Smart Keyless Locker Design Using Face Recognition Technology Based on the Internet of Things Jakarta Global University Classroom

Abstract

This research presents the design and implementation of an Internet of Things (IoT)-based Smart Keyless Locker integrated with face recognition technology to enhance security and efficiency in classroom locker management at Jakarta Global University. The system replaces conventional mechanical keys with biometric authentication to minimize risks associated with key loss, duplication, and unauthorized access. The hardware architecture consists of a Raspberry Pi as the primary processing unit for facial recognition, an Arduino Mega for actuator control, a camera module for image acquisition, solenoid door locks as locking mechanisms, load cell sensors for locker status detection, and an IoT-based notification system integrated with WhatsApp for real-time monitoring. The facial recognition process utilizes the Haar Cascade Classifier for face detection and the Local Binary Patterns Histograms (LBPH) algorithm for feature extraction and matching. System performance was evaluated under various conditions, including differences in lighting intensity, facial orientation, distance, and face coverings. Experimental results indicate that the system achieved a recognition success rate of 50% under the tested conditions, particularly within a distance range of 40–70 cm and adequate lighting. The average verification time ranged from 1.4 to 2.1 seconds depending on facial angle, while the WhatsApp notification system demonstrated reliable message delivery with an average delay of 4.75 seconds. Although recognition performance decreases when facial features are partially obstructed or when lighting is insufficient, the proposed system demonstrates the feasibility of integrating biometric authentication with IoT technology for modern classroom locker management applications.