Multi-Modal Biometrics Systems: Concepts, Strengths, Challenges and Solutions

Authentication is a very important aspect of computer security. Most systems employ strategies such as Password-based authentication, Multi-factor authentication, Certificate based authentication which are accompanied with a lot ofchallenges. To address this issue, most security systems have introduced the use of biometrics for authentication. Unimodal biometrics systems have many limitations regarding performance and accuracy. The use of Multimodal biometrics systems for authentication is recently attracting the attention of researchers due to its capacity to overcome most of the drawbacks of Unimodal biometric systems. This paper focuses on the use of biometric technology for authentication. The strengths of multimodal biometric systems, together with the challenges of multimodal biometric systems are presented. The paper also suggests solutions to the challenges of multimodal biometric systems.

Enhancing the Accuracy of Multimodal Biometric Systems

: Biometric acknowledgment frameworks have progressed altogether in the most recent decade and their utilization in explicit applications will increment sooner rather than later. The capacity to direct important correlations and evaluations will be urgent to fruitful organization and expanding biometric selection. Indeed, even the best methodology and unimodal biometric frameworks couldn't completely address the issue of exactness and execution as far as their bogus acknowledge rate (FAR) and bogus oddball rate (FRR). In spite of the fact that multimodal biometric frameworks had the option to moderate a portion of the restrictions experienced in unimodal biometric frameworks, like non-all inclusiveness, uniqueness, non-adequacy, loud sensor information, parody assaults, and execution, the issue of low exactness actually continues. In this paper, we survey research papers zeroed in on the precision improvement in data combination of face and finger impression biometric acknowledgment frameworks, decide the primary highlights of the chose techniques, and afterward call attention to their benefits and inadequacies. We propose a novel methodology in relieving the issue of exactness and execution of data combination of multimodal biometric frameworks. This methodology utilizes multilayer perceptron neural organizations in preparing and testing of the organization while additionally proposing the utilization of the most well-known utilized unique mark in biometric field.

CNN-based multimodal touchless biometric recognition system using Gait and speech

Person identification using biometric features is an effective method for recognizing and authenticating the identity of a person. Multimodal biometric systems combine different biometric modalities in order to make better predictions as well as for achieving increased robustness. This paper proposes a touchless multimodal person identification model using deep learning techniques by combining the gait and speech modalities. Separate pipelines for both the modalities were developed using Convolutional Neural Networks. The paper also explores various fusion strategies for combining the two pipelines and shows how various metrics get affected with different fusion strategies. Results show that weighted average and product fusion rules work best for the data used in the experiments.

Contemporary Biometric System Design

Individual's Identity Authentication depends on physical traits like face, iris, and fingerprint, etc., or behavioral traits like voice and signature. With the rapid advancement in the field of biometrics, multimodal biometric systems are replacing unimodal biometric systems. As the application of molecular biometric system removes certain errors like noisy data, interclass variations, spoof attacks, and unacceptable error rates as compared to unimodal biometric systems. Even the possibilities of multiple scenarios present in multimodal biometric systems are quite helpful for the consolidation of information using different levels of fusion. In this chapter, the authors try to analyze the technological change which is present due to growing field of biometrics with artificial intelligence and undergone a thorough research for multimodal biometric systems for effective authentication purpose. This study is quite helpful for getting different perception for the use of biometrics as a highest level of network security due to the fusion of many different modalities.