產品表面瑕疵檢測一直以來都是熱門的研究領域，所謂有生產活動就會有不良品的產生，問題在於如何更快速，更有效率地將瑕疵檢測出來，讓不良品不會流入消費者或客戶手中。近幾年隨著電腦科學(Computer Science)的普及，大量與人工智能(Artificial Intelligence)相關的應用充斥於我們的生活當中，像是以自然語言處理應用的聊天機器人，ChatGPT的出現不管是幫助個人獲得問題的解答，也幫助企業應用科技來改善公司運作流程，人員招募，減少大量且繁瑣的重複性任務上，讓企業可以將更多的時間及金錢放在更重要的業務上。在產品表面瑕疵上，自動光學檢測(Automated Optical Inspection)已經應用於工業界多年，不管是半導體業的晶圓缺陷、或是PCB業檢測外觀瑕疵都有不錯的效果，近幾年應用深度學習(Deep Learning)模型來協助產線自動化的提升上以FPN(Feature Pyramid Network)為基礎的Faster R-CNN與其之後的相關後續的改善模型及YOLO 系列為主軸，本文以YOLOv7 為架構，在此架構之上優化演算法來達到模型效能的提升 ; 並以K-fold 交叉實驗驗證模型的可靠度及堅固性。

The detection of product surface defects has always been a hot research field. The so-called production activities will produce defective products. The problem is detecting defects faster and more efficiently so that defective products will not flow into consumers. From the quality control perspective, the Seven Basic Tools of Quality and Quality Control Chart are the tools to help quality engineers monitor the process and take countermeasures in advance. In recent years, with the popularization of computer science, many applications related to artificial intelligence have flooded our lives, such as chatbots using natural language processing applications. The total solution provided by IT companies helps enterprises to apply technology to improve the operation process and personnel recruitment and reduce a large number of tedious as well as repetitive tasks; so that enterprises can spend more time on more important business. Regarding product surface defects, Automated Optical Inspection has been used in the industry for many years. It has successful examples whether it is wafer defects in the semiconductor industry or the detection of appearance defects in the PCB industry. The application of deep learning models to assist the improvement of automation. FPN-based models, like Faster R-CNN and its subsequent related improved models and YOLO series as mainstream. This paper used YOLOv7 as the framework, improved the algorithm to increase the model’s performance, and verified its robustness and reliability by K-fold cross-validation.

[1] Tao, X., Zhang, D., Ma, W., Liu, X., & Xu, D. (2018). Automatic metallic surface defect detection and recognition with convolutional neural networks. Applied Sciences, 8(9), 1575.
[2] Tsanakas, J. A., Chrysostomou, D., Botsaris, P. N., & Gasteratos, A. (2015). Fault diagnosis of photovoltaic modules through image processing and Canny edge detection on field thermographic measurements. International Journal of Sustainable Energy, 34(6), 351-372.
[3] Tastimur, C., Yetis, H., Karaköse, M., & Akin, E. (2016). Rail defect detection and classification with real time image processing technique. International Journal of Computer Science and Software Engineering, 5(12), 283.
[4] Jian, C., Gao, J., & Ao, Y. (2017). Automatic surface defect detection for mobile phone screen glass based on machine vision. Applied Soft Computing, 52, 348-358.
[5] Mak, K. L., Peng, P., & Yiu, K. F. C. (2009). Fabric defect detection using morphological filters. Image and Vision Computing, 27(10), 1585-1592.
[6] Li, X., Gao, B., Woo, W. L., Tian, G. Y., Qiu, X., & Gu, L. (2016). Quantitative surface crack evaluation based on eddy current pulsed thermography. IEEE Sensors Journal, 17(2), 412-421.
[7] Yuan, X. C., Wu, L. S., & Peng, Q. (2015). An improved Otsu method using the weighted object variance for defect detection. Applied Surface Science, 349, 472-484.
[8] Win, M., Bushroa, A. R., Hassan, M. A., Hilman, N. M., & Ide-Ektessabi, A. (2015). A contrast adjustment thresholding method for surface defect detection based on mesoscopy. IEEE Transactions on Industrial Informatics, 11(3), 642-649.
[9] Kalaiselvi, T., & Nagaraja, P. (2015). A rapid automatic brain tumor detection method for MRI images using modified minimum error thresholding technique. International Journal of Imaging Systems and Technology, 1(25), 77-85.
[10] Wang, L., Zhao, Y., Zhou, Y., & Hao, J. (2016). Calculation of flexible printed circuit boards (FPC) global and local defect detection based on computer vision. Circuit World.
[11] Bai, X.; Fang, Y.; Lin, W.; Wang, L.; Ju, B.F. (2014). Saliency-based defect detection in industrial images by using phase spectrum. IEEE Transactions on Industrial Informatics, 10, 2135–2145.
[12] Borwankar, R., & Ludwig, R. (2018). An optical surface inspection and automatic classification technique using the rotated wavelet transform. IEEE Transactions on Instrumentation and Measurement, 67(3), 690-697.
[13] Hu, G. H. (2015). Automated defect detection in textured surfaces using optimal elliptical Gabor filters. Optik, 126(14), 1331-1340.
[14] Susan, S., & Sharma, M. (2017). Automatic texture defect detection using Gaussian mixture entropy modeling. Neurocomputing, 239, 232-237.
[15] Cen, Y. G., Zhao, R. Z., Cen, L. H., Cui, L. H., Miao, Z. J., & Wei, Z. (2015). Defect inspection for TFT-LCD images based on the low-rank matrix reconstruction. Neurocomputing, 149, 1206-1215.
[16] Identification of outliers (monographs on statistics and applied probability), D. Hawkins(1st ed.), Springer, Netherlands (1980).
[17] Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3), 1-58.
[18] Chawla, Nitesh & Japkowicz, Nathalie & Kołcz, Aleksander. (2004). Editorial: Special Issue on Learning from Imbalanced Data Sets. SIGKDD Explorations. 6. 1-6. 10.1145/1007730.1007733.
[19] Phua, Clifton & Alahakoon, Damminda & Lee, Vincent. (2004). Minority Report in Fraud Detection: Classification of Skewed Data. SIGKDD Explorations. 6. 50-59.
[20] Fujimaki, Ryohei & Yairi, Takehisa & Machida, Kazuo. (2005). An approach to spacecraft anomaly detection problem using Kernel Feature Space. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 401-410. 10.1145/1081870.1081917.
[21] Ouali, Y., Hudelot, C, & Tami, M. (2020). An overview of deep semi-supervised learning. arXiv preprint arXiv:2006.05278.
[22] Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267-288.
[23] Van der Maaten, L., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of Machine Learning Research, 9(11).
[24] Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
[25] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
[26] Tan, M., & Le, Q. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning (pp. 6105-6114). PMLR.
[27] Wang, C. Y., Liao, H. Y. M., Wu, Y. H., Chen, P. Y., Hsieh, J. W., & Yeh, I. H. (2020). CSPNet: A new backbone that can enhance learning capability of CNN. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops (pp. 390-391).
[28] He, K., Zhang, X., Ren, S., & Sun, J. (2015). Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 37(9), 1904-1916.
[29] Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2117-2125).
[30] Zhang, C., Zou, Y., Chen, G., & Gan, L. (2020). Pan: Towards fast action recognition via learning persistence of appearance. arXiv preprint arXiv:2008.03462.
[31] Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time objectdetection with region proposal networks. Advances in Neural Information Processing Systems, 28.
[32] Dai, J., Li, Y., He, K., & Sun, J. (2016). R-fcn: Object detection via region-based fully convolutional networks. Advances in Neural Information Processing Systems, 29.
[33] He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 2961-2969).
[34] Yang, Z., Liu, S., Hu, H., Wang, L., & Lin, S. (2019). Reppoints: Point set representation for object detection. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 9657-9666).
[35] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). Ssd: Single shot multibox detector. In European conference on computer vision (pp. 21-37). Springer, Cham.
[36] Law, H., & Deng, J. (2018). Cornernet: Detecting objects as paired keypoints. In Proceedings of the European conference on computer vision (ECCV) (pp. 734-750).
[37] Duan, K., Bai, S., Xie, L., Qi, H., Huang, Q., & Tian, Q. (2019). Centernet: Keypoint triplets for object detection. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 6569-6578).
[38] Rashwan, A., Kalra, A., & Poupart, P. (2019). Matrix Nets: A new deep architecture for object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops (pp. 0-0).
[39] Tian, Z., Shen, C., Chen, H., & He, T. (2019). Fcos: Fully convolutional one-stage object detection. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 9627-9636).
[40] Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788).
[41] Redmon, J., & Farhadi, A. (2017). YOLO9000: better, faster, stronger. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7263-7271).
[42] Redmon, J., & Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767.
[43] Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2117-2125)
[44] Bochkovskiy, A., Wang, C. Y., & Liao, H. Y. M. (2020). Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934.
[45] Zhou, F., Zhao, H., & Nie, Z. (2021, January). Safety helmet detection based on YOLOv5. In 2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA) (pp. 6-11). IEEE.
[46] Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., ... & Wei, X. (2022). YOLOv6: A single-stage object detection framework for industrial applications. arXiv preprint arXiv:2209.02976.
[47] Wang, C. Y., Bochkovskiy, A., & Liao, H. Y. M. (2022). YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv preprint arXiv:2207.02696.
[48] Thrun, S., & Pratt, L. (1998). Learning to learn: Introduction and overview. In Learning to learn (pp. 3-17). Springer, Boston, MA. [40]
[49] Wang, C. Y., Yeh, I. H., & Liao, H. Y. M. (2021). You only learn one representation: Unified network for multiple tasks. arXiv preprint arXiv:2105.04206.
[50] Wang, C. Y., Bochkovskiy, A., & Liao, H. Y. M. (2021). Scaled-yolov4: Scaling cross stage partial network. In Proceedings of the IEEE/cvf conference on computer vision and pattern recognition (pp. 13029-13038).
[51] Dollár, P., Singh, M., & Girshick, R. (2021). Fast and accurate model scaling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 924-932).
[52] Tan, M., & Le, Q. (2019, May). Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning (pp. 6105-6114). PMLR.
[53] Dollár, P., Singh, M., & Girshick, R. (2021). Fast and accurate model scaling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 924-932).
[54] Ge, Z., Liu, S., Wang, F., Li, Z., & Sun, J. (2021). Yolox: Exceeding yolo series in 2021. arXiv preprint arXiv:2107.08430.
[55] Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., ... & Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 10012-10022).
[56] Adegbenjo, A. O. (2019). Non-destructive Assessment of Chicken Egg Fertility Using Hyperspectral Imaging Technique. McGill University (Canada).