作者: H.Y. Huang , S.H. Hsu
DOI:
关键词:
摘要: For face representation, the illumination effect is an important factor, in order to reduce we propose a facial feature location method based on improved active shape model (IASM) drawback of intensity contrast and enhance edge information. First, images are preprocessed by proposed normalization using Gabor wavelets. Then, features can fit more efficient fast kernels reduction feature-based weighted warping method. Based our method, it not only obtains better alignment but also overcomes (ASM) which caused failure for aligned target. Experimental results show that this approach achieve JAFFE database with seven type expressions Yale database_B varied conditions.