Using Particle Swarm Optimisation for Elastic Bunch Graph Matching to Recognise Faces

A new approach to face recognition, where elastic bunch graph matching technique optimised with particle swarm optimisation is proposed in this paper. It is a fully automatic algorithm and can be used for databases where only one image per person is available. A face is represented by a face graph. A face bunch graph is created as a generalised representation of faces of various individuals. The face graph similarity between the face bunch graph and the deformable graph that has to be fitted to the face in the image is maximised by particle swarm optimisation, to locate the landmarks and thereby to find the optimal graph to represent a face. Improved results were obtained by this method as apposed to elastic bunch graph matching without optimisation. Results were further improved by hybridising it with principal component analysis.