Trustworthy Method for Person Identification in IIoT Environments by Means of Facial Dynamics

Abstract

In Industrial Internet of Things environments, dependability of complex manufacturing process in which human operators play a key role can be improved by identity recognition/authentication of whoever is involved in the various stages of a production process, according to where and when he/she is supposed to be. To this aim we propose an approach that exploits the dynamic appearance and the time-dependent local features characterizing the face of an individual during speech utterance with regard to their spatial and temporal components.The proposed method models these dynamic facial patterns captured from edge IoT devices by means of LBP-TOP descriptor, which effectively extract both face's local features and movement at the fog level of the architecture. A deep feed-forward network available in the cloud, is trained and optimized to match the extracted features to a reference database. The achieved results highlight state-of-the-art performances of the proposed method with regard to robustness and trustworthiness of identification, especially for challenging IIoT scenarios.

Biometrics for Industry 4.0: a survey of recent applications

“…( 2018 ) De-authentication Identification Posture Patterns Private dataset: 30 subjects ML algorithms applied to time series of the force on the sensors True positive rate: 91.0% False positive rate: 0,33% False negative rate: 8.68% Multiple sessions True positive rate: 22%, False positive rate: 5.2%, False negative rate: 72.7% Castiglione et al. ( 2020 ) Identification Authentication Facial dynamics Private dataset: 48 subjects Local spatial–temporal descriptor for features Deep feedforward network for classification Identification Accuracy: 98.2% Authentication Accuracy: 99.49% EER: 0.05% Srivastava et al. ( 2019 ) Identification Voice Private datasets: 10 subjects LPC and MFCC feature extraction technique GMM classifier Isolated Hindi digit dataset Accuracy: 96.49% Hindi sentence dataset Accuracy: 94.97% Alkeem et al.…”

“…To face this kind of challenge, both in terms of identification but also authentication, the authors in Castiglione et al. ( 2020 ) analyzed a dynamic approach with respect to the classical static representations of the face by exploiting the dynamics associated with language. The experiments were conducted on a customized database containing short video clips capturing 48 subjects speaking short sentences.…”

“… 2020 Facial dynamics H M L M L H M Security Castiglione et al. 2020 Keystroke H M L L L L M WHP Ulinskas et al. 2018 Signature H H L L L L H Security Bhowal et al.…”

A Secure Lightweight Authentication Mechanism for IoT Devices in Generic Domain

“…An authentication scheme based on the edge-fogcloud was proposed that captured the dynamic facial pattern from the edge of the IoT devices. Although this scheme enhances the robustness of the presentation attack, it could not provide mutual authentication and it was not secure enough and had no key agreement [12]. One proposed lightweight authentication scheme based on numerical series cryptography for IoT environments was proposed to provide mutual authentication and session key agreement for IoT devices [13].…”

Person Recognition Optimization Method Relying on Radial Basis Function Network

“…Both RBF NNs and multi-layer perceptrons are general aroaches, and as with any RBF NN, it can be seen that multi-layer perceptrons can completely replace NNs with perceptron layers RBF [17,18]. But they also have the following differences:…”