Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Building of Informatics, Technology and Science

Sistem Keamanan Dua Lapis Dengan RFID dan Pendeteksi Objek Dengan Machine Learning Jhonatan, Jhonatan; Sekarsari, Kartika
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5747

Abstract

Conventional locking systems with physical keys are still widely used to secure the house door. Additionally, marketed security devices often come with various features but only have one security method used as the access key for the security system. This research designs a two-layer security system by sequentially applying two security methods: Radio Frequency Identification (RFID) and object detection, with Arduino UNO as the main microcontroller. In designing the two-layer security device for the room door, the RFID MFRCC22 module and OV2640 camera are used on the ESP32-CAM microcontroller. Testing results show that this device can function well using an RFID card in the first layer and small objects with maximum dimensions of 20 cm in length, width, and height in the second layer. With an operating voltage of 5Vdc and a current requirement of 150mA to 250mA, this system has high efficiency with low power consumption. The response time required to access this two-layer security system is 5.71 seconds to 6.57 seconds. The maximum distance between the RFID card and the RFID Reader is 5 cm, and between the ESP32-CAM camera and the object is between 5 cm and 40 cm. Additionally, the minimum number of image samples required for each object with different positions and angles to be applied to the ESP32-CAM microcontroller is 75 image samples with RGB color parameter configuration and 48x48 pixel image size, resulting in an F1-Score percentage of 100% so that the ESP32-CAM microcontroller can recognize objects between different object models. The F1-Score value in the Background column is 1.00, the Charger Hp column is 1.00, the Motorcycle Key column is 1.00, and the Leagoo column is 1.00.