Despite Nigeria's entry into a new electoral era, traditional voting systems continue to suffer from persistent problems, including lack of transparency, result manipulation, and voter intimidation. Existing digital alternatives also face challenges such as digital illiteracy, limited accessibility in rural areas, and vulnerability to cyber threats. To address these issues, this study proposes the Intelligent Decentralised Voting System (IDVS), a blockchain-based framework designed to support secure, transparent, and immutable election processes. The system incorporates multiple security, accessibility, and data integrity mechanisms, including facial recognition for voter authentication, QR codes for accessing voting results, and Fernet encryption for protecting data. The research followed a prototype-based design process. An abstract prototype of IDVS was developed using Figma and evaluated experimentally through the Useberry UX research platform. The Perception, Accessibility, Cost, and Technology (PACT) framework served as the analytical lens for assessing system performance based on survey responses from participants. Findings indicate an overall positive user response to the proposed system. Approximately 80% of participants completed a parallel registration and voting process, and over 70% reported confidence in the system's security and transparency features. Participants responded favourably to the facial recognition and QR code functionalities in particular. However, concerns were also raised regarding internet reliability and data privacy. In general, the study demonstrates the usability potential of the proposed IDVS framework and suggests that decentralised voting approaches offer a viable path toward improving electoral transparency and strengthening voter trust in Nigeria's electoral system.