Automatic Facial Image Analysis Based On Locality Preserving

Automatic facial image analysis has very large theoretic and practical values.How to extract facial features,implement the algorithm and obtain more efficient results through facial image analysis and recognition,is attracting a large number of researchers to study from multiple fields such as pattern recognition,computer vision,artificial intelligence and neural network.This thesis focuses on locality preserving manifold learning algorithms of facial image analysis and recognition technology.The main research contents of this thesis can be summarized as follows:(1)Research of facial image analysis on the basis of locality preserving manifold learning in spectral clustering experiments.The cropped and normalized face images are used as the experimental dataset,which contains the interference factors such as varying illumination conditions,differing mask conditions,differing facial expressions,varying number of samples and random pixel noise.Based on the spectral clustering experimental results,the effect of interference factors on the locality preserving manifold learning algorithms is discussed.The performance of these manifolds preserving methods in a variety of interference situations is analyzed.Experiments show that some locality preserving manifold learning algorithms have superior robust,such as multiple graph regularized method,elastic net hypergraph learning,-graph learning.(2)Propose a novel collaborative representation based face recognition method.Many face recognition methods have been proposed,among them,the recently proposed collaborative representation based face recognition has attracted the attention of researchers.Many variants and extensions of collaborative representation based classification(CRC)have been presented.However,most of CRC methods do not consider data locality,which is crucial for classification task.In this thesis,a novel collaborative representation based face recognition method,LP-CRC,is proposed,which balances data locality and collaborative representation.The proposed method incorporates locality adaptor term into the robust collaborative representation based classification framework,leading to a novel unified objective function.The Augmented Lagrange Multiplier is used to optimize the objective function.Tests on standard benchmarks demonstrate that the proposed face recognition method is superior to existing methods and robust to noise and outliers.