Research On Infrared Face Recognition Approaches

Infrared face recognition is a technology of identification and classification on infrared face images which are collected by thermal infrared imaging device and processed using the methods of pattern recognition and image processing etc.. In recent years, more attention has been paid to infrared face recognition for its special properties such as illumination invariant and insensitive to disguise.In this paper, eigenface and linear discriminant analysis methods are analyzed deeply based on the characteristics of infrared face images. The recognition performance of current methods are compared with the infrared face databases available. A new infrared face recognition method based on the feature extraction and regularization is proposed for information losing problems exist in linear discriminant analysis methods. This new method can effectively extract the features from infrared face images, and better performance is achieved. Finally, the compressed sensing theory is discussed. The feature regularization and extraction method is combined with compressed sensing for infrared face recognition. Experiment results show this method achieves higher recognition rate compared to other methods.