<![CDATA[The attendance system at Bencah Village has been conducted manually, creating vulnerability to data manipulation, recording errors, and lack of transparency in managing village official attendance data. This study aimed to design and develop a face recognition-based attendance system for Bencah Village officials using deep learning technology, implemented as a web-based platform. The system was developed using the Waterfall Software Development Life Cycle (SDLC) model with a layered microservice architecture, integrating face detection using the lightweight YOLOv8n model, face feature extraction using InsightFace buffalo_l with the ArcFace approach generating 512-dimensional embedding vectors, and identity matching using cosine similarity with a threshold of 0.5. The backend was built with Laravel 12 and Python Flask, containerized using Docker and communicating via REST API. Functional testing using the black-box testing method on 25 main system functions confirmed that all functions performed as expected. Face recognition performance evaluation on 30 test data yielded a Recognition Accuracy of 93.33%, a False Acceptance Rate (FAR) of 0%, and a False Rejection Rate (FRR) of 10%. The FAR of 0% confirmed the system successfully prevented unauthorized identity acceptance, which is critical for attendance data integrity. These results demonstrate the system is feasible as a biometric attendance solution for village government environments.]]>
Copyrights © 2026
