近數年來旋積類神經網路(Convolutional Neural Network)的在電腦視覺 領域帶起一股旋風，使得此架構幾乎在影像辨識和偵測上幾乎無往不利。目 前可供使用的旋積類神經網路函式庫中，較熱門的有Caffe、TensorFlow 等。
由於本論文未能自行實作出成功的旋積類神經網路，因此列出一些本研究可 再以旋積類神經網路提升水準的方向：
以 CNN 取代 Viola-Jones 方法擷取出人臉 ROI
抓取臉部的準確程度會大幅影響本研究的量化表現與實用性，而CNN 已 經被證實 [26]抓取臉部的表現超越 Viola-Jones 方法。
改用 CNN 與其他表情合併多類訓練
本論文在實測過程中，發現訓練出的良好模型對微笑表情偶爾會有假警 報(false alarm)，雖然本論文已在 PSPI=0 的樣本中加入一些微笑表情樣 本改善此誤差。但更加徹底的做法，是將疼痛表情估測，融入到目前已 經成熟的各種表情辨識中。
50
參考文獻
[1] "Microsoft Project Oxford Emotion API," Microsoft, [Online]. Available:
https://www.projectoxford.ai/emotion. [Accessed 2016].
[2] Theodoridis Sergios, Koutroumbas Konstantinos, in Pattern Recognition (Fourth Edition), Academic Press, 2009, p. 7.
[3] KRIZHEVSKY, Alex; SUTSKEVER, Ilya; HINTON, Geoffrey E.,
"Imagenet classification with deep convolutional neural networks,"
Advances in neural information processing systems, pp. 1097-1105., 2012.
[4] Lucey, P., Cohn, J. F., Prkachin, K. M., Solomon, P. E., Matthews, I.,
"Painful data: The UNBC-McMaster shoulder pain expression archive database," in Automatic Face & Gesture Recognition and Workshops, 2011 IEEE International Conference on (pp. 57-64), 2011.
[5] "Facial Action Coding System," Wikimedia Foundation, Inc, [Online].
Available: https://en.wikipedia.org/wiki/Facial_Action_Coding_System.
[Accessed 25 2 2016].
[6] Prkachin, K. M., Solomon, P. E., "The structure, reliability and validity of pain expression: Evidence from patients with shoulder pain," Pain, 139(2), pp. 267-274, 2008.
[7] Lyons, M. J., Akamatsu, S., Kamachi, M., Gyoba, J., Budynek, J., "The Japanese female facial expression (JAFFE) database," 1998. [Online].
Available: http://www. kasrl, org/jaffe.html.
[8] Chen, Li-Fen, and Yu-Shiuan Yen, "Taiwanese facial expression image
51
database, Brain Mapping Laboratory, Institute of Brain Science, National Yang-Ming University, Taipei, Taiwan," 2007. [Online]. Available:
http://bml.ym.edu.tw/tfeid/. [Accessed 16 6 2016].
[9] Loris Nannia, Alessandra Luminia, Sheryl Brahnam, "Local binary patterns variants as texture descriptors for medical image analysis," Artificial
intelligence in medicine 49(2), pp. 117-125.
[10] Gholami, Behnood, Wassim M. Haddad, and Allen R. Tannenbaum.,
"Agitation and pain assessment using digital imaging," in IEEE
Engineering in Medicine and Biology Society. Conference (Vol. 2009), 2009.
[11] Kaltwang Sebastian, Ognjen Rudovic, and Maja Pantic, "Continuous pain intensity estimation from facial expressions," Advances in Visual
Computing, pp. 368-377, 2012.
[12] 王佳琪, Recognition of Painful Facial Expression using Multiple Kernel Learning, 臺灣新竹: 國立清華大學電機工程學系碩士論文, 2011.
[13] MLA Chen, Junkai, Zheru Chi, and Hong Fu., "A new approach for pain event detection in video," in Affective Computing and Intelligent
Interaction (ACII), International Conference, 2015.
[14] Florea, C., Florea, L., Boia, R., Bandrabur, A., Vertan, C., "Pain Intensity Estimation by a Self--Taught Selection of Histograms of Topographical Features," 26 5 2016. [Online]. Available: http://arxiv.org/abs/1503.07706.
[15] 周揚賀, 基於深度摺積神經網路之影像檢索技術, 臺灣桃園: 國立中央 大學通訊工程學系碩士論文, 2015.
52
[16] "CS231n Convolutional Neural Networks for Visual Recognition,"
[Online]. Available: http://cs231n.github.io/neural-networks-3/. [Accessed 15 6 2016].
[17] 演算法筆記, "Regression," [Online]. Available:
http://www.csie.ntnu.edu.tw/~u91029/Regression.html. [Accessed 26 6 2016].
[18] Mu Li, Tong Zhang, Yuqiang Chen, Alexander J. Smola, "Efficient mini-batch training for stochastic optimization," in Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data
mining, 2014.
[19] McCaffrey, James, "How To Standardize Data for Neural Networks," 15 1 2014. [Online]. Available:
https://visualstudiomagazine.com/Articles/2014/01/01/How-To-Standardize-Data-for-Neural-Networks.aspx?Page=2. [Accessed 31 5 2016].
[20] Liu, P. H., Su, S. F., Chen, M. C., Hsiao, C. C., "Deep learning and its application to general image classification," in Informative and Cybernetics for Computational Social Systems (ICCSS), 2015.
[21] Paul Viola, Michael J. Jones, "Robust Real-Time Face Detection,"
International Journal of Computer Vision, pp. 137-154, 5 2004.
[22] 廖文宏, "基於三元化樣式的通用型區域特徵描述方法," 國立政治大學 資訊科學系行政院國家科學委員會專題研究計畫期末報告, 臺灣台北, 2011.
[23] T. Ojala, M. Pietikainen and T. Maenpaa,, "Multi-resolution Gray-scale and
53
Rotation Invariant Texture Classification with Local Binary Patterns," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24(7), pp.
971-987, July 2002.
[24] "Emgu CV: OpenCV in .NET (C#, VB, C++ and more)," [Online].
Available: http://www.emgu.com/wiki/index.php/Main_Page.
[25] 周志成, "相關係數," [Online]. Available:
https://ccjou.wordpress.com/2011/03/09/相關係數/. [Accessed 10 6 2016].
[26] Hong, X., Zhao, G., Zafeiriou, S., Pantic, M., Pietikäinen, M., "Capturing correlations of local features for image representation," Neurocomputing, pp. 99-106, 2016.
[27] H. e. a. Li, "A convolutional neural network cascade for face detection," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015.