Expert Conciliation for Multi Modal Person Authentication Systems by Bayesian Statistics
نویسندگان
چکیده
We present an algorithm functioning as a supervisor module in a multi expert decision making machine. It uses the Bayes theory in order to estimate the biases of individual expert opinions. These are then used to calibrate and conciliate expert opinions to one opinion. We present a framework for simulating decision strategies using expert opinions whose properties are easily modiiable. By using real data coming from a person authentication system using image and speech data we were able to connrm that the proposed supervisor improves the quality of individual expert decisions by reaching success rates of 99.5 %.
منابع مشابه
Multi - Modal Person
This paper deals with the elements of a multi-modal person authentication systems. Test procedures for evaluating machine experts as well as machine supervisors based on leave-one-out principle are described. Two independent machine experts on person authentication are presented along with their individual performances. These experts consisted of a face (Gabor features) and a speaker (LPC featu...
متن کاملFusion of Face and Speech Data for Person Identity Verification Fusion of Face and Speech Data for Person Identity Verification
Multi modal person identity authentication is gaining more and more attention in the biometrics area Combining di erent modalities increases the performance and robustness of identity authentication systems The authentication problem is a binary classi cation problem The fusion of di erent modalities can be therefore performed by binary classi ers We propose to evaluate di erent binary classi c...
متن کاملConstruction of a Multi-Modal Database for the Evaluation of Person Authentication Systems operating under Near Infrared Illumination
Nowadays there is a high demand on non-intrusive, user-friendly authentication systems. A strong focus of the research community lies in face recognition systems. Many of the proposed systems operate under visible illumination. There are several databases for training and evaluation of those systems available. However, we claim that operating a face recognition system under near infrared illumi...
متن کاملMHIDCA: Multi Level Hybrid Intrusion Detection and Continuous Authentication for MANET Security
Mobile ad-hoc networks have attracted a great deal of attentions over the past few years. Considering their applications, the security issue has a great significance in them. Security scheme utilization that includes prevention and detection has the worth of consideration. In this paper, a method is presented that includes a multi-level security scheme to identify intrusion by sensors and authe...
متن کاملOn the use of multi-agent systems for the monitoring of industrial systems
The objective of the current paper is to present an intelligent system for complex process monitoring, based on artificial intelligence technologies. This system aims to realize with success all the complex process monitoring tasks that are: detection, diagnosis, identification and reconfiguration. For this purpose, the development of a multi-agent system that combines multiple intelligences su...
متن کامل