Implementation of an access control system based on bimodal biometrics with fusion of global characteristics: application to facial recognition and fingerprints / Bopatriciat Boluma Mangata ... [et al.]

• Main Authors: Mangata, Bopatriciat Boluma, Dominique, Ilunga N’kashama, Ryan, Tebua Tene Patience, Parfum, Bukanga Christian
• Format: Article
• Language: English
• Published: Universiti Teknologi MARA, Perlis 2022
• Subjects: Detectors. Sensors. Sensor networks; Applications of electronics
• Online Access: https://ir.uitm.edu.my/id/eprint/68849/1/68849.pdf; https://ir.uitm.edu.my/id/eprint/68849/; https://crinn.conferencehunter.com/index.php/jcrinn

Single-mode biometric systems suffer from several problems that make them unsuitable for current biometric applications that require high levels of reliability and security. These problems include the use of a single biometric trait that is prone to noise, poor capture, lack of biometric points, and deterioration of biometric input quality. In this paper, we are interested in decision fusion access control on a biometric bimodal pattern recognition system based on fingerprints and facial recognition. To realize this access control system based on facial recognition and fingerprints, we used an embedded system under Arduino, we programmed electronic systems for the automatic opening of doors without human action being. The performance evaluation of decision fusion access control on a biometric bimodal pattern recognition system is realized by means of the confusion matrix, the calculations of the evaluation parameters (Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value and False Negative). From a sample of 500 individuals, 250 of whom were registered and 250 non-registered, our access control system obtained the results of 248 true positives, 2 false negative, 1 false positive and 249 true negatives which constitute our confusion matrix. However, from the set of tests performed we can conclude that by taking advantage of the fusion of these two modalities, we increase the verification performance of system as the verification performance of bimodal system (fingerprint decision fusion and facial recognition) is applied to give even better results compared to single mode systems.