You are granted permission for the non-commercial reproduction, distribution, display, and performance of this technical report in any format. However, this permission is only for a period of 45 (forty-five) days from the most recent time that you verified that this technical report is still available from the Department of Computer Science of the University of Maryland at College Park under terms that include this permission. All other rights are reserved by the author(s).
Rama Chellappa. C.L. Wilson. S. Sirohey. C.S. Barnes. Human and Machine Recognition of Faces: A Survey. August 1994.
The goal of this paper is to present a critical survey of the literature on human and machine recognition of faces. Machine recognition of faces has several applications ranging from static matching of controlled photographs as in mugshot matching and credit card verification to surveillance video images. These applications have different constraints in terms of the complexity of their processing requirements and thus present a wide range of technical challenges. Over the last twenty years researchers in psychophysics, neural sciences and engineering, image processing, analy sis and computer vision have investigated a number of issues related to face recognition by humans and machines. The ongoing research activities have been given renewed emphasis over the last five years. The existing techniques and systems have been tested on different sets of images of varying complexities. But very little synergism exists between studies in psychophysics and Engineering literature. Most importantly, there exist no evaluation or benchmarking studies using large databases with the image quality that arises in law enforcement/commercial applications. In this paper, we first present different applications of face recognition in the law enforcement and commercial sectors. Special constraints that are present in these applications are pointed out. This is followed by a brief overview of the literature o n face recognition in the psychophysics community. We then present a detailed overview of more than twenty years of research done in the engineering community. Techniques for segmentation/location of the face, feature extraction and recognition are review ed Global transform and feature based methods using statistical, structural and neural classifiers are summarized. A brief summary of recognition using face profiles and range image data is also given. Real-time recognition from video images acquired in a cluttered scene such as an airport is probably the most challenging face recognition problem. As not much has been reported on this problem, we discuss several existing technologies in the image under standing literature that could potentially impact this problem. Given the numerous theories and techniques that are applicable to face recognition, it is clear that evaluation and benchmarking of these algorithms is crucial. We discuss relevant issues such as data collection, performance metrics and evaluation of systems and techniques. Finally, a summary and conclusions are given. The postscript version of this TR is available from the Center for Automation Research via anonymous ftp at ftp.cfar.umd.edu; or via the WWW at http://www.cfar.umd.edu/CfAR/TRs. (Also cross-referenced as CAR-TR-731) Department of Computer Science, University of Maryland, Center for Automation Research,
(Also cross-referenced as CAR-TR-691) October 1993.
Evaluation of Pattern Classifiers for Fingerprint and OCR Applications. J.L. Blue. G.T. Candela. P.J. Grother. Rama Chellappa. C.L. Wilson. Computer Vision Laboratory, Center for Automation Research, Department of Computer Science, University of Maryland, In this paper we evaluate the classification accuracy of four statistical and three neural network classifiers for two image based pattern classification problems. These are fingerprint classification and optical character recognition (OCR) for isolated handprinted digits. The evaluation results reported here should be useful for designers of practical systems for these two important commercial applications. For the OCR problem, the Karhunen-Loeve (K-L) transform of the images is used to generate the inp ut feature set. Similarly for the fingerprint problem, the K-L transform of the ridge directions is used to generate the input feature set. The statistical classifiers used were Euclidean minimum distance, quadratic minimum distance, normal, and knearest neighbor. The neural network classifiers used were multilayer perceptron, radial basis function, and probabilistic. The OCR data consisted of 7,480 digit images for training and 23,140 digit images for testing. The fingerprint data consisted of 9,000 trai ning and 2,000 testing images. In addition to evaluation for accuracy, the multilayer perceptron and radial basis function networks were evaluated for size and generalization capability. For the evaluated datasets the best accuracy obtained for either pro blem was provided by the probabilistic neural network, where the minimum classification error was 2.5% for OCR and 7.2% for fingerprints.
Last Generated Fri Aug 11 04:01:01 EDT 2000