Provided are an apparatus and a method for generating a complex feature vector based on discriminant analysis for face recognition to generate a complex feature vector by using local features extracted from a face image consisting of pixels extracted based on discriminant information of local features to which weight is applied by using global features extracted from a whole face image and a ReliefF method. The apparatus for generating a complex feature vector secures a higher recognition rate than when using the global features and local features independently and when simply combining the global features and the local features in all databases used in face recognition, and increases a face recognition rate in a database having lighting variation compared to an existing apparatus.;COPYRIGHT KIPO 2017
展开▼