Cyber Physical Customer Management of Internet Of Robotic Things Enabled Banking

Abstract

In-person banking continues to play a vital role in financial services, particularly for document-based transactions and customer identity verification. This work presents an enhanced cyber-physical banking framework that integrates Internet of Robotic Things (IoRT) with an advanced automated Know-YourCustomer (KYC) mechanism. The proposed system extends existing decentralized KYC architectures by incorporating a Vision Transformer (ViT)-based deep biometric model for improved feature extraction and identity validation. Multiple biometric modalities, including facial images, fingerprints, voice, and signatures, are securely captured through humanoid robotic agents and processed using deep learning techniques. To ensure data confidentiality and integrity, a hybrid cryptographic approach combining symmetric and asymmetric encryption is employed alongside blockchain-based decentralized storage and verification. Furthermore, a high-capacity fragile watermarking scheme based on integer wavelet transform and lattice vector quantization is utilized to embed biometric information into banking documents, enhancing tamper detection and document security. A hierarchical privacy-preserving module is also introduced to anonymize sensitive visual data collected in banking environments. Experimental insights demonstrate that the proposed framework improves accuracy, security, and efficiency compared to conventional KYC systems, making it a robust and scalable solution for next-generation hybrid banking applications.