Feature Level Fusion of Face and voice Biometrics systems using Artificial Neural Network for personal recognition

Abstract

Lately, human recognition and identification has acquired much more attention than it had before, due to the fact that computer science nowadays is offering lots of alternatives to solve this problem, aiming to achieve the best security levels. One way is to fuse different modalities as face, voice, fingerprint and other biometric identifiers. The topics of computer vision and machine learning have recently become the state-of-the-art techniques when it comes to solving problems that involve huge amounts of data. One emerging concept is Artificial Neural networks. In this work, we have used both human face and voice to design a Multibiometric recognition system, the fusion is done at the feature level with three different schemes namely, concatenation of pre-normalized features, merging normalized features and multiplication of features extracted from faces and voices. The classification is performed by the means of an Artificial Neural Network. The system performances are to be assessed and compared with the Knearest-neighbor classifier as well as recent studies done on the subject. An analysis of the results is carried out on the basis Recognition Rates and Equal Error Rates. Povzetek: Z nevronsko mrežo so kombinirani obraz in glas za biometrično identifikacijo.

Damaged Fingerprint Recognition by Convolutional Long Short-Term Memory Networks for Forensic Purposes

“…Comparative models have proven their performance in similar fields such as scene text recognition [21], dynamic gesture recognition [22], hyperspectral image classification [23], emotion recognition [24], event surveillance [25], traffic forecasting [26], air quality prediction [27], facial expression recognition [28], etc. These good results encourage us to explore the recognition, by comparable networks, of other biometric forms such as hand or earlobe morphology, retinal and iris physiognomy [29], voice [30,31], and signatures [32], especially those that have been forged or falsified or altered [33]. These biometric patterns play as crucial a role as fingerprints in forensic investigative procedures and in establishing proof of innocence or indictment.…”

Intelligent course recommendation based on neural network for innovation and entrepreneurship education of college students

“…ANN is characterized by a strong self-learning capability that enables the algorithm to achieve higher accuracy [12]. It has an extensive application in fields such as face recognition [13]. The algorithm first performs weighted summation on all the signals before processing the previous input, and the calculation formula is:…”

Dynamic Audio-Visual Biometric Fusion for Person Recognition

“…Thus, matching score fusion achieved 0.62% equal error rate (EER), whereas feature fusion achieved 2.81% EER. In [29], a face-voice multimodal recognition approach was proposed and a number of experiments were conducted in three feature fusion mechanisms, concatenation of pre-normalized features, merging normalized features, and multiplication of features. The results showed that the merging fusion was the most effective mechanism.…”