. 41-49, 2017.
[14] S. Arivazhagan, R. Newlin Shebiah, S. Ananthi, and S. Vishnu Varthini, "Detection of unhealthy region of plant leaves and classification of plant leaf diseases using texture features," CIGR, vol. 15, 2013.
[15] D. Aiger and H. Talbot, "The Phase Only Transform for unsupervised surface defect detection," 2010.
[16] M. Liu, Y. Liu, H. Hu, and L. Nie, "Genetic algorithm and mathematical morphology based binarization method for strip steel defect image with non-uniform illumination," Journal of Visual Communication and Image Representation, vol. 37, pp. 70-77, 2016.
[17] R. C. Gonzalez and R. E. Woods, "Digital image processing [M]," Publishing house of electronics industry, vol. 141, no. 7, 2002.
[18] W. Liu, Y. Cai, M. Zhang, H. Li, and H. Gu, "Scene Background Estimation Based on Temporal Median Filter with Gaussian Filtering," presented at the International Conference on Pattern Recognition (ICPR), Cancún Center, Cancún, México, 2016.
[19] D. M. Tsai and D. E. Rivera Molina, "Morphology-based defect detection in machined surfaces with circular tool-mark patterns," Measurement, vol. 134, pp. 209-217, 2019.
[20] D. Huang, S. Liao, A. I. Sunny, and S. Yu, "A novel automatic surface scratch defect detection for fluid-conveying tube of Coriolis mass flow-meter based on 2D-direction filter," 2018.