Abstract
Biometric recognition refers to the process of recognizing a person’s identity using physiological or behavioral modalities, such as face, voice, fingerprint, gait, etc. Such biometric modalities are mostly used in recognition tasks separately as in unimodal systems, or jointly with two or more as in multimodal systems. However, multimodal systems can usually enhance the recognition performance over unimodal systems by integrating the biometric data of multiple modalities at different fusion levels. Despite this enhancement, in real-life applications some factors degrade multimodal systems’ performance, such as occlusion, face poses, and noise in voice data. In this paper, we propose two algorithms that effectively apply dynamic fusion at feature level based on the data quality of multimodal biometrics. The proposed algorithms attempt to minimize the negative influence of confusing and low-quality features by either exclusion or weight reduction to achieve better recognition performance. The proposed dynamic fusion was achieved using face and voice biometrics, where face features were extracted using principal component analysis (PCA), and Gabor filters separately, whilst voice features were extracted using Mel-Frequency Cepstral Coefficients (MFCCs). Here, the facial data quality assessment of face images is mainly based on the existence of occlusion, whereas the assessment of voice data quality is substantially based on the calculation of signal to noise ratio (SNR) as per the existence of noise. To evaluate the performance of the proposed algorithms, several experiments were conducted using two combinations of three different databases, AR database, and the extended Yale Face Database B for face images, in addition to VOiCES database for voice data. The obtained results show that both proposed dynamic fusion algorithms attain improved performance and offer more advantages in identification and verification over not only the standard unimodal algorithms but also the multimodal algorithms using standard fusion methods.
Highlights
Biometrics has long been known as a robust approach for person recognition that uses different physiological or behavioral traits, such as face, voice, fingerprint, iris, gait, and many others [1]
In some systems feature level fusion is more difficult to implement than other fusion levels, because of the relationship between the features spaces of different biometric systems may not be known and concatenating two feature vectors might lead to the dimensionality problem
Kawakami et al [24] proposed a method for speaker identification in a noisy environment and investigated the effect of pitch synchronous phase information when combined with Mel-Frequency Cepstral Coefficients (MFCCs) for speaker identification by combining Gaussian Mixture Model (GMM) based on MFCC with GMM based on phase information
Summary
Biometrics has long been known as a robust approach for person recognition that uses different physiological or behavioral traits, such as face, voice, fingerprint, iris, gait, and many others [1]. The greatest majority of existing real-life biometric systems are unimodal, which means they make use of a single biometric modality and need to be accurately enrolled in a database to train a discriminative algorithm, and to be sufficiently acceptable and usable in the recaptured probe or test samples for achieving successful recognition Such systems mostly suffer from different limitations against some challenges, such as noise in sensed data, intra-class variation, inter-class similarity, and non-universality [2]. Motivated by the challenging context of recognizing low-quality face and voice probe data besides the increased and urgent needs for developing such robust capabilities, we propose two dynamic featurelevel fusion algorithms for improved and adaptive multimodal biometric identification/verification. An investigation of the effects and performance of using the proposed dynamic feature level fusion of face and voice biometrics for person identification and verification tasks.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
