使用權重鄰近分類法之臉部表情辨識

Recognition of Facial Expressions by Using Weighted Neighbor Classification

林志瀚(Zhi-Han Lin)

指導教授 ： 廖梨君

朝陽科技大學/資訊學院/資訊工程系/碩士(2015年)

摘要

本論文提出一權重鄰近分類法(Weighted Neighbor Classification, WNC)，可快速且正確的辨識人類臉部表情。此方法透過影像處理，於每一張臉部影像擷取出七個特徵點及特徵向量，並根據不同表情所得之依據座標，進行初步的鄰近區域分類。在獲得初步之分類結果後，再依據各表情之主要影響特徵，給予相關特徵向量不同的權重值，用以修正分類誤差，並訓練此辨識系統。經由實驗結果可知，本文所提之權重鄰近分類法，可降低辨識系統之訓練時間及計算複雜度，同時，亦可正確辨識驚訝、快樂、生氣及悲傷四種表情達95%之正確率。

關鍵字

表情辨識 ； 最鄰近分類法 ； 特徵向量 ； 膚色偵測

並列摘要

In this thesis, the weighted neighbor classification (WNC) is proposed to recognize the facial expressions. Each facial image was extracted 7 feature points and 7 feature vectors. The 7 feature vectors of an image were synthesized to be one characterized point. The characterized points were first classified by experienced threshold and compared with the neighbor distances between other characterized points. Based on the results of the first classification, the weightings were assigned to the corresponding feature vectors to modify the characterized points to increase the correct recognition rate. The experimental results demonstrate that the proposed WNC can decrease the training cycles of the recognition system and reduce the computing complexities. The correct recognition rate can be 95% above.

並列關鍵字

recognition of facial expressions ； nearest neighbor classifier ； feature vectors ； skin color detection

參考文獻

[4]H. Wu, Q. Chen and M. Yachida, “Face Detection From Color Images Using a Fuzzy Pattern Matching Method,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, no. 6, pp.557-563, June 1999.

[7]Hui Sung Lee, Jeong Woo Park, MyungJin Chung, “A linear affect–expression space model and control points for mascot-type facial robots,” IEEE Transactions on Robotics, VOL. 23, Issue 5, pp. 863 – 873, October 2007.

[9]M.Pantic and M. Rothkrantz, “Automatic Analysis of Facial Expressions: The State of the Art,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 12, pp. 1424-1445, Dec. 2000.

[10]M. Hamouz, J. Kittler, J.K. Kamarainen, P. Paalanen, H. Kalviainen and J. Matas, “Feature-Based Affine-Invariant Localization of Faces,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 9, pp.1490-1495, September 2005.

[12]Li Zhang, Ming Jiang, DewanFarid, M.A. Hossain, “Intelligent facial emotion recognition and semantic-based topic detection for a humanoid robot, ” Expert Systems with Applications, Vol. 40, Issue 13,pp. 5160–5168, 1 October 2013.