INPUT
Angry Disgust Fear Happy Sad Surprise Other Total
Correct 5 5 5 5 5 5 13 43
Wrong
0 0 0 0 0 0 2 2Total
5 5 5 5 5 5 15 45Recognition Rate
100% 100% 100% 100% 100% 100% 86.67% 95.56%
表 4-8、人臉表情強度辨識與問卷比較結果(1 Stdev)
INPUT
Angry Disgust Fear Happy Sad Surprise Other Total
Correct 3 4 5 5 4 4 11 36
員，在經過一一詢問後得到了總共 25 人的協助。下面列出全部的測試與問卷調 查結果，其中 Stdev 項為標準差。
圖 4-6、混合表情測試圖像(1)
表 4-9、混合表情測試辨識輸出結果(1)
Angry Disgust Fear Happy Sadness Surprise
0% 50% 0% 0% 50% 0%
表 4-10、混合表情測試問卷調查結果(1)
Angry Disgust Fear Happy Sadness Surprise
Average 23.2% 37.6% 0% 0% 39.2% 0%
Stdev 26.1% 29.45% 0% 0% 39.26% 0%
圖 4-7、混合表情測試圖像(2)
表 4-11、混合表情測試辨識輸出結果(2)
Angry Disgust Fear Happy Sadness Surprise
0% 0% 80.1% 0% 19.9% 0%
表 4-12、混合表情測試問卷調查結果(2)
Angry Disgust Fear Happy Sadness Surprise Average 6.8% 19.6% 38% 0% 32.8% 2.8%
Stdev 17.01% 21.89% 41.33% 0% 33.85% 8.43%
圖 4-8、混合表情測試圖像(3)
表 4-13、混合表情測試辨識輸出結果(3)
Angry Disgust Fear Happy Sadness Surprise
70% 30% 0% 0% 0% 0%
表 4-14、混合表情測試問卷調查結果(3)
Angry Disgust Fear Happy Sadness Surprise Average 61.2% 36% 0.8% 0.4% 1.2% 0.4%
Stdev 29.91% 30.55% 2.77% 2% 4.4% 2%
圖 4-9、混合表情測試圖像(4)
表 4-15、混合表情測試辨識輸出結果(4)
Angry Disgust Fear Happy Sadness Surprise
0% 0% 50% 0% 50% 0%
表 4-16、混合表情測試問卷調查結果(4)
Angry Disgust Fear Happy Sadness Surprise Average 1.2% 14.2% 43.6% 2.4% 38.6% 0%
Stdev 3.32% 20.29% 39.36% 10.12% 41.22% 0%
圖 4-10、混合表情測試圖像(5)
表 4-17、混合表情測試辨識輸出結果(5)
Angry Disgust Fear Happy Sadness Surprise
100% 0% 0% 0% 0% 0%
表 4-18、混合表情測試問卷調查結果(5)
Angry Disgust Fear Happy Sadness Surprise
Average 55.4% 31.6% 2.8% 0% 7% 2.8%
Stdev 33.97% 34.6% 8.91% 0% 14.29% 10.61%
圖 4-11、混合表情測試圖像(6)
表 4-19、混合表情測試辨識輸出結果(6)
Angry Disgust Fear Happy Sadness Surprise
0% 0% 0% 0% 100% 0%
表 4-20、混合表情測試問卷調查結果(6)
Angry Disgust Fear Happy Sadness Surprise
Average 16.2% 31.8% 2% 0% 50% 0%
Stdev 25.71% 37.83% 7.07% 0% 42.91% 0%
表 4 Angry Disgust
0% 0%
表 4 Angry Average 0%
Stdev 0%
圖 4-12、混合表情測試圖像(7)
-21、混合表情測試辨識輸出結果(7) Fear Happy Sadness
50% 0% 0%
-22、混合表情測試問卷調查結果(7) Disgust Fear Happy Sadness
2.8% 21.6% 2.8% 0.4%
6.78% 22.49% 10.61% 2%
Surprise 50%
Sadness Surprise 72%
21.21%
表 4 Angry Disgust
0% 0%
表 4 Angry Average 0.4%
Stdev 2%
圖 4-13、混合表情測試圖像(8)
-23、混合表情測試辨識輸出結果(8) Fear Happy Sadness
0% 55.7% 0%
-24、混合表情測試問卷調查結果(8) Disgust Fear Happy Sadness
1.2% 2.8% 37.8% 2.4%
3.32% 10.21% 34.94% 12%
Surprise 44.3%
Sadness Surprise 55.4%
32.4%
4.6 結論 結論 結論 結論與討論 與討論 與討論 與討論
在辨識基本表情測試時，由表 4-1 之結果可以看出辨識系統可以準確的辨認 出基本表情的情況。在設定判定標準為最高比例正確且大於 50%並且第二高比例 不大於 30%時，表 4-1 的平均辨識率可達 93.33%。與其他使用 Cohn-Kanade 資 料庫的研究[34]相比較，其他研究[34]的平均辨識率為 88.9%，比本研究的平均 辨識率 93.33%略低。
在辨識表情強度測試時，由表 4-5 可以觀察到在以測試結果與問卷調查結果 在 1.5 個標準差之內作為判斷標準之時，平均辨識率可達 95.56%。在以測試結果 與問卷調查結果在 1 個標準差之內作為判斷標準之時，如表 4-6 所示，平均辨識 率仍有 80%。
在辨識混合表情比例測試時，由問卷調查中標準差的數值可以得知，每個人 對表情比例的判斷頗有分歧，但我們比較問卷結果的平均與系統輸出可以觀察到 問卷結果與辨識系統輸出有明確的正相關性。
第五章
合比例與表情強度。這套系統利用主動外觀模型(Active Appearance Model, AAM) 擷取出人臉特徵值，接著使用我們所提出的根據面部動作編碼組合與表情關聯性參考文獻 參考文獻 參考文獻 參考文獻
[1]內政部戶政司全球資訊網(http://www.ris.gov.tw/ch4/static/y1s400000.xls) [2] Meng-Ju Han, Chia-How Lin and Kai-Tai Song, “Autonomous Emotional
Expression Generation of a Robotic Face,” in Proc. of IEEE SMC 2009, St Antonio, TX, USA, 2009, pp. 2501-2506.
[3] Peng Yang, Qingshan Liu, ,Xinyi Cui and Dimitris N.Metaxas, “Facial Expression Recognition Using Encoded Dynamic Features,” in Proc. of IEEE Computer
Society Conference on Computer Vision and Pattern Recognition, 2008,
Anchorage, Alaska, USA, pp.1-8.[4] Maria Pateraki, Haris Baltzakis, Polychronis Kondaxakis and Panos Trahanias,
“Tracking of Facial Features to Support Human-Robot Interaction,” in Proc. of
IEEE International Conference on Robotics and Automation, 2009, Kobe, Japan,
pp.3755-3760.[5] 陳奕彣,
人臉辨識及表情辨識之整合設計
,碩士論文, 國立交通大學電機與 控制工程學系, 2010.[6] A.M. Martinez and R. Benavente, The AR Face Database, CVC Technical Report
#24, June 1998.
[7] Akshay Asthana,Roland Goecke,Novi Quadrianto and Tom Gedeon, “Learning Based Automatic Face Annotation for Arbitrary Poses and Expressions from Frontal Images Only,” in Proc. of IEEE Computer Society Conference on
Computer Vision and Pattern Recognition, 2009, Miami, Florida, USA,
pp.1635-1642.[8] Michel Valstar, Brais Martinez and Xavier Binefa, “Facial Point Detection using Boosted Regression and Graph Models,” in Proc. of IEEE Computer Society
Conference on Computer Vision and Pattern Recognition, 2010, San Francisco,
CA, USA, pp. 2729-2736.
[9] Liya Ding and Aleix M. Martinez, “Precise Detailed Detection of Faces and Facial Features,” in Proc. of IEEE Computer Society Conference on Computer Vision
and Pattern Recognition, 2008, Anchorage, Alaska, USA, pp.1-7.
[10] Paul Ekman and Wallace V. Friesen, Unmasking the face, Prentice-Hill Inc. , Englewood Cliffs, New Jersey, USA, 1975.
[11] Peng Yang, Qingshan Liu and Dimitris N. Metaxas, “Exploring Facial
Expressions with Compositional Features,” in Proc. of IEEE Computer Society
Conference on Computer Vision and Pattern Recognition, 2010, San Francisco,
CA, USA, pp. 2638-2644.[12] Peng Yang, Qingshan Liu and Dimitris N. Metaxas, “Boosting Coded Dynamic Features for Facial Action Units and Facial Expression Recognition,” in Proc. of
IEEE Computer Society Conference on Computer Vision and Pattern
Recognition, 2007, Minneapolis, Minnesota, USA, pp. 1-6.
[13] Zhi Zhang, Vartika Singh, Thomas E. Slowe, Sergey Tulyakov, and Venugopal Govindaraju, “Real-time Automatic Deceit Detection from Involuntary Facial Expressions,” in Proc. of IEEE Computer Society Conference on Computer
Vision and Pattern Recognition, 2007, Minneapolis, Minnesota, USA, pp. 1-6.
[14] Kai-Yueh Chang, Tyng-Luh Liu and Shang-Hong Lai, “Learning
Partially-Observed Hidden Conditional Random Fields for Facial Expression Recognition,” in Proc. of IEEE Computer Society Conference on Computer
Vision and Pattern Recognition, 2009, Miami, Florida, USA, pp.1-6.
[15] Feng Zhou, Fernando De la Torrey and Jeffrey F. Cohnz, “ Unsupervised Discovery of Facial Events,” in Proc. of IEEE Computer Society Conference on
Computer Vision and Pattern Recognition, 2010, San Francisco, CA, USA, pp.
2574-2581.
[16] Tomas Simon, Minh Hoai Nguyen, Fernando De La Torre and Jeffrey F. Cohn, “ Action Unit Detection with Segment-based SVMs,” in Proc. of IEEE
Computer Society Conference on Computer Vision and Pattern Recognition,
2010, San Francisco, CA, USA, pp. 2737-2744.[17] Lifeng Shang and Kwok-Ping Chan, “ Nonparametric Discriminant HMM and Application to Facial Expression Recognition,” in Proc. of IEEE Computer
Society Conference on Computer Vision and Pattern Recognition, 2009, Miami,
Florida, USA, pp. 2090-2096.[18] Paul Viola and Michael Jones, “ Robust Real-time Face Detection,” in Proc.of
the Eighth International Conference On Computer Vision, 2001, Vancouver,
British Columbia, Canada, pp.747.[19] Yu-Li Xue, Xia Mao, and Fan Zhang, “Beihang University Facial Expression Database and Multiple Facial Expression Recognition,” in Proc.of the Fifth
International Conference on Machine Learning and Cybernetics, 2006, Dalian,
China, pp. 3282 – 3287.[20] 莊蕙如,
萃取表情重要特徵進行表情辨識與表情強度分析
,碩士論文, 國立清 華大學資訊工程學系, 2009.[21] Hui Zhao, Zhiliang Wang, Jihui Men, “Facial Complex Expression Recognition Based on Fuzzy Kernel Clustering and Support Vector Machines,” in Proc.of the
Third International Conference on Natural Computation, 2007, Haikou, Hainan,
China, pp.562-566.[22] Paul Viola and Michael Jones, “Rapid Object Detection Using a Boosted
Cascade of Simple Features,” in Proc. of IEEE Computer Society Conference on
Computer Vision and Pattern Recognition, 2001, Kauai Marriott, Hawaii, USA,
pp. I-511-I-518.[23] Rainer Lienhart and Jochen Maydt. “An Extended Set of Haar-like Features for
Rapid Object Detection,” in Proc. of International Conference on Image
Processing, 2002, Rochester, New York, USA, pp. 900-903
[24] C. C. Chiang, W. K. Tai, M. T. Yang, Y. T. Huang and C. J. Huang, “A Novel Method for Detecting Lips, Eyes and Faces in Real Time,” Real-Time Imaging, vol. 9, no. 4, pp. 277-287, 2003.
[25]T. F. Cootes, G. J. Edwards and C. J. Taylor, “Active appearance models.” in Proc.
of 5th European Conference on Computer Vision, 1998, Springer, Berlin, pp
484–498.[26] T. F. Cootes, G. J. Edwards and C. J. Taylor, “Active Appearance Models,” In
IEEE Transactions on Pattern Analysis And Machine Intelligence, Vol. 23, No. 6,
2001, pp. 681-685[27] T. F. Cootes and C. J. Taylor, Technical Report: Statistical Models of Appearance
for Computer Vision, The University of Manchester, School of Medicine, 2004.
[28] I. Matthews and S. Baker, “Active Appearance Models Revisited,” International
Journal of Computer Vision, vol. 60, no.2, 2004, pp.135-164.
[29] C. Goodall, “Procrustes Methods in the Statistical Analysis of Shape,” Journal of
the Royal Statistical Society B, Vol 53, no.2, pp.285-339, 1991.
[30] J.R. Shewchuk, “Triangle: Engineering a 2D Quality Mesh Generator and Delaunay Triangulator,” In Applied Computational Geometry, FCRC'96
Workshop, pp. 203-222. Springer-Verlag, 1996.
[31] I. Matthews and S. Baker, “Lucas-Kanade 20 Years On: A Unifying Framework,”
International Journal of Computer Vision, Vol. 56, No. 3, 2004, pp. 221 - 255.
[32] 羅華強,
類神經網路－ MATLAB 的應用
, 高立圖書, 2005.[33 ] Kanade, T., Cohn, J. F., & Tian, Y. (2000). Comprehensive database for facial expression analysis. Proceedings of the Fourth IEEE International Conference
on Automatic Face and Gesture Recognition (FG'00), Grenoble, France, 46-53.
[34] M. Obaid, R. Mukundany, R. Goeckezx, M. Billinghurst and H. Seichter “A Quadratic Deformation Model for Facial Expression Recognition” in Proc. of Digital Image Computing: Techniques and Applications, 2009, Melbourne, Australia, pp. 264-270.