Fuzzy-appearance manifold and fuzzy nearest distance for face recognition on various poses and degraded images

This paper introduces an approach to recognize face from 3D space on 2D image using fuzzy vector manifolds and nearest distance. We employ fuzzy vector to help the system minimize negative effect coming from noise and image degradation. On the training set, crisp vector representation of images will be transformed to its fuzzy vector representation using a specific triangle fuzzification method. Then, a linear interpolation method will be used to construct a manifold, making the system able to cope with pose variation across data. In the testing phase, we transform every unknown data image to its fuzzy-vector representation using the parameter we obtained from training phase. We then project the unknown fuzzy vector to the manifolds using a technique called fuzzy nearest distance. The output of the system will be the index of manifold that the data mostly belong to, in this case the prediction of person. This system is applied to recognize photos on our databases which some of them are influenced by noises. Experiment result show that the system is able to recognize person on 98% success rate, with a 3% reduction if noises were added.