Face Recognition and Super-Resolution of Low-Resolution Faces

In this work we introduce a new framework to face recognition that uses super-resolution techniques to enchance classification performance of low-resolution faces. As we study the different issues of classification when the input sample is of lower resolution than the images in the training set used, we propose guidelens for the use of super-resolution methods with the purpose of improving face recognition.

Wavelet Packet Correlation Filter Classifiers

We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimizing some criterion function of the image training set. Instead we perform classification in wavelet spaces that have training set representations which provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object recognition task.

Palmprint Recognition Using Correlation Filter Classifiers

This study introduces the application of correlation filter classifiers for palmprint identification and verification. Correlation filter classifiers have been previously applied to other biometric classification tasks, but not to classification of palmprint images. We discuss how the extraction of an appropriate region of interest in the palmprint surface can be used to design correlation filters that accomplish very high levels of accuracy (for example, we have shown that for a database of 50 persons it is possible to achieve perfect separation of authentic and impostor scores).

Steganography for Reduced-Complexity Correlation Filter Classifiers

This study introduces an application of steganography for hiding cancelable biometric data based on quad-phase correlation filter classification. The proposed techniques can perform two tasks: (1) embed an encrypted (cancelable) template for biometric recognition into a host image or (2) embed the biometric data required for remote (or later) classification, such as embedding a transformed face image into the host image, so that it can be transmitted for remote authentication or stored for later use.