Research

Iris biometrics uses the texture of the iris to create a unique personal identifier. The iris image to the near right shows moderate pupil dilation and only minor occlusion of the iris region by the lower eyelid. The image to the far right shows a cosmetic contact lens, which obscures nearly all of the natural iris texture. We have published results dealing with many different aspects of iris image analysis, and provided advice to the government of Somaliland on the use of iris recognition to de-duplicate the voter registration roll.

To the left is an aerial image of a Florida community before Hurricane Dennis in 2005. To the right is an image taken after the storm. Detecting changes based on before- and after-storm images can enable automated and consistent damage assessment. My work in this area is a collaboration with Professor Ahsan Kareem in Civil Engineering. This problem is difficult because the images are not necessarily registered, structures that appear in one image may not appear in the other image, and images taken at different times may have different color properties.

We collected a large biometric dataset from volunteer "identical twins" subjects at the Twins Days Festival in Twinsburg, Ohio. We have worked on identification of identical twins from 2D images, 3D face shape and iris recognition. One unusual result is that while iris recognition does not 'see' the similarity in the iris textures of twins, human observers can tell if a pair of iris images come from identical twins or from unrelated persons.

WNDU news video on biometrics and twins.

Biometrics.
My research group has done extensive work in iris recognition, 2D, 3D and infra-red face recognition, face recognition in video, 2D and 3D ear biometrics, multi-modal biometrics and other topics in biometrics. We have been involved in support of the Human ID Gait Challenge, the Face Recognition Grand Challenge, the Face Recognition Vendor Test 2006, the Iris Challenge Evaluation, and the Multiple Biometric Grand Challenge.

Media Forensics.
My work in this area is focused on creating algorithms that aim to detect if an image is original or has been altered. A simple example of this is to detect whether a face image is original or has been "photoshopped". A more complex example would be to tell how a "fake" image was created from elements of various original images.

Data Mining.
My research in data mining is focused on creating ensembles of classifiers for problems that exhibit "extreme" characteristics such as a high imbalance between classes in the training data, unusually large size of training data, and noise in the class labels of the training data.

My most highly-cited paper introduced the synthetic minority over-sampling technique, or "SMOTE". SMOTE is an approach to dealing with class imbalance in the training data by creating additional synthetic training samples. (code for example SMOTE implementation)

Medical Image Analysis.
I previously worked in several areas of medical image analysis, including detection of signs of cancer in mammograms, algorithms for estimating shunted blood flow in the heart, and CT stereotaxis calculations used by neurosurgeons operating on the brain.

Object Recognition.
I made a series of contributions in two areas of object recognition: object recognition through reasoning about functionality and aspect graphs for object representation.

Publications.

A list of recent and selected publications is available, with links to pdf versions of most. My Google Scholar "h index" is currently in the 70s; see Google Scholar profile.

The Handbook of Iris Recognition (click on the image on the right) is now in its second edition. The first edition was reviewed in ACM Computing Reviews by Creed Jones (review #: CR141221). The closing paragraph of the review - "In the interest of testing this excellent volume, I posed several questions about iris recognition and attempted to find the answers within. I was pleased to see that either relevant information was present and quickly locatable, or there were sufficient references for further study. Since the preface claims that this is "the first book to be devoted entirely to iris recognition," it should be as accurate and complete as a reasonable length allows. It succeeds on both counts. For anyone interested in iris recognition, this book is indispensable."

PhD graduates.
I have so far had the pleasure of mentoring twenty-four wonderful colleagues for their PhD, with more to graduate soon. Two of my PhD graduates, Deborah Thomas and Karen Hollingsworth, appear in the picture to the left, taken at graduation. PhD students currently working with me include Sandipan Banerjee, Aparna Bharati, Joel Brogan, Andrey Kuehlkamp and Rosaura Vidal-Mata.