Semester
Fall
Date of Graduation
2010
Document Type
Thesis
Degree Type
MS
College
Statler College of Engineering and Mineral Resources
Department
Lane Department of Computer Science and Electrical Engineering
Committee Chair
Donald A Adjeroh
Abstract
Biometrics is now a more established and sophisticated field. It studies the use of physiological characteristics or behavioral traits to identify and recognize a person automatically. It facilitates theft control and increased security.;Metrology has been one of the well-studied topics in computer vision. Absolute measurement values of humans can be obtained from a fully calibrated camera. These measurements are stored as a database and studied in detail to assess their significance as a biometric. In this thesis we want to assess the performance of human body measurements as a soft biometric. Every human has distinct biometric characters. They can be classified using biometric measurements.;Here the performance of Biometric systems is measured empirically without explicitly measuring the available information. We make use of soft biometric traits like height, weight, gender, age to measure the discriminative capability of metrology in human recognition. Analysis of human body measurements can be applied in various domains like video surveillance, video retrieval, human-computer interaction systems, and medical diagnosis. We establish the performance of human metrology in distinguishing between humans using a database of such measurements. We characterize the performance using measures such as distance plots, precision and recall, genuine acceptance rate (GAR), and false acceptance rate (FAR).
Recommended Citation
Sepuri, Rakesh, "Measuring the Discriminative Capability of Metrology in Human Recognition" (2010). Graduate Theses, Dissertations, and Problem Reports. 4654.
https://researchrepository.wvu.edu/etd/4654