Abstract
The Intellilink Synthesis Network (ISN) Terminal Enhancement and Comprehensive Facial Identification System introduce a transformative approach to
law enforcement in Malawi by integrating advanced facial recognition technology for real-time identity verification and suspect tracking. These systems address critical challenges such as misidentification, outdated databases,
and delays in suspect apprehension, offering scalable, secure, and efficient
solutions tailored to resource-constrained environments. The ISN Terminal
Enhancement, an Android-based mobile application, extends the ISN facial
recognition system, enabling law enforcement officers to verify identities in
the field. It employs OpenCV for face detection, DLib for facial landmark detection, and face_recognition for generating 128-dimensional facial encodings, which are securely processed by a Django-based backend using HTTPS
and JWT authentication. Asynchronous processing with Celery ensures scalability, while audit logging enhances accountability.
The Comprehensive Facial Identification System complements this by integrating facial recognition with live video feeds to detect wanted individuals
in real time. It includes modules for suspect registration, real-time alerts,
and video feed integration, supported by a robust database design with tables for suspect information, video logs, and officer activity records. Officers
receive automated alerts for matched suspects, and administrators manage
the system through a user-friendly interface. The systems leverage deep
learning for high-accuracy matching, with lightweight processing optimized
for mobile use and encryption protocols ensuring privacy. Key features include real-time facial recognition, citizen and suspect data access, manual
lookup capabilities, and comprehensive audit logging.
The combined systems offer significant benefits: rapid identity verification
in seconds, field accessibility via Android devices, high accuracy through
deep learning, and secure operations with encrypted data. These solutions
empower law enforcement with efficient tools, reducing operational delays
and enhancing public safety. The project lays a foundation for future biometric innovations, with potential enhancements including additional biometric modalities (e.g., fingerprint or iris recognition) and cross-border collaboration. By leveraging open-source technologies and secure architectures,
this work contributes to the growing body of AI-driven law enforcement
research in Africa, demonstrating the potential of mobile technology to address critical security challenges in Malawi and beyond.
This research, conducted as part of a Bachelor of Engineering in Computer
Science at DMI St. John the Baptist University, underscores the importance
3
of developing contextually relevant solutions for African law enforcement.
It aligns with global trends in biometric technology while addressing local
needs, offering a model for scalable and ethical AI applications in public
safety