Application of DSmT in Facial Expression Recognition
Chengzhi Zhao, Jinyu Wei, Z. G. Xing, Ziqi Wei · 2012
As the sample in facial expression database is small, the influence of the environment and the actual expression image processing can cause face feature information uncertainty and conflict of different information. This paper presents how to solve the small sample problem and the fusion of global feature recognition results and local feature recognition results based on DSmT (Dezert-Smarandache Theory) by matlab, the results show that DSmT can better handle the face expression of uncertainty information and contradictory information than DST (Dempster-Shafer Theory), recognition effect has better performance.