人臉表情自動辨識系統之研製

Development of Automatic Recognition System for Facial Expression

林建成(Chien-Cheng Lin)

指導教授 ： 吳明川

國立臺北科技大學/機電學院/自動化科技研究所/碩士(2006年)

https://doi.org/10.6841/NTUT.2006.00045

摘要

本研究以機器視覺方式偵測人臉表情，有鑑於網路攝影機使用日益普及，比起一般的CCD裝置有價格低廉與重量輕巧的優點，故採用網路攝影機配合個人電腦建構一套表情自動辨識系統。人臉表情主要是以五官人臉動作特徵，表達出內心情緒，而嘴巴的特徵為表情主要指標，所以本研究擷取人臉動作特徵，並建立嘴巴特徵結構線，進行人臉表情辨識。本系統並不需對人臉資訊進行大量的學習訓練，即能辨識一般、高興、驚訝、生氣、傷心等五種表情，成功率初步可達80%以上。同時也針對人臉偵測、五官定位、特徵擷取提出改善的方法進行。本系統未來也可運用在互動式電子寵物、伺服遊樂器、疲勞監測、智慧型教學系統等裝置上。

關鍵字

人臉動作特徵系統 ； 表情辨識 ； 特徵結構線

並列摘要

This research used the machine vision to detect the facial expression. Webcams have an increasing popularity in many applications. The purpose of this study is to develop a facial expression recognition system. The basic techniques of the system are capture the facial expression by webcam, then create the mouth feature structure lines for human facial express recognition based on the images captured. The system are designed to be able to recognize five different types of facial expresses like general motion, happy, surprised, angry, and sad without mass learning. The recognition success rate of the system is expected to reach 80% or higher. It improves the efficiency of human face detection, facial feature location and feature extraction. It is expected that the desired system can also be used in the field of interactive electronic pet, TV game, fatigue monitor, intelligent teaching system.

並列關鍵字

Facial Action Coding System (FACS) ； Facial Expression ； Feature Structure Lines

參考文獻

[12] 黃泰祥，具備人臉追蹤與辨識功能的一個智慧型數位監視系統，私立中原大學電子工程學系，碩士論文，民國93年6月。

[4] T.Kondo and H.Yan, “Automatic human face detection and recognition under non-uniform illumination,” Pattern Recognition Society, no.10, pp. 1707-1718, 1999.

[5] Si-Cheng Zhang and Zhi-Qiang Liu, “A real-time face detector,” IEEE International Conference on Systems, Man and Cybernetics, 2004.

[7] Rein-Lien Hsu, Mohamed Abdel-Mottalab, and Anil K. Jain, “Face Detection in Color Images,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 24, no. 5, 2002.

[8] Son Lam Phung; Chai, D.; Bouzerdoum, A. “A universal and robust human skin color model using neural networks,” International Joint Conference on Neural Networks, vol.4, pp. 2844-2849, 2001.